Journal of Diabetes and Its Complications 29 (2015) 59–63

Contents lists available at ScienceDirect

Journal of Diabetes and Its Complications

j ourna l homepage: WWW.JDCJOURNAL.COM

Elevated beta2-glycoprotein I-low-density lipoprotein levels are

associated with the presence of diabetic microvascular complications

Ruijie Yu 1, Yunlong Yuan 1, Dongmei Niu, Jiaxi Song, Ting Liu, Jia Wu ⁎, Junjun Wang ⁎⁎Department of Clinical Laboratory, Jinling Hospital, School of Medicine, Nanjing University, Nanjing, China1 These authors have “equal contributions” to this work.

Disclosure of interest: The authors declare that theconcerning this article.⁎ Correspondence to: J. Wu, Department of Clinical La

East Zhongshan Rd., Nanjing, 210002, China. Tel.: +86 2⁎⁎ Correspondence to: J.J. Wang, Department of Clinical LEast Zhongshan Rd., Nanjing, 210002, China. Tel.: +884815775.

E-mail addresses: wujia0801@126.com (J. Wu), jjwa1 These authors have “equal contributions” to this work.

http://dx.doi.org/10.1016/j.jdiacomp.2014.09.0101056-8727/© 2015 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 15 June 2014Received in revised form 30 August 2014Accepted 18 September 2014Available online 28 September 2014

Keywords:β2-Glycoprotein-IOxidized low-density lipoproteinType 2 diabetes mellitusDiabetic complicationsMicroangiopathy

Aims: To investigate serum beta2-glycoprotein I-low-density lipoprotein (β2-GPI-LDL) and oxidized low-density lipoprotein (ox-LDL) levels in type 2 diabetes mellitus (T2DM) patients, and to further evaluate theassociations of β2-GPI-LDL with ox-LDL in vivo and with the presence of diabetic microvascular complications.Methods: We determined β2-GPI-LDL, ox-LDL and small dense low density lipoprotein cholesterol (sdLDL-C)levels in 236 T2DM patients with or without microvascular complications and 75 controls. The correlationanalyses, multiple linear regression analyses and logistic regression analyses were performed, respectively.Results: Compared with controls, β2-GPI-LDL and ox-LDL levels were significantly elevated in both groups ofT2DMpatients and thosewithmicrovascular complications exhibited themore significant increase than thosewithout complications. Serum β2-GPI-LDL levels were positively correlated with ox-LDL as well as sdLDL-Clevels in T2DM patients. Multiple linear regression analyses showed that ox-LDL was one of the independentdeterminants of β2-GPI-LDL levels. Logistic regression analyses indicated that elevated β2-GPI-LDL and ox-LDL

levels had significant predictive values for diabetic microvascular complications.Conclusions: Elevated serum β2-GPI-LDL levels may be a serological hallmark of enhanced LDL oxidationin vivo and closely associated with the presence of diabetic microvascular complications.

© 2015 Elsevier Inc. All rights reserved.

1. Introduction

Type 2 diabetes mellitus (T2DM) is known to be closely associatedwith the dysregulation of glucolipid metabolism and aggravateddevelopment of vascular complications (Pinhas-Hamiel & Zeitler,2007). It has been demonstrated that hyperglycemia and hyperlipid-emia in T2DM may enhance systematic oxidative stress, resulting inthe excessive production of lipid peroxides and subsequentlycontributing to the pathogenesis of atherothrombosis and microan-giopathy, which was the common pathogenetic mechanism underly-ing diabetic vascular complications (Davì, Falco, & Patrono, 2005;Pinhas-Hamiel & Zeitler, 2007).

Small, dense low-density lipoprotein (sdLDL), the generallyacknowledged component of atherogenic lipoproteins, has beencharacterized by the great susceptibility to oxidative modification

y have no conflicts of interest

boratory, Jinling Hospital, 3055 80860181.aboratory, Jinling Hospital, 3056 25 80861177; fax: +86 25

ng9202@gmail.com (J. Wang).

