LDL subclass lipidomics in Atherogenic Dyslipidemia: Effect of Statin therapy on Bioactive lipids and Dense LDL

M. John Chapman, BSc (Hons), PhD, DSc*, Alexina Orsoni, PhD, Ricardo Tan, BSc, Natalie A. Mellett, BSc, Anh Nguyen, BSc, Paul Robillard, BSc, Philippe Giral, MD, Patrice Thérond, PhD and Peter J. Meikle, PhD* 1 Endocrinology Metabolism Division, Pitie–Salpetriere University Hospital, Sorbonne University and National Institute for Health and Medical Research (INSERM), Paris, France (Drs Chapman and Robillard); 2 Service de Biochimie, AP–HP, HUPS, Bicetre University Hospital, Le Kremlin Bicetre (Drs Orsoni and Therond), and EA 7357, Paris–Sud University and Paris–Saclay University, Chatenay–Malabry, France (Dr Therond); 3 Metabolomics Laboratory, Baker Heart and Diabetes Institute, Melbourne, Victoria 3004, Australia (Drs Chapman, Tan, Nguyen, Mellett, and Meikle; and 4 INSERM UMR1166 and Cardiovascular Prevention Units, ICAN–Institute of CardioMetabolism and Nutrition, AP–HP, Pitie–Salpetriere University Hospital, Paris, France (Dr Giral)

glucose (FPG) ≥5.6 mmol/l (100 mg/dl). Key exclusion criteria were fasting triglyceride levels >400 mg/dl (4.5mmol/l), LDL-C >190 mg/dl (4.9 mmol/l) and excessive obesity defined as BMI > 35 kg/m². Recruitment was focused on subjects whose lifestyle and dietary profile was evaluated as consistent over time. Importantly, comprehensive lifestyle and dietary counseling throughout the study time course were reflected in a stable BMI (31.7±0.5 and 31.8±0.7 kg/m² at baseline and 180 days respectively) and by a stable homeostasis model of insulin resistance (2.7±0.5 and 2.2±0.3 at baseline and 180 days respectively; NS). Waist circumference was unchanged over the course of the study. All participants had been non-smokers for at least 12 months prior to inclusion and had previously smoked less than 25 cigarettes /day on a regular basis. Study participants had no history of cardiovascular disease or type 2 diabetes.

Exclusion criteria:
All subjects included in the study did not meet any of the following exclusion criteria: • Women • Non-Caucasian • Excessive obesity defined as BMI above 35 kg/m2, rounded to the nearest whole number • LDL-C >190 mg/dl at screening • Fasting TGs >400 mg/dl at screening • Diabetes mellitus, defined as a fasting glucose >7 mM, or taking diabetic therapy at screening their hypertension controlled with a calcium channel blocker but must not receive treatment with a diuretic, beta-blocker, angiotensin converting enzyme inhibitor, or angiotensin II Receptor blocker. If the patient has previously received treatment with these therapies, they must have been discontinued at least 2 months previously.
• Any conditions that cause secondary dyslipidemia or increase the risk of statin therapy including alcoholism, autoimmune disease, nephrotic syndrome, uremia, any viral hepatitis clinically active within 12 months before study entry, obstructive hepatic or biliary disease, dysglobulinemia or macroglobulinemia, multiple myeloma, glycogen storage disease, porphyria, and uncontrolled hypothyroidism or hyperthyroidism. Controlled thyroid disease [normal serum thyroid stimulating hormone and stable therapy for at least 3 months] is permitted.
• History of pancreatic injury or pancreatitis, or impaired pancreatic function/injury as indicated by abnormal lipase • Liver injury as indicated by serum transaminase levels (alanine aminotransferase/serum glutamic pyruvic transaminase, aspartate aminotransaminase/serum glutamic oxaloacetic transaminase) >3× upper limit of the reference range (ULRR).
• Impaired renal function as indicated by serum creatinine levels >1.5× ULRR at screening or estimated glomerular filtration rate (eGFR) by Cockroft formula <60 ml/min.
• History of any muscle disease or unexplained elevation (>3× ULRR) of serum creatine kinase • Any surgical or medical condition that might significantly alter the absorption, distribution, metabolism, or excretion of the study drug, including the following: history of major gastrointestinal tract surgery (e.g., gastrectomy, gastroenterostomy or small bowel resection), gastritis or inflammatory bowel disease, current active ulcers, or gastrointestinal or rectal bleeding.
• Current obstruction of the urinary tract or difficulty in voiding likely to require intervention during the course of the study • Severe acute illness or severe trauma in the preceding 3 months • Evidence of symptomatic heart failure (New York Heart Association class III or IV): significant heart block or cardiac arrhythmia • History of uncontrolled complex ventricular arrhythmias, uncontrolled atrial fibrillation/flutter or uncontrolled supraventricular tachycardias with a ventricular response rate of >100 beats/min at rest. Patients whose electrophysiological instability is controlled with a pacemaker or implantable cardiac device are eligible.
• History of drug abuse

