A METHOD FOR DIAGNOSING METABOLIC DISORDER

Abstract

The present invention relates to a method of diagnosing or prognosing a metabolic disorder in a subject; and in particular to a method comprising determining the quantitative or qualitative level of a biomarker in a biological sample; and diagnosing or prognosing the metabolic disorder based on the quantitative or qualitative level of the biomarker.

Claims

1. A method of diagnosing or prognosing a metabolic disorder in a subject, the method comprising the steps of: (a) determining the quantitative or qualitative level of one or more biomarkers in a biological sample from the subject; and (b) diagnosing or prognosing the metabolic disorder in the subject based on the quantitative or qualitative level of the or each biomarker in the biological sample; wherein the or each biomarker is selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC-III; HV304; APOA-I; APOA-IV; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; CRP; A2GL; CO6; FA10; IGKV1D-33; ITIH1; I3L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.

2. The method of claim 1, wherein the determining step (a) comprises determining the quantitative or qualitative level of all of the biomarkers in the biological sample from the subject.

3. The method of claim 1, wherein the determining step (a) comprises determining the quantitative or qualitative level of each of the biomarkers in the biological sample from the subject.

4. The method of claim 1, wherein the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC-III; HV304; APOA-I; APOA-IV; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; CRP; A2GL; CO6; FA10; IGKV1D-33; ITIH1; I3L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.

5. The method of claim 1, wherein the or each biomarker is a protein encoded by a gene selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC-III; HV304; APOA-I; APOA-IV; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; CRP; A2GL; CO6; FA10; IGKV1D-33; ITIH1; I3L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.

6. The method of claim 5, wherein the or each biomarker is a protein selected from: Complement C4-B; Heparin cofactor 2; Protein IGKV2D-28; Apolipoprotein C-I; Apolipoprotein C-III; Ig heavy chain V-III region 23; Apolipoprotein A-I; Apolipoprotein A-IV; Alpha-2-antiplasmin; Gelsolin; Complement factor I; Protein IGHV4-31; Ceruloplasmin; Serum paraoxonase/arylesterase 1; Apolipoprotein D; Ig lambda-3 chain C regions; Immunoglobulin lambda variable 2-8; Complement C2; Inter-alpha-trypsin inhibitor heavy chain H4; Ig kappa chain C region; Inter-alpha-trypsin inhibitor heavy chain H2; Ig gamma-2 chain C region; Uncharacterized protein (B4E1Z4); Protein IGKV1-17; C-reactive protein; Leucine-rich alpha-2-glycoprotein; Complement component C6; Coagulation factor X; Protein IGKV1-33; Inter-alpha-trypsin inhibitor heavy chain H1; Sex hormone-binding globulin; Ig heavy chain V-I region 5; Serum amyloid P-component; Vitamin K-dependent protein S; Ig gamma-1 chain C region; Phosphatidylinositol-glycan-specific phospholipase D; Plasma protease C1 inhibitor; Ig lambda chain V-I region 51; Actin, cytoplasmic 2; Protein IGHV3-74; Corticosteroid-binding globulin; Protein IGKV2D-29; Vitamin K-dependent protein C; and Ig lambda chain V-II region BUR.

7. The method of claim 1, wherein the determining step (a) comprises determining the quantitative or qualitative level of one or more subsets of one or more biomarkers in the biological sample from the subject.

8. The method of claim 7, wherein the first subset comprises CO4B; HEP2; IGKV2D-28; K7ERI9; APOC3; HV304; APOA1; APOA4; A2AP; and GELS.

9. The method of claim 7, wherein the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC3; HV304; APOA1; APOA4; A2AP; and GELS.

10. The method of claim 7, wherein the or each biomarker is a protein selected from: Complement C4-B; Heparin cofactor 2; Protein IGKV2D-28; Apolipoprotein C-I; Apolipoprotein C-III; Ig heavy chain V-III region 23; Apolipoprotein A-I; Apolipoprotein A-IV; Alpha-2-antiplasmin; and Gelsolin.