(Hoogeveen et al., 2014). Oxidized low-density lipoprotein (ox-LDL),as a recognized hallmark of in vivo lipids peroxidation, has beenconsidered to play a pivotal role in the initiation and progression ofatherosclerosis (Koenig et al., 2011). Beta2-glycoprotein I (β2-GPI), themain antigenic target for antiphospholipid antibodies, can bind ox-LDLwith higher affinity than native low-density lipoprotein (LDL) to formstable and indissociable β2-GPI/ox-LDL complexes in vitro, consequentlybeing involved in antibody-mediated atherothrombosis in patients withsystemic autoimmune diseases (Kobayashi et al., 2003; Lopez, Simpson,Hurley, & Matsuura, 2005). In vivo studies also revealed that increasedβ2-GPI/ox-LDL complexes existed in the bloodstream of patients withnon-autoimmune diseases accompanied by premature or acceleratedatherogenesis, indicating its potentially pathogenetic role in vascularthromboembolic events (Greco et al., 2010; Kasahara et al., 2004; Lopez,Hurley, Simpson, & Matsuura, 2005).

Enhanced oxidative stress in T2DM patients may lead toendothelial cell damage and vascular dysfunction through variousmechanisms, particularly promoting microvascular thromboembo-lism and leading to the onset and development of diabeticmicroangiopathy (Martín-Gallán, Carrascosa, Gussinyé, & Domínguez,2003). Lopez, Hurley, et al. (2005) reported that the lower levels ofserum β2-GPI/ox-LDL complexes in T2DM patients taking cholesterol-lowering statins were in agreement with the antioxidant and anti-thrombotic properties of statins, suggesting that high β2-GPI/ox-LDL

60 R. Yu et al. / Journal of Diabetes and Its Complications 29 (2015) 59–63

levelsmight be a consequence of LDL atherogenicmodificationmediatedby oxidative stress. Nevertheless, no study has been published to showdirect association between serumβ2-GPI/ox-LDL and ox-LDL levels inT2DM. Furthermore, it still needed to study the differences in serumlevels of β2-GPI/ox-LDL between T2DM patients with and withoutmicrovascular complications and the association of β2-GPI/ox-LDLwith the occurrence of microangiopathy. Therefore, this study wasundertaken to investigate serum β2-GPI-LDL, ox-LDL levels and theirassociations in T2DM patients and to further evaluate the clinicalvalues of β2-GPI-LDL for predicting the presence of diabeticmicrovascular complications.

2. Materials and methods

2.1. Study subjects

A total of 236 newly admitted T2DM subjects were randomlyenrolled from the department of endocrinology of Jinling Hospitalbetween September 2011 and April 2012. All the patients werediagnosed according to the fasting plasma glucose (FPG) concentra-tion and the presentation of clinical symptoms relevant to diabetesbased on the 1999 World Health Organization criteria (World HealthOrganization, 1999). The exclusion criteria included the presence oftype 1 diabetes mellitus, gestational diabetes, diabetic macrovascularcomplications (including hypertension, cardiovascular/cerebrovascularaccidents and lower extremity vascular diseases), chronic liver or renaldiseases, severe infections or other malignant diseases and previouslydiagnosed diseases during drugwithdrawal nomore than 3 months. Ofthese T2DM patients, 135 individuals diagnosed with microvascularcomplications (including diabetes nephrology, diabetic retinopathy anddiabetic neuropathy) formed group 1,whereas group 2 consisted of 101individuals diagnosed without any clinical signs and symptoms ofdiabetic vascular complications.

75 healthy subjects who had contemporaneously visited JinlingHospital for routine health examination constituted the control group.They were all found to be normal in physical, electrocardiography,ultrasonography examination and serologically biochemical testswithout any other definitive diseases such as hyperlipemia, hyper-tension, cardiovascular or cerebrovascular diseases, diabetes mellitus,severely impaired hepatic function and any recent surgery.

The blood was sampled at least 12 h after fasting and serum waspromptly separated by a 15 min centrifugation at 3000 rpm, andstored at −80 °C until analysis. This study protocol was approved bythe Ethics Committee of Jinling Hospital (2012GJJ-044) and all thesubjects provided written informed consent.

