Pre-β1 HDL in type 2 diabetes mellitus

Accepted Manuscript Pre-β1 HDL in type 2 diabetes mellitus S.W. Shiu, Y. Wong, K.C. Tan PII:

S0021-9150(17)30234-4

DOI:

10.1016/j.atherosclerosis.2017.05.031

Reference:

ATH 15076

To appear in:

Atherosclerosis

Received Date: 15 February 2017 Revised Date:

22 May 2017

Accepted Date: 24 May 2017

Please cite this article as: Shiu SW, Wong Y, Tan KC, Pre-β1 HDL in type 2 diabetes mellitus, Atherosclerosis (2017), doi: 10.1016/j.atherosclerosis.2017.05.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Pre-β1 HDL in type 2 diabetes mellitus

Shiu SW, Wong Y, Tan KC.

SC

RI PT

Department of Medicine, University of Hong Kong, Hong Kong

Hospital, Pokfulam Road, Hong Kong.

AC C

EP

TE D

E-mail: [email protected] (K. Tan)

M AN U

Corresponding author: Department of Medicine, University of Hong Kong, Queen Mary

1

ACCEPTED MANUSCRIPT

Abstract Background and aims: Pre-β1 HDL, being a major acceptor of free cholesterol from cells, plays an important role in reverse cholesterol transport. This study was performed to determine

RI PT

whether abnormalities in pre-β1 HDL concentration were present in type 2 diabetes irrespective of their HDL-cholesterol levels, and the impact on cholesterol efflux.

Methods: 640 type 2 diabetic patients with or without cardiovascular disease (CVD) and 360

SC

non-diabetic controls matched for serum HDL-cholesterol levels were recruited. Plasma pre-β1 HDL was measured by ELISA, and cholesterol efflux to serum, mediated by ATP-binding

M AN U

cassette transporter A1 (ABCA1), was determined by measuring the transfer of [3H]cholesterol from cultured cells expressing ABCA1 to the medium containing the tested serum. Results: Despite the diabetic subjects having matched HDL-cholesterol and total apoA1 as controls, plasma pre-β1 HDL was significantly reduced in both male (p<0.01) and female diabetic

TE D

patients (p<0.05), and patients with CVD had the lowest pre-β1 HDL level. Serum capacity to induce ABCA1-mediated cholesterol efflux was impaired in the diabetic group (p<0.01) and cholesterol efflux correlated with pre-β1 HDL (Pearson’s r = 0.38, p<0.01), and this association

EP

remained significantly even after controlling for age, gender, body mass index, diabetes status, smoking, apoA1, triglyceride and LDL.

AC C

Conclusions: Plasma pre-β1 HDL level was significantly decreased in type 2 diabetes and was associated with a reduction in cholesterol efflux mediated by ABCA1. Our data would suggest that low pre-β1 HDL might cause impairment in reverse cholesterol transport in type 2 diabetes.

Keywords: Pre-β1 HDL; ATP-binding cassette transporter A1; cholesterol efflux; type 2 diabetes; cardiovascular disease 2

ACCEPTED MANUSCRIPT

Introduction HDL particles are highly heterogeneous and differ in their size, structure and composition. Conventionally, HDL concentration is reported in terms of the cholesterol concentration

RI PT

measured within the ultracentrifugally defined density range of 1.063 to 1.21 g/L, and further divisions within this density range have given rise to specific terminology for HDL subclasses [1]. Experimental data have suggested that HDL has multiple protective functions against

SC

atherosclerosis. Various HDL subpopulations differ in their anti-atherogenic properties and the functional diversity of HDL is intrinsically related to the heterogeneity in HDL structure [1,2].

M AN U

Pre-β1 HDL is a specific HDL subfraction that migrates with a pre-β mobility on agarose gel electrophoresis and is distributed over a range of hydrated density from approximately 1.21 to 1.25 g/L. The main components of pre-β1 HDL are apolipoprotein (apo) A1 and phospholipids [1,3]. The interaction of these discoidal, pre-β-migrating particles with the ATP-binding cassette

TE D

transporter A1 (ABCA1) promotes cellular cholesterol efflux and pre-β1 HDL particles play an important role in reverse cholesterol transport [4]. Therapy that raises pre-β1 HDL concentration like apoA1 mimetic leads to increase in ABCA1-mediated cholesterol efflux [5].