11. The method of claim 7, wherein the second subset comprises CO4B; HEP2; IGKV2D-28; K7ERI9; APOC3; HV304; APOA1; APOA4; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; and CRP.

12. The method of claim 11, wherein the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC3; HV304; APOA1; APOA4; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; and CRP.

13. The method of claim 11, wherein the or each biomarker is a protein selected from: Complement C4-B; Heparin cofactor 2; Protein IGKV2D-28; Apolipoprotein C-I; Apolipoprotein C-III; Ig heavy chain V-III region 23; Apolipoprotein A-I; Apolipoprotein A-IV; Alpha-2-antiplasmin; Gelsolin; Complement factor I; Protein IGHV4-31; Ceruloplasmin; Serum paraoxonase/arylesterase 1; Apolipoprotein D; Ig lambda-3 chain C regions; Immunoglobulin lambda variable 2-8; Complement C2; Inter-alpha-trypsin inhibitor heavy chain H4; Ig kappa chain C region; Inter-alpha-trypsin inhibitor heavy chain H2; Ig gamma-2 chain C region; Uncharacterized protein (B4E1Z4); Protein IGKV1-17; and C-reactive protein [Cleaved into: C-reactive protein(1-205)].

14. The method of claim 1, wherein the metabolic syndrome is selected from one or more of abdominal obesity, high blood pressure (hypertension), high blood sugar (hyperglycaemia), high serum triglycerides (hypertriglyceridemia), and low high-density lipoprotein (HDL) levels (HDL-cholesterol (C)).

15. A method of diagnosing or prognosing a metabolic disorder in a subject, wherein the metabolic syndrome is selected from one or more of abdominal obesity, high blood pressure (hypertension), high blood sugar (hyperglycaemia), high serum triglycerides (hypertriglyceridemia), and low high-density lipoprotein (HDL) levels (HDL-cholesterol (C)), the method comprising the steps of: (a) determining the quantitative or qualitative level of one or more biomarkers in a biological sample from the subject; and (b) diagnosing or prognosing the metabolic disorder in the subject based on the quantitative or qualitative level of the or each biomarker in the biological sample; wherein the or each biomarker is selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC-III; HV304; APOA-I; APOA-IV; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; CRP; A2GL; CO6; FA10; IGKV1D-33; ITIH1; I3L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.

16. A method of diagnosing or prognosing a metabolic disorder in a subject, irrespective of the weight of the subject; the method comprising the steps of: (a) determining the quantitative or qualitative level of one or more biomarkers in a biological sample from the subject; and (b) diagnosing or prognosing the metabolic disorder in the subject based on the quantitative or qualitative level of the or each biomarker in the biological sample; wherein the or each biomarker is selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC-III; HV304; APOA-I; APOA-IV; A2AP; GELS; G3XAM2; IGHV4-31; CERU; PON1; APOD; IGLC3; IGLV2-8; CO2; ITIH4; IGKC; Q5T985; IGHG2; B4E1Z4; IGKV1-17; CRP; A2GL; CO6; FA10; IGKV1D-33; ITIH1; I3L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Reference will be made to the accompanying drawings in which:

[0073] FIG. 1 illustrates serum cholesterol efflux capacity (CEC) and PON1 activity within obese and normal weight (NW) cohorts; wherein J774 macrophages, labelled with .sup.3H-cholesterol (1 Ci/ml), were stimulated cAMP (0.3 mM) to drive ABCA1 protein expression; fasting serum samples were depleted of ApoB particles by PEG precipitation; and percentage .sup.3H-cholesterol efflux to 2.8% PEG-supernatant was monitored over 4h; and wherein (A) Total, (B) ABCA1-independent and (C) ABCA1-dependent efflux in NW and obese group is presented (NW n=131, obese n=108); and wherein (D) Serum PON1 activity was assessed enzymatically (NW n=44, obese n=97); and wherein the obese cohort was subdivided into MHO (n=43) and MUO (n=65) and (E) total, (F) ABCA1-independent and (G) ABCA1-dependent efflux presented; and wherein (H) Serum PON1 activity within the MHO (n=35) and MUO (n=58) sub-groups are presented; wherein statistical significance is presented as *p<0.05, **p<0.01, ***p<0.001 w.r.t. NW;