2.2. Laboratory methods

Serum β2-GPI-LDL levels were determined by a sandwich enzyme-linked immunosorbent assay using the polyclonal anti-human β2-GPIantibody as the capture antibody and the peroxidase conjugatedpolyclonal antibody against apolipoprotein B as the detectionantibody (Zhang et al., 2011). Briefly, 500 μL of serum was firstlyincubated with MgCl2 (final concentration 10 μmol/L) at 37 °C for 2 hand then polyethyleneglycol-6000 (Sigma-Aldrich, United States) wasadded to isolate β2-GPI-LDL from endogenous free form of β2-GPI. Thesamples were incubated overnight at 4 °C and then centrifuged at10,000 rpm for 20 min. The precipitates were resuspended in 500 μLwashing solution containing 0.5% gelatin and 0.05% Tween-20 in0.01 mol/L PBS buffer solutions. A pooled fresh-frozen plasma sample(mixed plasma from 50 healthy subjects) was used as the referenceserum of β2-GPI-LDL. Reference serum was also precipitated everytime as serum sample. The value of β2-GPI-LDL was expressed as 1relative absorbance unit (U/mL).

The measurement of ox-LDL was analyzed by a commerciallyavailable sandwich enzyme-linked immunosorbent assay using the

monoclonal anti-human ox-LDL antibody as the capture antibody andthe peroxidase conjugatedmonoclonal antibody against apolipoproteinB as the detection antibody (Mercodia, Sweden). Serum sdLDLcholesterol (sdLDL-C) levels were detected by use of the sdLDL-EXSeiken kit (Denka Seiken, Japan) based on a homogeneous assayadaptable to autoanalyzers (Ito, Fujimura, Ohta, & Hirano, 2011). Thelevels of FPG (Wako Pure Chemical Industries, Japan) and serumlipid/lipoprotein profiles including total cholesterol (TC), triglyceride(TG), high density lipoprotein cholesterol (HDL-C) and LDL cholesterol(LDL-C) (Daiichi Pure Chemicals, Japan) were measured by enzymaticprocedures both on a model 7600 automatic analyzer (Hitachi, Japan).

2.3. Statistical analysis

Data analyseswere performedusing SPSS version 16.0. Kolmogorov–SmirnovTestwasused to evaluate thenormalityof variables and skeweddata were log-transformed to create a more normal distribution.Normally distributed values were expressed as mean ± standarddeviation and the skewed distribution values were expressed asmedianand interquartile range (P25–P75). Comparisons of normal variablesamong groups were analyzed by one-way ANOVA test and thedifferences between groups were subsequently determined by FisherLSD test when appropriate. Comparisons of skewed variables which stillremained skewedly distributed after log-transformed among groupswere analyzed by Kruskal–Wallis H test. The Chi-square test was used tocompare differences of gender among groups. Correlations betweenvariables were calculated by non-parametric Spearman rank coefficienttest. The stepwise multiple linear regression analyses (Pin = 0.05,Pout = 0.10) were used to identify the influencing factors for serumβ2-GPI-LDL levels. The univariate and multivariate logistic regressionanalyses were used to calculate the approximation of the relative risk,odds ratio (OR) and 95% confidence interval (CI) for selected variables. Atwo-tailed P-value less than 0.05was considered statistically significant.

3. Results

3.1. Serum β2-GPI-LDL, ox-LDL and sdLDL-C levels in T2DM subjects

Compared with controls, serum β2-GPI-LDL and ox-LDL levelswere significantly increased in both groups of T2DM patients andsdLDL-C levels were elevated in patients with diabetic microvascularcomplications. The levels of β2-GPI-LDL, ox-LDL and sdLDL-C werehigher in T2DM patients with complications than in those withoutcomplications. The other lipid/lipoprotein status and the glucoselevels are shown in Table 1.

3.2. Associations among β2-GPI-LDL, ox-LDL, sdLDL-C and other lipid/lipoprotein parameters in T2DM subjects

To study the relationship of β2-GPI-LDL, ox-LDL, sdLDL-C withother lipid/lipoprotein parameters, Spearman rank correlation anal-yses were performed. Due to the similar associations among β2-GPI-LDL, ox-LDL, sdLDL-C and other lipid/lipoprotein parameters in T2DMpatients with or without microvascular complications (data notshown), we combined two groups for the next analyses (n = 236).