Plasma lipid and apolipoprotein profiles
Plasma levels of triglycerides (TG), total cholesterol (TC), HDL-C, LDL-C, and Lp(a) at baseline (D0) and at 180 days of statin treatment (D180) were quantified spectrophotometrically using Randox ® (Randox Laboratories, Crumlin, N. Ireland) reagent kits (2). Non-HDL-C was calculated as TC minus HDL-C. Plasma concentrations of apoAI and apoB were determined by immunonephelometry on a Dade-Behring ® autoanalyzer with Siemens ® reagents. Intra-and inter-assay coefficients of variation were <10% for all assays.

Determination of plasma LCAT activity and plasma CETP mass and activity
The activity of LCAT in plasma was determined using an assay kit involving a fluorescent substrate (Roar Biomedical®) according to the instruction manual (2). CETP mass in plasma was measured using a sandwich immunoassay (ALPCO Diagnostics®), according to the instruction manual (2).
Plasma CETP activity was determined using a fluorescent ex-vivo CETP activity assay kit (Roar Biomedical®) according to the manufacturer's instructions (2); this assay measures CETP activity against an exogenous substrate.

Determination of secretory phospholipase A2 mass in plasma
The plasma mass of secretory PLA2 (human type IIA) mass was determined using the colorimetric sPLA2 assay kit (Cayman Chemical Company, Ann Arbor, MI, USA) based on 96 well plates according to the manufacturer's instructions. The synthetic substrate is the 1,2-dithio analogue of diheptanoyl thiophosphorylcholine. Upon hydrolysis of the thioester bond at the sn-2 position by PLA2, free thiols are detected using DTNB (5,5-dithio-bis-(2-nitrobenzoic acid) at 414 nm with a microwell plate reader (Dynex). Bee venom PLA2 was used as a positive control. Results are expressed as time-dependent absorbance increase using 5 points in time with 1-minute increments.
Values are means of duplicate or triplicate measurements. Performance characteristics conformed to those described in the manufacturer's instruction manual.

Determination of oxidized LDL concentration in plasma
Circulating levels of oxidized LDL were quantitated with a solid phase two-site ELISA in a sandwich technique based on two different monoclonal antibodies and in accordance with the manufacturer's instruction manual (Mercodia, Sweden). This assay determines the content of aldehyde-conjugated epitopes in LDL-apoB100, and is therefore an indirect measure of the oxidative modification of LDL.
Samples were assayed in duplicate or triplicate in multiwell plates; following reaction with peroxidase-conjugated anti-oxidised LDL, the absorbance at 450nm was read in each sample well using a Dynex multiwell plate reader (Perkin-Elmer). Intra-and inter-assay coefficients of variation were <10% for this assay.

Determination of paraoxonase activity in serum
Paraoxonase activity was assayed as the organophosphatase activity of serum at baseline and after statin treatment (D180) using the EnzChek paraoxonase assay kit (Invitrogen, Thermo Fisher, Carlsbad, CA, USA). This assay is based on the hydrolysis of a fluorigenic organophosphate analogue; assays were performed in duplicate according to the manufacturer's instructions, and readout conducted on a microwell plate reader (TECAN) at 350nm.