[0074] FIG. 2 illustrates fractionation and assessment of HDL sub-particle function; wherein a sub-group of age and sex-matched NW (n=12) and obese (n=17) subjects were selected and serum separated by FPLC; wherein (A) Cholesterol levels within fractions were determined enzymatically and a lipoprotein profile curve generated; and wherein (B) Cholesterol levels within fractions 32, 34 and 37 presented; and wherein J774 macrophages were labeled with .sup.3H-cholesterol (1 Ci/ml), and stimulated with cAMP (0.3 mM) prior to evaluation of efflux to 30% v/v FPLC fraction over 4h; and wherein Total efflux to (C) HDL-fraction 32 (large), (D) HDL-fraction 34 (medium) and (E) HDL-fraction 37 (small) are presented; and wherein (F) SAA protein levels were determined within fraction 32 and 37 by ELISA; wherein statistical significance is presented as *p<0.05, **p<0.01, ***p<0.001 w.r.t. NW;

[0075] FIG. 3 illustrates functional analysis of the HDL proteome; wherein serum lipoproteins (NW n=12, Obese n=17 (MHO n=7; MUO n=10), n=29 total) were separated by FPLC and protein within fraction 38 was precipitated, digested and proteomics characterised using an Orbitrap Q Exactive mass spectrometer; and wherein raw data was processed using MaxQuant and analysed using Perseus software; wherein gene ontology (GO) analysis was performed on the data from the total cohort (n=29); and wherein (A) Statistically enriched categories (p<0.05) were functionally grouped and visualised using a pie chart; and wherein the most significant terms in each functional group are labelled; and wherein (B) Non parametric correlations were performed between ApoA-I and the HDL proteome across the total cohort, and significant correlations are presented;

[0076] FIG. 4 illustrates visualisation of differentially regulated proteins on HDL-fraction 38 between NW and MUO groups; wherein (A) the differentially regulated proteins (p<0.05) on HDL fraction 38 comparing LFQ intensities between NW (n=12) and MUO (n=10) were visualised using a heat map in Perseus; wherein, for visualisation, Log 2 transformed LFQ intensities are first imputed to replace missing values with values from a normal distribution and subsequently z-score normalized; wherein label free quantitative (LFQ) intensities of (B) CF1, CO2, CO4B and C06, (C) CRP, HEP2, HV304, IGLV2_8 and PHLD, (D) A2AP, CERU, GELS and PON1 and (E) APOA-I, APOA-II, APOA-IV and APOC-III are illustrated (*p<0.05, **p<0.01, ***p<0.001 w.r.t. NW; ##p<0.01 w.r.t. MHO);

[0077] FIG. 5 illustrates visualisation of differentially regulated proteins on HDL between MHO and MUO groups; wherein (A) the differentially regulated proteins (p<0.05) on HDL fraction 38 comparing LFQ intensities between MHO (n=7) and MUO (n=10) were visualised using a heat map in Perseus; wherein, for visualisation, Log 2 transformed LFQ intensities are first imputed to replace missing values with values from a normal distribution and subsequently z-score normalized; wherein label free quantitative (LFQ) intensities of (B) CERU and GELS, (C) IGLV2_8 and IGKC and (D) ANGT, FA-X and KLKB1 were illustrated, (*p<0.05, **p<0.01, w.r.t. NW);

[0078] FIG. 6 illustrates metabolic HDL index scorecorrelation to metabolic health parameters; wherein (A) a Metabolic HDL Index (MHI) score was developed using the 44 significantly different proteins between NW and MUO groups and the composite score across groups is presented, *p<0.05 and ***p<0.001; wherein (B) non parametric correlations were performed between the MHI score and clinical characteristics; and wherein (C) a heat-map was generated in Perseus based on the contributing proteins to the MHI score and were ranked according to decreasing MHI score;