In all the T2DM patients, the β2-GPI-LDL levels were positivelycorrelated with ox-LDL, sdLDL-C, TC and LDL-C; ox-LDL levels werepositively related with sdLDL-C, TC and LDL-C; sdLDL-C levelsexhibited positive correlations with TC, TG and LDL-C (Table 2).

To further explore the possible factors affecting β2-GPI-LDL andox-LDL levels in T2DM patients, the multiple linear regressionanalyses were performed. Consequently, the ox-LDL, TC accountedfor 42.0% of the variation of β2-GPI-LDL levels, when all lipid/lipoprotein parameters were included as independent variables(Table 3). In addition, only β2-GPI-LDL (β-coefficient = 0.382,

Table 1The clinical and biochemical characteristics in T2DM patients and controls.

Variables Group 1 (n = 135) Group 2 (n = 101) Controls (n = 75)

Age, (years) 57.14 ± 15.81 55.96 ± 15.24 54.32 ± 16.42Male, n (%) 83 (61.48%) 62 (61.39%) 46 (61.33%)FPG (mmol/L) ⁎ 6.90 (5.60–9.00) §§ 7.70 (6.30–10.60) §§ 4.60 (4.30–4.8)TC (mmol/L) 4.65 (3.80–5.30) 4.30 (3.70–5.00) 4.38 (4.03–4.78)TG (mmol/L) ⁎ 1.53 (1.14–2.43) §§ 1.44 (0.97–1.92) §§ 0.87 (0.64–1.05)HDL-C(mmol/L) ⁎

1.02 (0.88–1.25) §§ 1.07 (0.89–1.26) §§ 1.61 (1.38–1.87)

LDL-C(mmol/L)

2.96 ± 1.12 §§ 2.68 ± 0.82 2.55 ± 0.62

sdLDL-C(mmol/L) ⁎

1.02 (0.62–1.38)§§,# 0.79 (0.58–1.08) 0.70 (0.57–0.93)

ox-LDL (U/L) ⁎ 46.62 (33.54–60.36)§§, # 41.08 (28.42–54.23)§§ 25.83(20.42–36.25)β2-GPI-LDL(U/mL) ⁎

1.10 (0.78–1.43) §§, ## 0.86 (0.65–1.20)§ 0.80 (0.59–1.06)

⁎ These variables were log-transformed before analyses. Group 1: T2DM patientswith microvascular complications; Group 2: T2DM patients without complications.

§ Compared with Controls: P b 0.05.§§ Compared with Controls: P b 0.01.# Compared with Group 2: P b 0.05;## Compared with Group 2: P b 0.01.

Table 3Multiple linear regression analyses of possible factors affecting β2-GPI-LDL levels inT2DM patients (n = 236).

Unstandardizedcoefficients

Standardized coefficient P-value

B SE Beta

Independent variables in the modelConstant −0.354 0.138 0.011TC 0.277 0.031 0.558 b0.001ox-LDL 0.004 0.001 0.208 0.001

Independent variables excluded from the modelTG −0.092 0.143HDL-C 0.035 0.570LDL-C 0.140 0.163sdLDL-C −0.121 0.180

The dependent variable was β2-GPI-LDL (adjusted R2 = 0.420). All lipid/lipoproteinvariables on the correlation analyses were used for independent variables. P b 0.05 wasconsidered statistically significant.

61R. Yu et al. / Journal of Diabetes and Its Complications 29 (2015) 59–63

P b 0.001) was the significantly independent predictor for ox-LDLlevels (adjusted R2 = 0.141) and both TC (β-coefficient = 0.690,P b 0.001) and TG (β-coefficient = 0.200, P b 0.001) were thesignificantly independent determinants of sdLDL-C levels (adjustedR2 = 0.605).