Determination of serum amyloid A (SAA) in plasma
As an acute phase reactant, SAA was assayed in plasma samples at baseline and after statin treatment (D180). As 3 subjects displayed hsCRP levels at baseline superior to 5mg/L, they were excluded from SAA assays. Plasma SAA levels were determined with a solid phase two-site ELISA based on a biotin-streptavidin-peroxidase system in 96 well plates according to the manufacturer's instructions, which included a standard curve based on human recombinant SAA (Invitrogen, Thermo Fisher, Carlsbad, CA, USA). Readout was performed at 450nm on a microwell plate reader (Dynex) following reaction with DTNB as substrate. Values are means of duplicate or triplicate assays.

Determination of plasma high sensitivity C-reactive protein, pro-and anti-inflammatory cytokines, soluble cytokine receptors, growth factors, and adhesion proteins
Determination of high sensitivity C-reactive protein (hsCRP) in plasma samples was performed using the MetS Array II, a multiplex (biochip) array, according to the manufacturer's instruction manual (Randox Laboratories, Crumlin, N. Ireland); quantitation was based on a chemiluminescent signal. After development, the multiplex arrays were read on a Randox Investigator BioChip Reader (Randox Laboratories, Crumlin, N. Ireland). The lower limit of the sensitivity of the hs-CRP assay was <0.67 μg/mL. Intra-and inter-assay CVs were in the range from 4 to 15%. It is noteworthy that 3 subjects displayed baseline hsCRP values in excess of 5mg/L, potentially due to mild infection; these subjects were excluded.
Pro-and anti-inflammatory cytokines, soluble cytokine receptors, growth factors and adhesion proteins were assayed as described earlier using multiplex biochip arrays (Randox Laboratories, Crumlin, N. Ireland) by the same technology as that indicated above for hsCRP, and in accordance with the manufacturers' instructions (13). Additional details of assay sensitivities are included in the Manufacturer's instruction manual.

Preparative isolation of plasma lipoprotein subfractions at baseline and following statin treatment
A single step, isopycnic non-denaturing density gradient procedure was employed to preparatively fractionate LDL subpopulations from plasma samples at each timepoint (D0 and D180) on the basis of their hydrated densities by ultracentrifugation in a Beckman SW41 Ti rotor at 40,000 rpm for 44 hours in a Beckman Optima XPN-80 ultracentrifuge at 15°C (14). Circulating LDL particles exhibit hydrated densities ranging from 1.019 to 1.063 g/mL (14).  14).

Chemical analysis of lipoprotein subfractions isolated by isopycnic density gradient ultracentrifugation:
Total cholesterol (TC), free cholesterol (COH), phospholipid (PL), triglyceride (TG) and total protein (TP) concentrations in the VLDL + IDL subfraction and in LDL1 to LDL5 at baseline (D0) and after treatment (D180) were determined by spectrophotometry with the following assay kits (for TC, FC, PL, Diasys Diagnostics Systems GmbH; for TG, Biomerieux and for protein, the BCA method of Thermo Scientific Pierce) as described previously (2,3,6,12) ; it is noteworthy that the phospholipid assay estimates choline enzymatically liberated from phosphatidylcholines and sphingomyelins, and thus non-choline-containing phospholipids, and notably those containing ethanolamine, inositol, or serine are not estimated. It is equally noteworthy that enzymatic assay of TG as performed here estimates glycerol content after lipolysis; this assay therefore includes glycerol derived from partial glycerides, such as DAG (see below). The concentration of cholesteryl esters was calculated from the formula (TC-FC) x1.67 (14). The total mass of each subfraction was calculated as the sum of FC, PL, TG, CE and TP. ApoB concentration in individual apoB-containing subfractions was determined by immunoturbidimetry on the Konelab autoanalyser 20® instrument with Apolipoprotein B reagents, calibrators and "Lipotrol" as control (Thermo Scientific). Lp(a) concentration in each lipoprotein subfraction was determined on the Konelab ® autoanalyser 20 instrument using a commercial assay kit (Randox ® ). Coefficients of intra-and inter-assay variation for the individual assays ranged from 2% to 9%. Concentrations were expressed as mg/dL plasma following correction for increase in plasma volume resulting from addition of solid KBr to raise the density to 1.21 g/mL for gradient fractionation (14).