[0079] FIG. 7 illustrates cholesterol efflux capacity normalised to HDL-C and characterisation of gender effects on CEC within NW and obese cohort J774 macrophages, labeled with 3H-cholesterol (1 Ci/ml), were stimulated cAMP (0.3 mM) to drive ABCA1 protein expression; wherein fasting serum samples were depleted of ApoB particles by PEG precipitation; and wherein serum HDL-C levels were determined enzymatically; and wherein percentage .sup.3Hcholesterol efflux to 2.8% PEG-supernatant was monitored over 4h and normalised to HDLC input; wherein (A) Total, ABCA1-independent and ABCA1-dependent efflux capacity after normalisation to HDL-C; (B) Serum HDL-C levels within NW (n=129) and obese (n=110) cohort and male (NW n=129, Obese n=110) and female (NW n=129, Obese n=110) subgroups; and wherein non-normalised efflux in male and female (C) NW and (D) obese cohorts presented; and wherein effects of obesity on efflux in (E) female and (F) male cohorts presented compared to NW; wherein statistical significance is presented as *p<0.05, **p<0.01, ***p<0.001 w.r.t. NW, ##p<0.01, male v female;

[0080] FIG. 8 illustrates functional analysis of the differentially expressed proteins on the HDL proteome; wherein a sub-group of age and sex-matched NW (n=12) and Obese n=17 (MHO n=7; MUO n=10) subjects were selected and serum separated by FPLC (A) 146 proteins were identified; and wherein enriched categories from GO biological processes analysis of the genes were functionally grouped using GO Term Fusion; and wherein the percentage genes associated with the functional pathways were identified using GO analysis and presented in bar graph format (B);

[0081] FIG. 9 illustrates functional analysis of the differentially expressed proteins on the HDL proteome between NW and MUO groups; wherein a sub-group of age and sex-matched NW (n=12) and Obese n=17 (MHO n=7; MUO n=10) subjects were selected and serum separated by FPLC; wherein the functional pathways associated with the differentially expressed proteins on HDL between NW and MUO groups were investigated using GO analysis (A); and wherein enriched categories from GO biological processes analysis of the genes were functionally grouped using GO Term Fusion; wherein the primary pathways identified by GO analysis are presented in bar-graph format (B); and

[0082] FIG. 10 illustrates correlation between MHI score and components of the metabolic Syndrome; wherein non-parametric correlations were applied to assess the relationship between the MHI score and components of the Metabolic Syndrome; BMI (kg/m2) (A), DBP (mmHG) (B), SBP (mmHG) (C), Glucose (mmol/L) (D), HDL (mmol/L) (E) and TAG (mmol/L) (F).

EXAMPLES

[0083] Reference will now be made to the following non-limiting examples:

Materials and Methods

[0084] Materials: Cholesterol [1,2-3H(N)] was purchased from Perkin-Elmer Analytical Sciences (Ireland). Cell culture material was purchased from Lonza (Slough, UK). All other reagents, unless otherwise stated, were from Sigma Aldrich Ltd.

[0085] Study Population: The study subjects (n=108 obese and n=131 normal weight (NW)) were recruited by St. Vincent's University Hospital, University College Dublin and Tallaght Hospital, Dublin, Ireland. Overnight fasted serum samples were used for all analysis. The inclusion criteria for the obese subjects were: age (20 to 70 years), BMI 30 kg/m.sup.2, while the inclusion criteria for the normal weight (NW) control subjects were: age (20 to 70 years old), BMI<30 kg/m.sup.2, and absence of the MetS (MetS). Obese subjects were classified into metabolically healthy obese (MHO, n=43) (2 components of MetS) or metabolically unhealthy obese (MUO, n=65) groups (3 components of MetS) based on the following National Cholesterol Education ProgramAdult Treatment Panel III (NCEP-ATP III) guidelines [3]; (1) waist circumference >102 cm (men) and >88 cm (women) (2) Triglycerides levels 150 mg/dL, (3) HDL-C levels <40 mg/dL (men) and <50 mg/dL (women) (4) fasting glucose levels 100 mg/dL and (5) systolic blood pressure 130 mmHg and diastolic blood pressure 85 mmHg. A sub-group of age and sex-matched NW (n=12) and obese (n=17; n=7 MHO & n=10 MUO) were selected for HDL proteomics analysis. Ethical approval was obtained from University College Dublin, St. Vincent's University Hospital and Tallaght Hospital Human Research Ethics committees.