3.3. Associations of β2-GPI-LDL, ox-LDL and sdLDL-C with the presence ofdiabetic microvascular complications

Theunivariate andmultivariate logistic regression analyseswerenextperformed to evaluate the possible associations of β2-GPI-LDL, ox-LDL orsdLDL-C with diabetic microvascular complications, respectively.

As shown in Table 4, the univariate analyses revealed thatincreased β2-GPI-LDL and ox-LDL levels were significantly related tothe presence of T2DM with or without microvascular complications,and high sdLDL-C levels were only associated with the presence ofdiabetic microvascular complications, respectively. In multivariateanalyses, adjusting for age, gender and other serum lipid/lipoproteinlevels, elevated β2-GPI-LDL and ox-LDL still had significantlypredictive values for the presence of diabetic microvascular compli-cations, and only high ox-LDL levels were associated with thepresence of T2DM without complications, respectively.

4. Discussion

The present study found that both β2-GPI-LDL and ox-LDL levelswere significantly increased in T2DM patients with microvascularcomplications. The β2-GPI-LDL levelswere independent correlatedwithox-LDL in T2DM patients. Elevated β2-GPI-LDL and ox-LDL levels wereassociated with the presence of microvascular complications in T2DM.

Table 2Spearman's correlation coefficients among β2-GPI-LDL, ox-LDL, sdLDL-C and other lipid/lipoprotein parameters in T2DM patients (n = 236).

Variables TC TG HDL-C LDL-C sdLDL-C ox-LDL β2-GPI-LDL

β2-GPI-LDL r = 0.557 r = 0.168 r = 0.112 r = 0.495 r = 0.387 r = 0.406 –

P b 0.001 P = 0.015 P = 0.105 P b 0.001 P b 0.001 P b 0.001 –

ox-LDL r = 0.296 r = 0.064 r = 0.008 r = 0.330 r = 0.167 – r = 0.406P b 0.001 P = 0.369 P = 0.908 P b 0.001 P = 0.024 – P b 0.001

sdLDL-C r = 0.682 r = 0.635 r = −0.082 r = 0.485 – r = 0.167 r = 0.387P b 0.001 P b 0.001 P = 0.243 P b 0.001 – P = 0.024 P b 0.001

β2-GPI, as a member of the plasma complement control proteinsuperfamily, has been characterized by its ability to bind lipoproteinsand other negatively chargedmaterials (Kobayashi et al., 2003; Lopez,Simpson, et al., 2005). It has been demonstrated that β2-GPI couldinteract with ox-LDL via 7-ketocholesterol having a w-carboxyl acylchain to form covalently bound β2-GPI/ox-LDL complexes in vitro(Kobayashi et al., 2003; Lopez, Simpson, et al., 2005). Circulatingβ2-GPI/ox-LDL complexes also existed in patients with significantatherothrombotic vascular involvement (Greco et al., 2010; Lopez,Hurley, et al., 2005; Lopez, Simpson, et al., 2005). In the present study,serum β2-GPI-LDL and ox-LDL levels were found to be significantlyincreased in both groups of T2DM patients. It has been known thatdysregulation of glucolipid metabolism in T2DM predisposed individ-uals to premature atherosclerosis and further contributed to theaccelerated thrombosis (Davì et al., 2005; Pinhas-Hamiel & Zeitler,2007). The ox-LDL has beenwidely recognized as themajor atherogeniclipoproteins (Koenig et al., 2011). The β2-GPI/ox-LDL complexes maypromote the accelerated intracellular accumulation of ox-LDL in thepresence of β2-GPI mediated by anti-β2-GPI antibodies and furthercontribute to foam cell formation (Kajiwara, Yasuda, &Matsuura, 2007),which revealed the pathogenetic relevance ofβ2-GPI/ox-LDL complexesin atherogenesis. Consistently, the colocalization of β2-GPI and ox-LDLhas also been found in the arterial intima of atherosclerotic lesions(George et al., 1999). Thus, elevated β2-GPI/ox-LDL complexes andox-LDL levelsmay be relatedwith the pathogenesis of atherothromboticevents in T2DM patients. Interestingly, we also observed the significantcorrelation between serum β2-GPI-LDL and ox-LDL levels in T2DM.Furthermore, ox-LDL was one of the independent determinants ofβ2-GPI-LDL levels and only β2-GPI-LDL was the independent predictorfor ox-LDL levels. Since the sandwich ELISA for detecting β2-GPI-LDL inthis study could capture all of β2-GPI complexes with oxidizationmodified LDL, other modified forms of LDL (such as MDA-modified LDLand acetylated LDL) as well as native LDL (if existing), the independentlinkage ofβ2-GPI-LDLwith ox-LDL but notwith native LDL (representedby LDL-C) may directly provide the striking evidence supporting thatox-LDL was the preferential form interacted with β2-GPI and high

