NMR-Based Lipid Profiling of High Density Lipoprotein Particles in

Subscriber access provided by University of Colorado Boulder

Article

NMR-based lipid profiling of High Density Lipoprotein particles in healthy subjects with low, normal and elevated HDL-cholesterol Christina E. Kostara, Vasilis Tsimihodimos, Moses S. Elisaf, and Eleni T. Bairaktari J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00975 • Publication Date (Web): 13 Mar 2017 Downloaded from http://pubs.acs.org on March 13, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

NMR-based lipid profiling of High Density Lipoprotein particles in healthy subjects with low, normal and elevated HDL-cholesterol

1

Christina E. Kostara, Ph.D., Chemist

2

Vasilis Tsimihodimos, MD, Ph.D., Assistant Professor of Internal Medicine

2

Moses S. Elisaf, MD, FRSH, Professor of Internal Medicine

1,

* Eleni T. Bairaktari, Ph.D., Associate Professor of Clinical Chemistry

1

Laboratory of Clinical Chemistry, 2Department of Internal Medicine, Faculty of Medicine,

School of Health Sciences, University of Ioannina, 451 10, Ioannina, Greece

*Address for correspondence and reprint requests Eleni T. Bairaktari, Ph.D., Eur Clin Chem Head of Biochemistry Laboratory, University Hospital of Ioannina Associate Professor of Clinical Chemistry, Laboratory of Clinical Chemistry, Faculty of Medicine, School of Health Sciences, University of Ioannina 451 10 Ioannina, GREECE Phone: +30-2651007620, Fax: +30-2651007871 E-mail: [email protected]

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Recent studies suggest that the cholesterol content of HDL (High Density Lipoproteins) lipoproteins may provide limited information on their antiatherogenic properties and that the composition and particles’ structure provide more information on their functionality. We used NMR-based (Nuclear Magnetic Resonance-based) lipidomics to study the relationships of serum HDL-C (HDL-Cholesterol) levels with the lipid composition of HDL particles in three groups of subjects selected on the basis of their HDL-C levels. Subjects with low and high HDL-C levels exhibited differences in HDL lipidome compared to those with normal HDL-C levels. In Pattern Recognition analysis, the discrimination power among all groups was of high significance. Low HDL-C group presented enrichment of the core in triglycerides and depletion in cholesterol esters, whereas high HDL-C group showed a decrease in triglycerides content. Additionally, as HDL-C increases, all lipid classes are esterified with higher percentage of unsaturated than saturated fatty acids. In addition to the aforementioned differences, the surface layer is enriched in sphingomyelin and free cholesterol at high HDL-C level group. NMR–based lipidomic analysis of HDL can be particularly useful since it provides insights into molecular features, helps in the characterization of the atheroprotective function of HDL lipoproteins and in the identification of novel biomarkers of cardiovascular risk.

Keywords NMR, Lipidomics, Lipoproteins, HDL, lipid composition, phospholipids, phosphatidylcholine, sphingomyelin


Page 2 of 38

Page 3 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Cardiovascular disease (CVD) remains the main cause of mortality worldwide, representing 31% of all global deaths.1 The cause of CVD is multifactorial, with both environmental and genetic factors contributing. Patients with established cardiovascular disease and individuals at high cardiovascular risk (due to the presence of one or more risk factors such as hypertension, diabetes and hyperlipidaemia) need early detection and proper management.1-3 Dyslipidemia, that is defined as abnormal blood lipid levels, [that is elevated total cholesterol, low-density lipoprotein (LDL), and triglyceride, and decreased high-density lipoprotein (HDL)], is one of the most important and well-known risk factors for CVD.2-3 Epidemiological and clinical studies have over the years shown that low HDL-cholesterol (HDL-C) concentration in plasma is associated with an increased risk for the development of CVD.4-5 The pathophysiological basis of this inverse relationship of HDL-C with cardiovascular risk was thought to be the ability of HDL-C levels to reflect the efficiency of the main antiatherogenic function of the HDL particles which is the clearance of excess cholesterol from peripheral tissues. On the other hand, experimental studies suggested that the antiatherogenic functionality of HDL particles is not determined solely by the quantity of their cholesterol content6-7 but also from the particles’ quality which is closely related to their compositional and structural features.8-12 This hypothesis was further supported by the failure of pharmaceutical interventions that increase HDL-C levels to prevent subsequent cardiovascular events.13-14 Thus, the need for a systematic analysis of the different components of HDL and their modifications was reinforced so as to better understand the contribution of HDL components in promoting reverse cholesterol transport. This mechanism, also termed "cholesterol efflux capacity"14-15 occurs by accepting cholesterol from lipid-loaded macrophages, thus protecting against the development of atherosclerosis. HDL particles consist of a hydrophobic core containing non-polar lipids, primarily cholesterol esters and triglycerides, surrounded by a hydrophilic surface monolayer, which is mainly constituted of phospholipids and proteins. Unesterified sterols are predominantly located on the surface monolayer and partially penetrating the core.16 In the everyday clinical practice, the HDL particles are estimated only by the measurement of their cholesterol content. However, recently, the evolution of available