[0086] Paraoxonase-1 (PON1) Activity Assay: PON1 activity was determined by the conversion rate of phenylacetate to phenol in the presence of serum as described in Osto, E., et al., 2015. 131 (10): p. 871-81.25. Activity of PON-1 is expressed as per mol/min/L.

[0087] Fast Protein Liquid Chromatography (FPLC): Lipoproteins from frozen serum samples (150 L) were separated using size exclusion chromatography on FPLC (Amersham Pharmacia Biotech) using two sequential Superose 6 10/300 columns (GE Healthcare Lifesciences, UK) and phosphate buffer saline containing 1 mM Ethylenediaminetetraacetic acid as the eluent. Cholesterol concentration in each fraction was determined enzymatically using LabAssay Cholesterol (WAKO chemicals, Germany). FPLC fractions were stored at 80 C. prior to proteomics analysis.

[0088] Cholesterol efflux capacity (CEC): J774 macrophages were labeled for 24h with .sup.3H-cholesterol (1 Ci/ml) and equilibrated overnight in Dulbecco's modified eagle medium (DMEM) containing 0.2% bovine serum albumin (BSA)cAMP (0.3 mM) to drive ABCA1 expression. ApoB-containing lipoproteins were removed from human serum by polyethylene glycol (PEG) precipitation as described in Vikari, J., 1976. 36 (3): p. 265-8 or HDL-fractions were isolated by FPLC. Ex vivo efflux from labeled macrophages to 2.8% HDL supernatant or 30% v/v FPLC fraction in minimal essential media (MEM) was measured over 4h. The difference in efflux from cells stimulated in the presence or absence of cAMP represents ABCA1-dependent efflux. ABCA1-independent efflux was derived from untreated (cAMP) cells.

[0089] Proteomics analysis: Lipoproteins from serum samples were separated by FPLC and proteins from HDL-containing fraction 38 were precipitated using trichloroacetic acid (TCA). Protein pellets washed with ice-cold acetone and re-suspended in buffer of 8M Urea in 50 mM Ammonium Bicarbonate (NH.sub.4HCO.sub.3, Sigma Aldrich). Protein concentration was determined using Bradford Protein Assay. Cysteines of plasma protein samples were reduced using dithiothreitol followed by alkylation with iodoacetamide before addition of trypsin (trypsin singles TM proteomic grade, Sigma Aldrich). Digestion was carried out overnight at 37 C. After drying in vacuum centrifuge, peptides were acidified by trifluoroacetic acid (TFA), desalted with c18 STAGE tips as described in Rappsilber, J., 2007. 2 (8): p. 1896-906 and re-suspended in 0.1% TFA. The samples were run on a Thermo Scientific Q Exactive mass spectrometer connected to a Dionex Ultimate 3000 (RSLC nano) chromatography system. Raw data was processed using MaxQuant version 1.5.5.1 incorporating the Andromeda search engine. MS/MS spectra was searched against a human uniprot database using the default settings of MaxQuant. Label free quantitative (LFQ) ion intensities of peptides and proteins were generated by MaxQuant and analysed using Perseus software. Data was log transformed and t-test comparison of fractions carried out. For visualization using heat maps, missing values were imputed with values from a normal distribution and the dataset was normalized by z-score as described by Tyanova, S., et al., 2016. 13 (9): p. 731-40.