Table 4Univariate and multivariate logistic regression analyses for the clinical predictivevalues of β2-GPI-LDL, ox-LDL and sdLDL-C.

Group⁎ Univariate analysis& Multivariate analysis$

Unadjusted OR (95% CI) P-value OR (95% CI) P-value

Group 1: T2DM patients with microvascular complications (n = 135)β2-GPI-LDL 6.83 (2.64, 17.68) b0.001 4.34 (1.05, 18.05) 0.043ox-LDL 1.07 (1.04, 1.09) b0.001 1.04 (1.01, 1.07) 0.009sdLDL-C 6.16 (2.50, 15.19) b0.001 0.42(0.07, 2.64) 0.352

Group 2: T2DM patients without complications (n = 101)β2-GPI-LDL 3.25 (1.25, 8.48) 0.016 2.62 (0.63, 10.93) 0.186ox-LDL 1.06 (1.03, 1.09) b0.001 1.05 (1.02, 1.08) 0.004sdLDL-C 2.11 (0.84, 5.32) 0.114 0.26(0.04, 1.70) 0.160

In univariate and multivariate logistic regression analyses, group 1, group 2 or thecontrol group was treated as a dependent three-category variable.

& Only one variable of β2-GPI-LDL, ox-LDL and sdLDL-C was included in the model.$ The age, gender and serum lipid/lipoprotein levels were adjusted in the model. OR

was considered significant when the lower limit of the 95% CI was N1.0. P b 0.05 wasconsidered statistically significant.⁎ Reference category: the control group.

62 R. Yu et al. / Journal of Diabetes and Its Complications 29 (2015) 59–63

β2-GPI-LDL levels may be mainly attributed to the increase of ox-LDLlevels. Elevated β2-GPI-LDL may serve as a serologically relevanthallmark of enhanced LDL oxidation. Additionally, itwas also sufficientlyavailable formeasuring the complexes of β2-GPI binding to LDL particleswith different oxidation degrees in different sites (from minimallyox-LDL to extensively ox-LDL). Thus, ourβ2-GPI-LDL assayhadawiderdetection range than the reported β2-GPI/ox-LDL method (Kobayashiet al., 2003; Lopez, Hurley, et al., 2005; Lopez, Simpson, et al., 2005) andwould better reflect the levels of pathogenic LDL in vivo. Similarly, wealso analyzed β2-GPI-Lp(a) levels using the same capture antibodyagainst human β2-GPI in patients with T2DM and coronary arterydisease (Wang et al., 2012; 2013).