EP

It has been suggested that pre-β1 HDL is a risk marker of cardiovascular disease (CVD) [6]. In cross sectional case control studies, higher plasma levels of pre-β1 HDL have been

AC C

reported in patients with prevalent coronary heart disease [7-9], and is also associated with increase in carotid intima media thickness (IMT) in non-diabetic subjects [10]. Type 2 diabetes is associated with high risk of CVD but there are relatively little data on pre-β1 HDL in subjects with type 2 diabetes. Hirayama et al have shown that pre-β1 HDL is elevated in a small group of type 2 diabetic patients and pre-β1 HDL is associated with carotid IMT [11]. In contrast, no change or a decrease in pre-β1 HDL concentration has been reported in patients with type 2

3

ACCEPTED MANUSCRIPT

diabetes by other groups [12-14]. The available studies so far in type 2 diabetes are all limited by the small number of subjects. The aim of this study is therefore to evaluate pre-β1 HDL in a much larger cohort of type 2 diabetic patients with and without CVD, and to determine whether

RI PT

abnormalities in pre-β1 HDL concentration are present irrespective of their HDL-cholesterol

SC

levels and the impact on cholesterol efflux.

Materials and methods

M AN U

Type 2 diabetic patients with and without CVD were recruited from the diabetes clinics at Queen Mary Hospital. Healthy non-diabetic control subjects were recruited from the community. Diabetic patients were invited to participate when they attended their annual screening visit for diabetic complications. Cardiovascular disease was defined as documented evidence of coronary

TE D

heart disease (angina, myocardial infarction, significant stenosis on coronary angiography or coronary revascularisation), ischaemic stroke, and/or peripheral vascular disease according to clinical history. All diabetic patients must have stable glycemic control with no change in anti-

EP

diabetic therapy for the preceding 3 months. In order to determine whether any observed changes in pre-β1 HDL concentration were independent of HDL-cholesterol, control subjects were

AC C

matched with diabetic subjects (without CVD and not on lipid lowering therapy) for age, gender and HDL-cholesterol level. The study was approved by the Ethics Committee of the University of Hong Kong and informed consent was obtained from all subjects. A total of 500 type 2 diabetic patients without CVD, 140 male diabetic patients with CVD and 360 healthy control were recruited. Fasting blood samples were taken for the measurement of glucose, HbA1c, lipids, apoA1 and B, and pre-β1 HDL. Cholesterol efflux to serum mediated by ABCA1 was determined in a random

4

ACCEPTED MANUSCRIPT

subgroup of patients and control. Blood samples were kept at 4oC after collection and EDTAplasma and serum was separated from whole blood within 2 hours by centrifugation and stored at -80oC until assayed.

RI PT

Plasma total cholesterol and triglyceride were determined enzymatically on a Hitachi 912 analyzer (Roche Diagnostics, GmbH, Mannheim, Germany). HDL-cholesterol was measured using a homogenous method with polyethylene glycol-modified enzymes and alpha-

SC

cyclodextrin. LDL cholesterol was calculated by the Friedewald equation or measured directly if plasma triglyceride was >4.5 mmol/L. Plasma apoA1 and B were measured by rate nephelometry

M AN U

using the Beckman Array System (Beckman Instruments). Plasma pre-β1 HDL level was determined by pre-β1 HDL ELISA kit using EDTA-plasma diluted with the stabilization buffer containing 50% sucrose solution according to the manufacturer’s instructions (American Diagnostica GmbH, Pfungstadt, Germany). The intra-assay and inter-assay coefficient of

TE D

variations of the ELISA were 3.6% and 8.4% respectively. HbA1c was measured in whole blood using ion-exchange high performance liquid chromatography with the Bio-Rad Variant Haemoglobin Testing System (Bio-Rad Laboratories Inc., California, USA). Cholesterol efflux to

EP

serum mediated by ABCA1 was determined as previously described [13]. In brief, the transfer of [3H]cholesterol from Fu5AH cells expressing ABCA1 (induced by 22(R)-hydroxycholesterol

AC C

and 9-cis-retinoic acid) to the medium containing the tested serum was measured. Results were expressed as mean and standard deviation, or as median and interquartile range

if the distribution of the data were found to be skewed. Data that were not normally distributed were logarithmically transformed before analyses were made. Comparisons between two groups were done using independent sample t-test. Analysis of variance was used to compare multiple groups followed by post hoc multiple comparisons using Dunnett t-tests with the non-diabetic control as the

5

ACCEPTED MANUSCRIPT

reference group. Pearson’s correlation coefficient and multivariate regression analysis were used to test the relationships between variables.