[0090] Serum analysis: Serum insulin was measured by ELISA (Crystal Chem Inc, USA). Plasma and PEG-supernatant triacylglycerol (TAG), cholesterol, phospholipid (Wako Chemicals GmbH, Germany), were measured enzymatically as per manufacturers' guidelines.

[0091] Statistical analysis and metabolic HDL index (MHI) scoring: Statistical analysis was performed using GraphPad Prism 5 (GraphPad Sofware Inc, CA) and SPSS software (IBM analytics). Normality tests were conducted using Shapiro-Wilk tests prior to analysis. In the event of normally distributed data, a one-way ANOVA with Bonferroni post-hoc test was applied to data-sets with multiple groups, and an unpaired t-test applied to data-sets of two groups. An adjusted General Linear Model was used to assess confounding. If violation of normal distribution was observed, non-parametric Kruskal-Wallis with a Dunn's post-hoc test was applied to data-sets with multiple groups. Bivariate correlations were performed using Pearson's (normal data) or Spearman's (non-normal data) tests as appropriate. Variables are expressed as meanSEM. To generate a MHI score from the proteomics data-base, z-scores were generated from raw LFQ values and the sum of the z-scores of proteins that increased in MUO was subtracted from the sum of the z-scores that decreased in MUO.

Example 1

Serum ABCA1-Independent Efflux Capacity and PON1 Activity Reduced in Obese Subjects Compared to Normal-Weight (NW) Control

[0092] Clinical characteristics of NW and obese cohorts are highlighted in Table 1 and the medication list is provided within the supplement (Supplement Table 1A&B). The obese group exhibited significant increases in BMI and waist to hip ratio, triglyceride (TAG), fasting glucose, hsCRP, fasting insulin, and HOMA-IR while HDL-C was significantly reduced compared to NW. A significant reduction in total and ABCA1-independent efflux to ApoB-depleted serum was observed in the obese group compared to NW group. No effect on ABCA1-dependent efflux was observed (FIG. 1A-C). Reduced HDL-C was a critical determinant of reduced efflux as evident upon normalisation of results to HDL-C (FIGS. 7A&B). Total and ABCA1-independent efflux negatively correlated with BMI, systolic BP, diastolic BP, TAG and insulin and positively correlated with HDL-C across the total cohort (Table 2A/B). ABCA1-dependent efflux positively correlated with HbA1c (r=0.212) while ABCA1-independent efflux negatively correlated with HbA1c (r=0.214) indicative of a shift from efflux to larger particles towards smaller particles with insulin resistance. A significant correlation between efflux and gender was also observed (Table 2A/B) with NW females exhibiting significantly higher CEC compared to NW malesthis beneficial effect was abrogated within the obese setting (FIGS. 7C&D). Furthermore the reduction in efflux within the obese cohort compared to lean cohort was more pronounced in females than in males (FIGS. 7E&F). These findings suggest that the protective effect of female gender is negated in the obese setting.

TABLE-US-00002 TABLE 1 Participant Characteristics Lean Obese n = 131 n = 108 Mean SEM Mean SEM P* Age (years) 40.23 0.82 45.10 1.14 <0.001 Gender (M, F %) 49, 51 42, 58 0.267 BMI (kg/m.sup.2) 25.05 0.24 45.77 0.97 <0.001 Waist to Hip Ratio 0.85 0.01 0.96 0.02 0.010 Systolic Blood Pressure (mmHG) 121.68 2.07 131.53 1.25 0.113 Diastolic Blood Pressure (mmHG) 74.70 1.34 82.40 0.87 0.053 Total Cholesterol (mmol/L) 4.67 0.08 4.73 0.09 0.241 HDL (mmol/L) 1.45 0.03 1.19 0.03 0.003 LDL (mmol/L) 2.65 0.12 2.69 0.10 0.538 TAG (mmol/L) 0.99 0.04 1.58 0.08 0.001 HbA1c (mmol/mol) 32.64 1.22 41.43 1.12 0.253 Glucose (mmol/L) 5.17 0.06 5.91 0.25 <0.001 Insulin (mIU/L) 6.50 0.56 24.03 v 2.10 <0.001 HOMA-IR 1.54 0.15 6.45 0.83 <0.001 High sensitivity CRP (mg/L) 1.63 0.31 7.90 0.76 <0.001 *Differences assessed by Independent Samples T-Test Differences assessed by a X.sup.2 Test n = 68 (NW), 53 (Obese) n = 117 (NW), 98 (Obese)