In T2DM patients, glucolipid metabolism alterations caused byinsulin resistance may aggravate the imbalance of oxidant/antioxidantsystems, leading to increased reactive oxygen species and subsequentlipids peroxidation, which further played a crucial role in theprogression of microvascular complications (Martín-Gallán et al.,2003). In addition, the fat deposition in microcirculation may alsocontribute to the deterioration of microangiopathy (Davì et al., 2005;Martín-Gallán et al., 2003; Pinhas-Hamiel & Zeitler, 2007). In this study,the β2-GPI-LDL and ox-LDL levels showed themore significant increasein patients with diabetic microvascular complications than in thosewithout any vascular complications, while sdLDL-C levels were onlyelevated in patients with microvascular complications. The abovediscrepant results may be explained by the fact that T2DM patientswithmicrovascular complicationswere in the relatively advanced stageof diabetes and their dysregulation of lipid metabolism (represented byelevated LDL-C) was more severe than those without any vascularcomplications. Additionally, we found no significant change in TC levels(one of the independent determinants of sdLDL-C levels) between theT2DM patients with and without complications, which may be alsoresponsible for the no significant increase of sdLDL-C levels in patientswithout complications. It has been demonstrated that enhancedoxidative stress could occur in the early stage of developing diabetes(Martín-Gallán et al., 2003) and ox-LDL may function as theparticularly sensitive biomarker (Koenig et al., 2011; Yan, Mehta,Zhang, & Hu, 2011). Consequently, ox-LDL levels still had a significantincrease in patients without complications. The excessive generation ofcirculating lipid peroxides may contribute to the aggravation ofendothelial dysfunction and cells oxidative injuries induced bysystematic oxidative stress, further promoting the onsetand development of microvascular thromboembolism in T2DMpatients (Martín-Gallán et al., 2003). Strikingly, both β2-GPI-LDLand ox-LDL levels showed significant associations with the presenceof T2DM with microangiopathy and only ox-LDL levels exhibited

significant association with the presence of T2DM without compli-cations. Since high ox-LDL levels have been regarded as the directevidence for the enhanced oxidative stress in vivo, the imbalanceof oxidant/antioxidant systems in diabetes may initially lead tothe elevation in ox-LDL levels and subsequently result in theincreased β2-GPI-LDL levels. Thus, ox-LDL levels may consequentlyexhibit the more obvious association with the presence of T2DMwithout complications (the early stage of diabetes) than β2-GPI-LDL.The ox-LDL has been reported to exert intensive cytotoxic effects tohumanvascular endothelial cells and smoothmuscle cells, subsequentlyparticipating in the pathogenesis of diabetic microangiopathy (Koeniget al., 2011;Yanet al., 2011).Highβ2-GPI-LDL levels, as a consequence ofoxidative lipids storage, may reflect the oxidant/antioxidant imbalanceand consequently exhibited the close association with the presence ofdiabetic microvascular complications.

In conclusion, the present study demonstrated that serum β2-GPI-LDL and ox-LDL levels were significantly increased in T2DM patients,especially in those with microvascular complications. The β2-GPI-LDLlevels were independently correlated with ox-LDL in T2DM patients.Elevated β2-GPI-LDL levels may be a serologically relevant hallmark ofenhanced LDL oxidation. Furthermore, elevated β2-GPI-LDL levelsmay be closely associated with the presence of diabetic microvascularcomplications. These findings may contribute to the understanding ofthe pathogenetic role for circulating β2-GPI-LDL in T2DM. Furtherstudies are needed to validate these associations and to elucidatepathophysiologic mechanisms of circulating β2-GPI-LDL in microan-giopathy of T2DM.

Conflict of interest

The authors declared no conflict of interest.

Acknowledgments

This work was supported by grants from the National NaturalScience Foundation of China (NSFC 81271904), the Special-fundedProgram on National Key Scientific Instruments and EquipmentDevelopment of China (2012YQ 03026109) and the Jinling HospitalFoundation (No. 2014051).

References

Davì, G., Falco, A., & Patrono, C. (2005). Lipid peroxidation in diabetes mellitus.Antioxidants and Redox Signaling, 7, 256–268.

George, J., Harats, D., Gilburd, B., Afek, A., Levy, Y., Schneiderman, J., et al. (1999).Immunolocalization of beta2-glycoprotein I (apolipoprotein H) to human athero-sclerotic plaques: Potential implications for lesion progression. Circulation, 99,2227–2230.

Greco, T. P., Conti-Kelly, A. M., Anthony, J. R., Greco, T., Jr., Doyle, R., Boisen, M., et al.(2010). Oxidized-LDL/β2-glycoprotein I complexes are associated with diseaseseverity and increased risk for adverse outcomes in patients with acute coronarysyndromes. American Journal of Clinical Pathology, 133, 737–743.

Hoogeveen, R. C., Gaubatz, J. W., Sun, W., Dodge, R. C., Crosby, J. R., Jiang, J., et al. (2014).Small dense low-density lipoprotein-cholesterol concentrations predict risk forcoronary heart disease: The Atherosclerosis Risk in Communities (ARIC) Study.Arteriosclerosis, Thrombosis, and Vascular Biology, 34, 1069–1077.