RI PT

Results

The clinical characteristics of the subjects are shown in Table 1 and 2. Data on male and female subjects are analyzed separately because of gender differences in HDL-cholesterol. Both

SC

male and female diabetic patients had a greater body mass index (BMI) and waist circumference than their matched non-diabetic control. Male diabetic subjects with CVD were significantly older,

M AN U

had longer duration of diabetes and majority of them were on lipid lowering agents (114 on statins and 6 on fibrates). Plasma lipid and apolipoprotein levels are shown in Table 3 and 4.The diabetic patients and control subjects were well-matched for their HDL-cholesterol, and plasma total apoA1 levels were also similar. Despite having matched HDL-cholesterol levels and apoA1, plasma pre-β1

TE D

HDL level was significantly reduced in both male and female diabetic patients compared to their matched control. Diabetic patients with CVD had the lowest plasma level of pre-β1 HDL. Fiftynine percent of the diabetic patients without CVD had metabolic syndrome and there were no

EP

significant differences in pre-β1 HDL level between those with or without metabolic syndrome. There were also no significant differences in pre-β1 HDL level between patients receiving

AC C

insulin therapy and those on oral anti-diabetic agents. However, patients on statin therapy had lower pre-β1 HDL than those diabetic subjects without CVD and not receiving statins (p = 0.04). In the whole diabetic cohort, log(pre-β1 HDL) level correlated inversely with BMI (r = -0.17, p<0.01), waist circumference (r = -0.21, p<0.01) and log(triglyceride) (r = -0.15, p<0.01), and there were no association with parameters like age, duration of diabetes, HbA1c or eGFR. A significant association between HbA1c and log(pre-β1 HDL) was only seen when diabetic

6

ACCEPTED MANUSCRIPT

subjects and control were analyzed together (r = -0.13, p<0.01). On multiple logistic regression analysis, the differences in pre-β1 HDL due to diabetes status persisted even after adjusting for age, gender, BMI, smoking and triglyceride (Supplementary Table 1). Subjects on lipid lowering

RI PT

agents were excluded in the analysis to avoid the confounding effect of statin therapy.

Cholesterol efflux to serum mediated by ABCA1 was determined in a random subgroup of subjects (180 diabetic patients without CVD and 120 control subjects, male to female ratio 1:1) and

SC

was significantly reduced in the diabetic group (1.42 ± 0.52% versus 1.72 ± 0.50 respectively, p<0.01). Cholesterol efflux to serum mediated by ABCA1 correlated significantly with pre-β1 HDL

M AN U

in all subjects (r = 0.38, p<0.01, Fig. 1), or when diabetic patients and control were analyzed separately (DM: r = 0.31, p<0.01; Control: r = 0.28, p<0.05). However, there was no significant relationship with HDL-cholesterol and only a trend towards a weak association with total apoA1 (r = 0.12, p = 0.058) and HbA1c (r = -0.11, p = 0.065). Multiple linear regression analysis including

TE D

age, gender, BMI, diabetes status, smoking, apoA1, triglyceride and LDL in the model showed that pre-β1 HDL remained significantly associated with cholesterol efflux (Supplementary Table 2). We have also performed a parallel assessment of efflux capacity with the use of either whole

EP

serum or apo B-depleted serum as acceptor in a small subgroup of subjects (43 diabetic patients and 27 control subjects). Cholesterol efflux obtained using whole serum was reduced by 21 ± 7%

AC C

upon removal of apo B lipoproteins and there was a very good correlation between cholesterol efflux to whole serum and cholesterol efflux to apo B-depleted serum (r = 0.94, p<0.01).