TABLE-US-00003 TABLE 2A Correlations between Efflux Capacity and Clinical Characteristics Total Efflux Total Cohort Lean Cohort Obese Cohort n = 239 n = 129 n = 110 R P R P R P Age (years) 0.022 0.731 0.188 0.031 0.082 0.402 Gender (M %, F %) 0.227 <0.001 0.241 0.006 0.258 0.007 BMI (kg/m.sup.2) 0.157 0.015 0.111 0.207 0.034 0.730 SBP (mmHG) 0.204 0.002 0.221 0.012 0.101 0.298 DBP (mmHG) 0.198 0.002 0.190 0.031 0.097 0.317 TC (mmol/L) 0.032 0.626 0.046 0.604 0.059 0.548 HDL (mmol/L) 0.286 <0.001 0.202 0.020 0.327 0.001 LDL (mmol/L) 0.105 0.208 0.183 0.190 0.005 0.961 TAG (mmol/L) 0.147 0.023 0.069 0.433 0.066 0.501 Glucose (mmol/L) 0.030 0.658 0.086 0.360 0.135 0.166 Insulin (uIU/ml) 0.235 0.009 0.009 0.942 0.042 0.769 HOMA-IR 0.165 0.069 0.036 0.767 0.052 0.710 HbA1c (mmol/mol) 0.018 0.834 0.002 0.987 0.046 0.678 hsCRP (mg/L) 0.127 0.064 0.184 0.045 0.012 0.908 Spearman Correlations

TABLE-US-00004 TABLE 2B Correlations between Efflux Capacity and Clinical Characteristics ABCA1-Independent Efflux Total Cohort Lean Cohort Obese Cohort n = 239 n = 131 n = 108 R P R P R P Age (years) 0.070 0.280 0.194 0.026 0.072 0.456 Gender 0.243 <0.001 0.339 <0.001 0.142 0.143 (M %, F %) BMI (kg/m.sup.2) 0.175 0.007 0.069 0.437 0.035 0.718 SBP (mmHG) 0.190 0.003 0.231 0.008 0.021 0.827 DBP (mmHG) 0.173 0.007 0.143 0.107 0.059 0.546 TC (mmol/L) 0.097 0.137 0.136 0.122 0.006 0.952 HDL (mmol/L) 0.308 <0.001 0.276 0.001 0.245 0.012 LDL (mmol/L) 0.208 0.012 0.429 0.001 0.067 0.525 TAG (mmol/L) 0.172 0.008 0.077 0.385 0.122 0.212 Glucose 0.075 0.266 0.177 0.059 0.092 0.346 (mmol/L) Insulin (uIU/ml) 0.179 0.050 0.125 0.303 0.225 0.113 HOMA-IR 0.143 0.114 0.086 0.478 0.141 0.314 HbA1c 0.214 0.012 0.338 0.013 0.031 0.781 (mmol/mol) hsCRP (mg/L) 0.141 0.039 0.135 0.143 0.075 0.468 Spearman Correlations

TABLE-US-00005 Supplement TABLE 1A Medication use in the obese cohort, split by MHO and MUO MHO MUO n = 43 n = 65 P* Medication Use 79.9 84.3 0.268 Medication Type Insulin-sensitising 15.4 35.3 0.029 Weight loss 25.6 25.5 0.588 Liraglitide 12.8 29.4 0.051 Statins 10.3 17.6 0.249 Hypertension 23.1 39.2 0.081 *Differences between groups assessed using X2 Test