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

lipidomic methodologies have facilitated the detailed characterization of all lipid classes present in plasma as well as in lipoprotein fractions.16-17 In our laboratory, we established a combination of analytical techniques that allows the quantification of all major lipid classes present in serum lipoprotein particles in a relatively short experimental time. We used this lipidomic methodology for the global characterization of lipid classes present in HDL lipoprotein fractions, derived from otherwise subjects with normal, low and high HDL-C levels. All lipid classes of the HDL particles, isolated by precipitation from non-HDL lipoproteins, were extracted and then a proton NMR-based approach provided a rapid and non-destructive lipid assay of surface and core particle’s lipids such as cholesterol, triglycerides and phospholipids as well as fatty acid pattern (saturated, unsaturated, linoleic and ω-3 fatty acids and degree of unsaturation of fatty acids).


Page 4 of 38

Page 5 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Materials and Methods Subjects Sixty healthy individuals receiving an annual medical checkup at the Outpatient Lipid Clinic of the University Hospital of Ioannina were included in the study. The total group was divided in three subgroups according to serum HDL-C levels: 20 individuals with low (< 40 mg/dl), 20 with normal (40-59 mg/dl) and 20 with high HDL-C levels ( 60 mg/dl) (Table 1). All groups were matched according to age and the rest of the lipid parameters e.g. total, LDLand non-HDL cholesterol and triglycerides levels to minimize the confounding effect of the serum lipid profile on HDL lipid metabolism (Table 1). No individual had any evidence of cardiovascular disease according to history, clinical examination, or electrocardiogram. None of the participants were taking lipid-lowering drugs or any other medication known to affect lipid metabolism, including hormonal replacement therapy, during the last 12 weeks. Individuals with hypertension (blood pressure 140/90 mmHg on repeated measurements), diabetes mellitus (fasting blood glucose 126 mg/dl), obesity (body mass index 30 kg/m2), metabolic syndrome or thyroid, hepatic, or renal disease as well as subjects known to ingest more than two alcoholic drinks daily or taking vitamin supplements were excluded from the study. The Ethics Committee of the University Hospital of Ioannina approved the study and written informed consent was obtained from each participant.

Sample Collection Venous blood samples were collected in the morning after an overnight fast from all healthy individuals. Serum was separated by centrifugation at 1500 g for 15 min for the determination of serum lipid parameters and one 1.5 ml aliquot was stored at –80oC until NMR analysis.

Determination of Serum Lipid Parameters Serum lipid parameters were measured on an Olympus AU5400 analyzer (Beckman, Hamburg, Germany) by standard procedures. Total cholesterol and triglycerides were determined enzymatically and HDL-cholesterol by a direct assay (Bechman, Hamburg,



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Germany). LDL-C was calculated by the Friedewald formula (provided that triglycerides levels were lower than 400 mg/dL or 4.5 mmol/L) and non-HDL-C was calculated by the equation: non-HDL-C = total cholesterol - HDL-C. Serum apolipoproteins AI and B were measured by immunonephelometry on a BN ProSpec System (Siemens, Marburg, Germany).

Isolation and Lipid Extraction of HDL lipoproteins HDL lipoprotein particles were isolated from non-HDL lipoproteins by precipitation with Dextran Sulfate/ MgCl2.18 The lipid extraction of HDL particles was performed as described previously.17 Briefly, the lipid content of HDL subfraction was extracted with methanol/chloroform (2:1) according to the modified Bligh and Dyer method19, dried in a stream of nitrogen and bubbled with nitrogen prior to recording spectrum in order to remove oxygen.