Discussion

Pre-β1 HDL is a molecular species of plasma HDL that contains apoA1 and phospholipids and has an apparent molecular mass of approximately 70 kDa. It has been

7

ACCEPTED MANUSCRIPT

estimated that some 5% of the apoA1 in human plasma is present in this form [1]. Despite the low concentration, this pool of pre-β1 HDL is important because it is particularly effective at mediating cellular cholesterol efflux [15]. Synthesis of apoA1 and the transfer of cell-derived

RI PT

free cholesterol contributes to the generation of pre-β1 HDL particles in the circulation [15,16]. Pre-β1 HDL is both a product and a substrate in the ABCA1-mediated reaction to efflux cellular cholesterol [17]. Unesterified cholesterol from cells incorporated into pre-β1 HDL via interaction

SC

with ABCA1 provides a substrate for esterification by lecithin:cholesterol acyltransferase (LCAT). Pre-β1 HDL then becomes incorporated into larger HDL species of alpha mobility as

M AN U

esterification of cholesterol proceeds and is regenerated during the transferof cholesteryl esters by cholesteryl ester transfer protein (CETP) from alpha HDL particles to acceptor lipoproteins. Plasma factors like phospholipid transfer protein (PLTP) and hepatic lipase (HL) are also involved in this process through the remodeling of HDL particles [18,19]. Hence, the metabolism

TE D

of pre-β1 HDL particles is complex and is governed by a number of plasma factors as well as the prevailing constellation of plasma lipoproteins.

The steady state level of pre-β1 HDL in plasma is influenced by the relative efficiencies

EP

of the major metabolic processes involved in its generation and removal, and alterations in lipoprotein metabolism associated with type 2 diabetes may affect one or more of the

AC C

determinants of levels of pre-β1 HDL in plasma. We have shown that the absolute levels of preβ1 HDL is decreased in patients with type 2 diabetes. Since low HDL-cholesterol level is commonly seen in type 2 diabetes, we have deliberately compared our group of diabetic patients with a group of non-diabetic control matched for their HDL-cholesterol levels so that any abnormalities observed in pre-β1 HDL concentration could not be attributed to the diabetic subjects having lower HDL-cholesterol level than control. By doing so, we avoid the

8

ACCEPTED MANUSCRIPT

confounding effect of differences in HDL-cholesterol between control and diabetic patients. However, this may limit the generalizability of our findings. Even though we have matched the groups for age, gender and HDL-cholesterol, the diabetic patients had higher BMI, waist

RI PT

circumference and systolic blood pressure than controls. The reduction in pre-β1 HDL in our study was seen in both male and female diabetic patients and in those with or without CVD. Our results differ from previous studies, which showed that pre-β1 HDL was increased in patients

SC

with coronary heart disease [7-9]. Differences in the selection of subjects being studied might partly explain the conflicting results, as the proportion of diabetic subjects in these studies was

M AN U

small. Another reason may be due to the fact that most of our patients with CVD were on statin therapy, and statin has been shown to decrease pre-β1 HDL [20,21]. We have shown that diabetic patients receiving statin therapy had significantly lower pre-β1 HDL than those patients not receiving statins. However, pre-β1 HDL concentration was significantly decreased even in

TE D

our diabetic patients not receiving statin compared to non-diabetic controls. This would suggest that there may be changes in the metabolism of pre-β1 HDL particles in type 2 diabetes which may account for these differences. A number of diverse determinants may operate to either

EP

increase or decrease levels of pre-β1 HDL in diabetes and changes in the activities of LCAT, CETP, PLTP and HL and in the expression of cholesterol transporters like ABCA1 have all been

AC C

reported in type 2 diabetes [22-24]. Hence, whether the decrease in pre-β1 HDL in patients with type 2 diabetes is the net result of a decrease in generation and/or an increase in catabolism remains to be determined. Furthermore, the lack of a standardized method to measure pre-β1 HDL also makes comparison between different studies difficult. A number of techniques have been used to assay pre-β1 HDL and these include ultrafiltration-isotope dilution [25], Western blotting of two-dimensional gels [8] and immunoassay [26]. All the available methods used to

9

ACCEPTED MANUSCRIPT

assess pre-β1 HDL do not reflect the dynamic processes that regulate pre-β1 HDL concentration. Since we want to measure pre-β1 HDL in a large number of subjects, we have used a sandwich enzyme immunoassay with a monoclonal antibody that is specific for pre-β1 HDL in our study.