Ito, Y., Fujimura, M., Ohta, M., & Hirano, T. (2011). Development of a homogeneousassay for measurement of small dense LDL cholesterol. Clinical Chemistry, 57,57–65.

Kajiwara, T., Yasuda, T., & Matsuura, E. (2007). Intracellular trafficking of beta2-glycoprotein I complexes with lipid vesicles in macrophages: Implications on thedevelopment of antiphospholipid syndrome. Journal of Autoimmunity, 29, 164–173.

Kasahara, J., Kobayashi, K., Maeshima, Y., Yamasaki, Y., Yasuda, T., Matsuura, E., et al.(2004). Clinical significance of serum oxidized low-density lipoprotein/beta2-glycoprotein I complexes in patients with chronic renal diseases. Nephron. ClinicalPractice, 98, c15–c24.

Kobayashi, K., Kishi, M., Atsumi, T., Bertolaccini, M. L., Makino, H., Sakairi, N., et al.(2003). Circulating oxidized LDL forms complexes with β2-glycoprotein I:Implication as an atherogenic autoantigen. Journal of Lipid Research, 44, 716–726.

Koenig, W., Karakas, M., Zierer, A., Herder, C., Baumert, J., Meisinger, C., et al. (2011).Oxidized LDL and the risk of coronary heart disease: Results from the MONICA/KORA Augsburg Study. Clinical Chemistry, 57, 1196–1200.

63R. Yu et al. / Journal of Diabetes and Its Complications 29 (2015) 59–63

Lopez, L. R., Hurley, B. L., Simpson, D. F., & Matsuura, E. (2005). Oxidized low-density lipoprotein/beta2-GPI complexes and antibodies in patients withtype 2 diabetes mellitus. Annals of the New York Academy of Sciences, 1051,97–103.

Lopez, L. R., Simpson, D. F., Hurley, B. L., & Matsuura, E. (2005). OxLDL/beta2GPIcomplexes and autoantibodies in patients with systemic lupus erythematosus,systemic sclerosis, and antiphospholipid syndrome: Pathogenic implicationsfor vascular involvement. Annals of the New York Academy of Sciences, 1051,313–322.

Martín-Gallán, P., Carrascosa, A., Gussinyé, M., & Domínguez, C. (2003). Biomarkers ofdiabetes-associated oxidative stress and antioxidant status in young diabeticpatients with or without subclinical complications. Free Radical Biology andMedicine, 34, 1563–1574.

Pinhas-Hamiel, O., & Zeitler, P. (2007). Acute and chronic complications of type 2diabetes mellitus in children and adolescents. Lancet, 369, 1823–1831.

Wang, J. J., Gong, J. B., Li, H. Q., Niu, D. M., Han, A. Z., Wu, J., et al. (2012). Lipoprotein(a)complexes with beta2-glycoprotein I in patients with coronary artery disease.Journal of Atherosclerosis and Thrombosis, 19, 81–89.

Wang, C., Niu, D. M., Hu, J., Guan, X. C., Yang, W., Wang, J. J., Zhang, C. Y., & Zhang, C. N.(2013). Elevated serum β2-glycoprotein-I-lipoprotein(a) complexes levels areassociated with the presence and complications in type 2 diabetes mellitus.Diabetes Research and Clinical Practice, 100, 250–256.

World Health Organization (1999). Definition, diagnosis, and classification of diabetesmellitus and its complications. Geneva: Switzerland.

Yan, M., Mehta, J. L., Zhang, W., & Hu, C. (2011). LOX-1, oxidative stress andinflammation: A novel mechanism for diabetic cardiovascular complications.Cardiovascular Drugs and Therapy, 25, 451–459.

Zhang, C., Li, X., Niu, D., Zi, R.,Wang, C., Han, A., et al. (2011). Increased serum levels ofβ2-GPI-Lp(a) complexes and their association with premature atherosclerosis in patients withrheumatoid arthritis. Clinica Chimica Acta, 412, 1332–1336.