1

H NMR Spectroscopy Analysis The lipid extracts of HDL lipoproteins prepared as described above, were redissolved in

500μl mixture of deuterated methanol/chloroform (2:1, v/v).20 The 1H NMR spectra were recorded at 298 Κ on a 500 MHz Bruker Avance DRX NMR spectrometer (NMR Center, University of Ioannina) operating at a field strength of 11.74 Tesla and running on TopSpin 2.1 suite. A standard Bruker “zgpr” pulse program that applies a presaturation pulse sequence for the suppression of the residual water signal at about 4.75 ppm during the relaxation delay was used. Spectra were acquired in the Fourier transform (FT) mode with 32K data points, 128 free induction decays (FID), 90° pulses, relaxation delay of 3 s and spectral widths of 5000 Hz. All FIDs were multiplied by an exponential weighting function corresponding to a 0.3Hz line-broadening factor prior to Fourier transformation. The acquired NMR spectra were manually corrected for phase and baseline distortions (by applying a simple polynomial curve fit) with TopSpin 2.1 suite and referenced to the methanol peak (δ 1Η 3.30). Structural assignment of lipid molecules were identified as described previously.17 Quantification of the lipids was based on the integration of selected signals, corrected for the number of protons and then normalized with respect to the C-18 methyl peak of cholesterol signal at 0.68 ppm. Given that this peak of cholesterol attributed to total HDL-


Page 6 of 38

Page 7 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


cholesterol is measured by the enzymatic assay, the absolute concentration of all HDL lipid classes was calculated and expressed as mmol/L (Table S1 in Supporting Information). In addition, the lipid composition of HDL lipoproteins was expressed as % percentage of the total lipids presented on HDL particles (see Results).

Data Analysis Statistical Analysis Statistical analysis of serum lipid parameters and HDL lipid composition was performed with Statistica Ver. 7.0 (StatSoft Inc. Tulsa, OK). Values were expressed as mean value  standard deviation (SD) and compared by using t-test. Significance levels were set at 0.05.

Multivariate Data Analysis NMR Data Reduction: The 1H NMR spectra were automatically reduced using AMIX 3.9 software (Bruker Biospin Corporation) to continuous integral bins of equal width of 0.03 ppm corresponding to the chemical shift range δ 1H, 0.49 – 5.98. The area between 4.58 and 5.00 ppm was excluded to remove any effect of variation from the suppression of the water resonance as well as the area between 3.24 and 3.50 ppm containing the resonance of the residual methanol solvent. All data was normalized by dividing each integral segment by the total area of the spectrum in order to compensate for the differences in overall concentration between individual HDL lipid extracts. The resulting data matrix was imported into SIMCA-P+ 14 software (Umetrics, Umea, Sweden) for multivariate data analysis (MVDA) analysis. Prior to the analysis, the NMR lipidomic data were mean centered scaled. "Partial Least Squares Discriminant Analysis (PLS-DA) and Orthogonal projection to PLSDA (OPLS-DA) were applied to construct statistical models of HDL lipidomic data. PLS-DA was used to find the best possible discriminant model that separates subjects with low from both those with normal HDL-C levels and high HDL-C levels on the basis of their variables X (NMR spectral bins).21 Orthogonal projection to PLS-DA, eliminated the uncorrelated systemic variation and described the maximum separation based on class membership.22 The interpretation of both PLS-DA and OPLS-DA models was based on the score and regression coefficient plots. The score plot was used to reveal observations lying outside the 0.95



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Hotteling’s T2 ellipse and to detect any grouping trend or separation, whereas the regression coefficient plot was used to show the contribution of all spectral regions (corresponding to lipid constituents) to the grouping trend or separation seen in the score plot. These spectral regions (loadings) were transferred in a table and ranked according to the coefficient value for better visualization of the most significant changes in lipid profiling. The quality of the models evaluated by three parameters (R2X, R2Y, and Q2) which were calculated by the default leaveone-out procedure. The quality of the models evaluated by three parameters (R2X, R2Y, and Q2) which were calculated by the default leave-one-out procedure. R2X and R2Y parameters were used to quantify the goodness of fit, whereas Q2 used to assess the predictive ability of the model. The values of the goodness of fit parameters vary between 0 and 1, where 1 means a perfectly fitting model and 0 no fit at all. When the value of parameter Q2, is higher than 0.5 the predictive capability of the model considered ‘good’ and when is higher than 0.9 considered ‘excellent’.21 The significance of the OPLS-DA HDL models was also examined via cross-validated ANOVA (CV-ANOVA) test. When the calculated CV-ANOVA p value is 0.5) as displayed in Figure 2a. The CV-ANOVA p value was

NMR-Based Lipid Profiling of High Density Lipoprotein Particles in

Recommend Documents