RI PT

Even though the mechanism(s) for the reduction in pre-β1 HDL remain unclear, our study confirms that plasma pre-β1 HDL concentration is an important determinant of the ability of plasma/serum to induce cholesterol efflux. Cholesterol efflux to serum mediated by ABCA1 was

SC

clearly related to plasma pre-β1 HDL concentration but not to HDL-cholesterol level, and there was only a weak association with plasma total apoA1. This is in keeping with the finding that

M AN U

HDL-cholesterol levels only explain a fraction of the variance in cholesterol efflux capacity [27]. Ex vivo measurement of the capacity of an individual’s HDL specimen or serum to remove cholesterol from cultured cholesterol-loaded macrophages is predictive of prevalent and incident coronary heart disease and it been suggested that cholesterol efflux capacity is potentially a novel

TE D

cardiovascular risk marker [28]. An independent inverse association between HDL cholesterol efflux capacity and incident cardiovascular events has been demonstrated both in the Dallas Heart Study and in the European Prospective Investigation of Cancer-Norfolk (EPIC-Norfolk)

EP

study [29,30]. The methodology used to measure cholesterol efflux was different from ours in those two studies. Cholesterol efflux capacity was measured using J774 macrophages, and apo

AC C

B-depleted plasma or serum in the Dallas Heart Study and EPIC-Norfolk Study respectively. There is currently no established standard for the measurement of cholesterol efflux capacity ex vivo. Differences in the cell lines and the means to label cholesterol (fluorescence-labelled or radiolabeled cholesterol) employed in the assay, as well as the use of either whole plasma/serum, apo B-depleted plasma/serum or isolated HDL specimens as acceptors, make direct comparisons between studies challenging [2,31]. Although we have used whole serum as acceptor in our

10

ACCEPTED MANUSCRIPT

assay, we have found that there was a very good correlation between cholesterol efflux to whole serum and to apo B-depleted serum. Hence, whether low pre-β1 HDL particles and the associated reduction in cellular cholesterol efflux is predictive of CVD in patients with type 2

RI PT

diabetes warrants further investigation.

The strength of our study lies in the large sample size, and the diabetic patients and control were matched for their HDL-cholesterol levels. Our study does have several limitations.

SC

We have only measured the static concentrations of pre-β1 HDL without assessing the dynamic processes regulating the concentrations of pre-β1 HDL in the circulation. To study the kinetics

M AN U

and metabolism of pre-β1 HDL will require the use of stable isotope methodology [32]. Because of the cross-sectional design, we can only demonstrate associations and not causal relationships between pre-β1 HDL and serum cholesterol efflux capacity. We have not evaluated the relationship between pre-β1 HDL and CVD as majority of the patients with CVD in our study

TE D

were receiving statin therapy and statin can affect pre-β1 HDL concentration [20,21]. The potential relationship of pre-β1 HDL to atherosclerosis in type 2 diabetes will need to be determined in a prospective manner.

EP

In conclusion, plasma pre-β1 HDL level was significantly decreased in type 2diabetes and was associated with a reduction in cholesterol efflux mediated by ABCA1. Our data would

AC C

suggest that low pre-β1 HDL might cause impairment in reverse cholesterol transport in type 2 diabetes.

11

ACCEPTED MANUSCRIPT

Conflict of interest The authors declared they do not have anything to disclose regarding conflict of interest with

RI PT

respect to this manuscript.

Financial support

SC

This study is supported by an Endowment Fund established for the “Sir David Todd

M AN U

Professorship in Medicine” awarded to KCT.

Author contributions

SWS performed the laboratory assays and analysed the data. YW recruited the subjects and

TE D

collected clinical data. KCT designed and supervised the study, analysed and interpreted the data, and critically revised the manuscript. All authors approved the final version of the

AC C

EP

manuscript.

12

ACCEPTED MANUSCRIPT

References 1.

Movva R, Rader DJ. Laboratory Assessment of HDL heterogeneity and function. Clin Chem 2008;54:788–800 Rosenson RS, Brewer HB Jr, Ansell B, et al. Translation of high-density lipoprotein

RI PT

2.

function into clinical practice: current prospects and future challenges. Circulation 2013;128:1256-67

Rosenson RS, Brewer HB Jr, Chapman MJ, et al. HDL measures, particle heterogeneity,

SC

3.

proposed nomenclature, and relation to atherosclerotic cardiovascular events. Clin Chem

4.