Lipoprotein-associated phospholipase A2 mass and activity in plasma and lipoprotein subfractions
The  Table S1; assays used in the clinical laboratory were employed, for which details are provided in the Methods section above.
Comparison of the MetS phenotype at baseline and following statin treatment with corresponding values in normal, healthy control subjects has been reported earlier and is presented in Suppl Table S1, indicating that MetS subjects were both hypertriglyceridemic and hypercholesterolemic at baseline, and that this dyslipidemia was largely normalised by statin treatment to lipid levels within the respective normal ranges (1,2,6).

Baseline plasma mass concentrations and % weight chemical compositions of the VLDL+IDL fraction and LDL1-5 subclasses in MetS subjects and comparison with normal healthy controls : Effect of pitavastatin treatment
A wide overall range in plasma LDL mass concentrations in healthy control subjects has been reported from our laboratory using comparable analytical methodologies (225 -303 mg/dL) (2,6). Consistent with this finding, we observed wide ranges in the mass concentrations of individual LDL subclasses in these earlier studies (Table S2)

Plasma concentrations of 23 lipid classes in the VLDL+IDL subfraction and the LDL1-5 subclasses: Effect of pitavastatin treatment
The absolute values for the plasma concentrations of the 23 lipid classes detected in the VLDL+IDL fraction and LDL1-LDL5 subclasses by mass spectrometry at baseline in obese, mixed dyslipidemic, male MetS subjects (expressed as pmol/mL plasma), and the effect of pitavastatin treatment (4mg/day; D180) on these concentrations (changes are expressed as %) are summarized in Suppl Table S3. These data are presented in schematic form as Figure 1 in the main manuscript.

Statistical comparisons between the Lipid class profiles normalized to apoB in the VLDL+IDL subfraction versus individual LDL1-5 subclasses at baseline: Effect of statin treatment.
In order to probe structural relationships between each lipid class and apoB in LDL particles,  Fig S2-AK). Furthermore, the CE/TAG ratio was up to some 10-fold lower in VLDL+IDL relative to LDL subfractions (Suppl Fig S2-AG).

Statistical comparisons between the Lipid class profiles normalized to apoB in the individual LDL1-5 subclasses at baseline: Effect of statin treatment.
Significant discontinuities between LDL subclasses were evident when the profiles of lipid/apoB ratios were considered at baseline ( Fig. 2A-V; Suppl Fig. S2-A-AF) Fig.S2-B (p<0.001), whereas this ratio in LDL1 was lower than that in both LDL2 and LDL3, giving a convex profile overall.
When the apoB-normalised profiles for the VLDL+IDL fraction and the LDL1-5 subclasses were considered following statin treatment, then an overall resemblance to the corresponding profiles at baseline was seen for each lipid class, with minor differences. The significant and preferential elevation in the lysolipid/apoB ratio for LPC, LPC(O) and LPI in LDL5 at both baseline and following statin treatment is noteworthy (range of p values <0.05 -0.001) (Suppl fig.S2, K-N, T).

Comparison of lipid class concentrations in the total LDL fraction in healthy, control subjects with those of MetS subjects at baseline: Normalisation by statin treatment
Important qualifications concerning the comparison of lipid class profiles in LDL (d1.019-1.063 g/mL) in control subjects with the corresponding LDL fraction in the MetS group at D0 and D180 are discussed in the main Results text. These profiles are presented in Suppl. Table S4; the salient findings are commented in the Results section.

REFERENCES :
1  Values are expressed either as means ± SEM (n=12) or as ranges (see below). Baseline corresponds to D0 and 180 days of statin treatment to D180.
Mass determinations of lipid and protein components were performed with absorptiometric assays as outlined in the Methods section; the mass of individual components was expressed as % of total mass for each subfraction, with the exception of plasma concentrations and mass ratios.
***p<0.001, **0.001<p<0.01 and *0.01<p<0.05 vs D0. Data from Chapman et al. (2). Values for ranges of plasma concentrations of VLDL+IDL and LDL subclasses in groups of healthy normal control subjects were obtained using the same methodology as that described herein, and are taken from refs (2,6,14).