M AN U

2011;57:392-410.

Phillips MC. Molecular mechanisms of cellular cholesterol efflux. J Biol Chem. 2014;289:24020-9.

5.

Kempen HJ, Asztalos BF, Moerland M, et al. High-density lipoprotein subfractions and capacities after

infusion

of MDCO-216

(Apolipoprotein

A-

TE D

cholesterol efflux

IMilano/Palmitoyl-Oleoyl-Phosphatidylcholine) in healthy volunteers and stable coronary artery disease patients. Arterioscler Thromb Vasc Biol. 2016 36:736-42. Kane JP, Malloy MJ. Prebeta-1 HDL and coronary heart disease. Curr Opin Lipidol

EP

6.

2102;23:367-71

Hattori H, Kujiraoka T, Egashira T, et al. Association of coronary heart disease with pre-

AC C

7.

beta-HDL concentrations in Japanese men.Clin Chem. 2004;50:589-95. 8.

Asztalos BF, Cupples LA, Demissie S, et al. High-density lipoprotein subpopulation profile and coronary heart disease prevalence in male participants of the Framingham Offspring Study. Arterioscler Thromb Vasc Biol. 2004;24:2181-7.

13

ACCEPTED MANUSCRIPT

9.

Sethi AA, Sampson M, Warnick R, et al. High pre-beta1 HDL concentrations and low lecithin: cholesterol acyltransferase activities are strong positive risk markers for ischemic heart disease and independent of HDL-cholesterol. Clin Chem. 2010;56:1128-37.

RI PT

10. de Vries R, Perton FG, van Tol A, Dullaart RP. Carotid intima media thickness is related positively to plasma pre ß-high density lipoproteins in non-diabetic subjects. Clin Chim Acta. 2012;413:473-7.

SC

11. Hirayama S, Miida T, Miyazaki O, Aizawa Y. Pre beta1-HDL concentration is a predictor of carotid atherosclerosis in type 2 diabetic patients. Diabetes Care.2007;30:1289-91.

M AN U

12. Plomgaard P, Dullaart RP, de Vries R, Groen AK, Dahlbäck B, Nielsen LB. Apolipoprotein M predicts pre-beta-HDL formation: studies in type 2 diabetic and nondiabetic subjects. J Intern Med. 2009;266:258-67.

13. Zhou H, Tan KC, Shiu SW, Wong Y. Cellular cholesterol efflux to serum is impaired in

TE D

diabetic nephropathy. Diabetes Metab Res Rev. 2008;24:617-23. 14. Pan J, Shilian P, Ishida B, Wu X, Kane JP, Malloy MJ, Charles MA. Effect of niacin on preβ-1 high-density lipoprotein levels in diabetes. Metabolism. 2011;60:292-7.

EP

15. Rosenson RS, Brewer HB Jr, Davidson WS, et al. Cholesterol efflux and atheroprotection: advancing the concept of reverse cholesterol transport. Circulation 2012;125:1905-19.

AC C

16. Lee JY, Parks JS. ATP-binding cassette transporter AI and its role in HDL formation. Curr Opin Lipidol. 2005;16:19-25. 17. Duong PT, Weibel GL, Lund-Katz S, Rothblat GH, Phillips MC. Characterization and properties of prebeta-HDL particles formed by ABCA1-mediated cellular lipid efflux to Apo A-I. J Lipid Res. 2008;49:1006–1014

14

ACCEPTED MANUSCRIPT

18. Santamarina-Fojo S, Lambert G, Hoeg JM, Brewer HB Jr. Lecithin-cholesterol acyltransferase: role in lipoprotein metabolism, reverse cholesterol transport and atherosclerosis. Curr Opin Lipidol. 2000;11:267-75.

RI PT

19. Rye KA, Barter PJ. Formation and metabolism of prebeta-migrating, lipid-poor apolipoprotein A-I. Arterioscler Thromb Vasc Biol.2004;24:421-8.

20. Kawano M, Nagasaka S, Yagyu H, Ishibashi S. Pitavastatin decreases plasma prebeta1-

patients. J Atheroscler Thromb 2008;15:41-6

SC

HDL concentration and might promote its disappearance rate in hypercholesterolemic

M AN U

21. Dallinga-Thie GM, van Tol A, Dullaart RP; Diabetes Atorvastatin lipid intervention (DALI) study group. Plasma pre beta-HDL formation is decreased by atorvastatin treatment in type 2 diabetes mellitus: role of phospholipid transfer protein. Biochim Biophys Acta 2009;1791:714-18

TE D

22. Borggreve SE, De Vries R, Dullaart RP. Alterations in high-density lipoprotein metabolism and reverse cholesterol transport in insulin resistance and type 2 diabetes mellitus: role of lipolytic enzymes, lecithin:cholesterol acyltransferase and lipid transfer proteins. Eur J Clin

EP

Invest. 2003;33:1051-69.

23. Tan KC, Shiu SW, Wong Y, Tam S. Plasma phospholipid transfer protein activity and

AC C

subclinical inflammation in type 2 diabetes mellitus. Atherosclerosis. 2005;178:365-370. 24. Machado-Lima A, Iborra RT, Pinto RS, et al. In type 2 diabetes mellitus glycated albumin alters macrophage gene expression impairing ABCA1-mediated cholesterol efflux. J Cell Physiol. 2015;230:1250-7.

15

ACCEPTED MANUSCRIPT

25. O'Connor PM, Naya-Vigne JM, Duchateau PN, et al. Measurement of prebeta-1 HDL in human plasma by an ultrafiltration-isotope dilution technique. Anal Biochem. 1997;251:234-40.

RI PT

26. Miyazaki O, Kobayashi J, Fukamachi I, Miida T, Bujo H, Saito Y. A new sandwich enzyme immunoassay for measurement of plasma pre-beta 1-HDL levels.J Lipid Res 2000;41:20838.

SC

27. Hutchins PM, Heinecke JW. Cholesterol efflux capacity, macrophage reverse cholesterol transport and cardioprotective HDL. Curr Opin Lipidol. 2015;26:388-93.

M AN U

28. Mody P, Joshi PH, Khera A, Ayers CR, Rohatgi A. Beyond coronary calcification, family history, and C-reactive protein: cholesterol efflux capacity and cardiovascular risk prediction. J Am Coll Cardiol. 2016;67:2480-7.

29. Rohatgi A, Khera A, Berry JD, et al. HDL cholesterol efflux capacity and incident

TE D

cardiovascular events. N Engl J Med 2014; 371:2383–2393.

30. Saleheen D, Scott R, Javad S, et al. Association of HDL cholesterol efflux capacity with incident CHD events: a prospective case control study. Lancet Diabetes Endocrinol

EP

2015;3:507-513

31. Brownell N, Rohatgi A. Modulating cholesterol efflux capacity to improve cardiovascular

AC C

disease. Curr Opin Lipidol. 2016;27:398-407 32. Li X, Stolinski M, Umpleby AM. Development of a method to measure preβ HDL and αHDL apoA-I enrichment for stable isotopic studies of HDL kinetics. Lipids. 2012;47:1011-8

16

ACCEPTED MANUSCRIPT

Table 1: Clinical characteristics of male subjects.

DM without CVD

DM with CVD

N=180

N=250

N=140

49.8 ± 6.7

49.4 ± 6.6

RI PT

Age (years)

Control

62.2 ± 8.8**

11 (7 – 14)

15 (10 - 21)#

BMI (kg/m2)

25.0 ± 2.9

25.9 ± 3.9**

26.7 ± 4.4**

Waist circumference (cm)

85.9 ± 8.5

89.9 ± 10.8**

93.3 ± 10.4**

Smoker (%)

27.7

25.6

17.1

Insulin (%)

-

Lipid lowering therapy (%)

-

M AN U

SC

-

Duration of diabetes (years)

42.8

62.8

0

85.7

120. 1 ± 17.0

126.9 ± 16.8*

141.9 ± 21.9**

DBP (mmHg

79.5 ± 9.8

79.5 ± 8.7

78.2 ± 10.1

FG (mmol/L)

5.11 ± 0.56

8.68 ± 1.60**

8.52 ± 2.11**

5.69 ± 0.50

8.68 ± 1.60**

8.45 ± 1.41**

78.9 ± 12.6

75.5 ± 11.2

58.4 ± 14.7**

HbA1c (%)

EP

eGFR (mL/min/1.73 m2)

TE D

SBP (mmHg)

AC C

Data are expressed as mean ± SD or median (interquartile range).*p<0.05 and ** p<0.01 versus healthy controls. # p<0.01 versus DM without CVD.

17

ACCEPTED MANUSCRIPT

Table 2: Clinical characteristics of female subjects.

DM

N=180

N=250

50.0 ± 5.7

49.8 ± 6.9 11 (8 -16)

BMI (kg/m2)

24.9 ± 3.7

26.8 ± 4.5**

Waist circumference (cm)

79.6 ± 9.3

SC

Duration of diabetes (years)

RI PT

Age (years)

Control

3.8

M AN U

Smoker (%)

86.8 ± 10.6** 3.2

-

35.2

-

0

SBP (mmHg)

118.6 ± 16.1

125.9 ± 22.0**

DBP (mmHg)

73.6 ± 9.1

76.5 ± 8.5

Insulin (%)

FG (mmol/L) HbA1c (%)

4.97 ± 0.51

8.60 ± 2.85**

5.60 ± 0.46

8.43 ± 1.84**

79.4 ± 15.0

79.0 ± 18.9

EP

eGFR (mL/min/1.73 m2)

TE D

Lipid lowering therapy (%)

AC C

Data are expressed as mean ± SD or median (interquartile range).** p<0.01 versus healthy controls.

18

ACCEPTED MANUSCRIPT

Table 3: Plasma lipids, apolipoproteins and pre-β1 HDL in male subjects.

DM without CVD

DM with CVD

TC (mmol/L)

5.09 ± 0.85

4.90 ± 0.91

4.14 ± 0.82**

TG (mmol/L)

1.40 (1.00 – 1.80)

1.40 (0.90 – 2.00)

1.35 (1.00 – 1.90)

LDL (mmol/L)

3.08 ± 0.75

2.99 ± 0.79

2.28 ± 0.71**

HDL (mmol/L)

1.17 ± 0.25

1.15 ± 0.28

ApoA1 (g/L)

1.28 ± 0.20

1.26 ± 0.22

1.25 ± 0.22

Apo B (g/L)

0.96 ± 0.20

0.94 ± 0.23

0.85 ± 0.24**

16.7 (11.6- 24.3)**

15.9 (10.4 - 23.2)**

SC

M AN U

Pre-β1 HDL (ug/mL)

RI PT

Control

19.0 (12.6- 26.5)

1.13 ± 0.30

Data are expressed as mean ± SD or median (interquartile range).**p<0.01 versus healthy

AC C

EP

TE D

controls.

19

ACCEPTED MANUSCRIPT

Table 4: Plasma lipids, apolipoproteins and pre-β1 HDL in female subjects.

DM

TC (mmol/L)

5.02 ± 0.86

4.96 ± 0.89

TG (mmol/L)

1.00 (0.78 – 1.30)

1.20 (0.83 – 1.90)

LDL (mmol/L)

3.06 ± 0.84

2.90 ± 0.80

HDL (mmol/L)

1.32 ± 0.24

1.37 ± 0.37

ApoA1 (g/L)

1.43 ± 0.20

Apo B (g/L)

0.85 ± 0.22

SC

RI PT

Control

1.39 ± 0.22

M AN U

0.88 ± 0.21

Pre-β1 HDL (µg/mL)

21.6 (14.6 – 30.5)

20.0 (14.0 – 27.2)*

Data are expressed as mean ± SD or median (interquartile range).* p<0.05 versus healthy

AC C

EP

TE D

controls.

20

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 1: Association between ABCA1-mediated cholesterol efflux to serum and plasma pre-β1 HDL.

21

ACCEPTED MANUSCRIPT Highlights Pre-β1 HDL was reduced in type 2 diabetes despite having similar HDL-C as control

AC C

EP

TE D

M AN U

SC

RI PT

ABCA1-mediated cholesterol efflux to serum was impaired in type 2 diabetes Low pre-β1 HDL level was associated with impaired cholesterol efflux