TABLE-US-00006 Supplement TABLE 1B Medication Types Insulin-sensitising Weight Loss Liraglitide Statins Anti-hypertensives Diaglyc Orlistat Victoza Atorvastatin Accupo Diamicron Reductil Lipitor ACE-1 DPP4 Subitramine Pravastatin Amlode Glicazide Xenical Simvastatin Amlodipine Glucophage Atenelol Junamet Bendroflumethiazide Metformin Benetor Novarapid Bisoprolol Trajenta Biopress Cardura XL Centyl K Cosartal Coversil Cozaar Doxatosin Frusemide Half-beta progone Istin Konverge Lercanidipine Lisinopril Losartan Metaprolol Metocor Micardis plus Micordis Modiuretic Nebivolol Omessar Plus Perindopril Ramic Rampiril Telmisartan Zestril

[0093] Serum PON1 activity was significantly reduced in the obese group relative to NW (FIG. 1D). The obese group was sub-divided into MHO and MUO sub-groups and clinical characteristics outlined in Table 3. A significant difference in HDL-C, TAG, fasting glucose and HbA1c was observed between MHO and MUO. Both groups exhibited reduced ABCA1-independent efflux and PON-1 activity compared to NW with no significant difference evident between MHO and MUO (FIG. 1E-H).

TABLE-US-00007 TABLE 3 Participant characteristics, split by MHO and MUO NW MHO MUO n = 131 n = 43 n = 65 Mean SEM Mean SEM Mean SEM P* P Age (years) 40.05 0.83.sup.a 43.49 1.73.sup.ab 46.42 1.48.sup.b <0.001 Gender (M, F %) 50, 50 36, 64 45, 55 0.260 BMI (kg/m.sup.2) 25.06 0.24.sup.a 44.34 1.72.sup.b 46.11 1.18.sup.b <0.001 <0.001 Waist to Hip Ratio 0.85 0.01.sup.a 0.92 0.02.sup.b 0.98 0.03.sup.b <0.001 <0.001 SBP (mmHG) 121.61 2.09.sup.a 126.91 2.21.sup.ab 134.55 1.41.sup.b <0.001 0.002 DBP (mmHG) 74.67 1.35.sup.a 81.02 1.37.sup.b 83.18 1.15.sup.b <0.001 <0.001 Total Cholesterol (mmol/L) 4.67 0.08 .sup.4.70 0.11 4.74 0.12.sup. 0.902 0.978 HDL (mmol/L) 1.44 0.03.sup.a 1.33 0.05.sup.a 1.12 0.03.sup.b <0.001 <0.001 LDL (mmol/L) 2.67 0.12 .sup.2.68 0.14 2.67 0.15.sup. 0.997 0.927 TAG (mmol/L) 0.99 0.04.sup.a 1.14 0.04.sup.a 1.87 0.12.sup.b <0.001 <0.001 HbA1c (mmol/mol) 34.79 3.41.sup.a 38.25 1.23.sup.b 43.35 1.71.sup.c <0.001 <0.001 Glucose (mmol/L) 5.18 0.06.sup.a 4.99 0.24.sup.a 6.52 0.37.sup.b <0.001 <0.001 Insulin (uIU/ml) 6.27 0.52.sup.a 22.54 3.08.sup.b 24.34 2.75.sup.b <0.001 <0.001 HOMA-IR 1.49 0.14.sup.a 4.96 1.00.sup.b 7.11 1.11.sup.b <0.001 <0.001 High sensitivity CRP (mg/L) 1.65 0.31.sup.a 6.96 1.12.sup.b 8.29 1.11.sup.b <0.001 <0.001 *Differences across groups were assessed using a one-way ANOVA with Bonferroni Correction Differences across groups were assessed using a General Linear Model adjusted for Age .sup.a,b,cMean values with unlike superscript letters are significantly different between groups (P < 0.05) Differences across groups were assessed using a X.sup.2 Test n = 68 (NW), 20 (MHO), 33 (MUO) n = 117 (NW), 40 (MHO), 58 (MUO)

Example 2

[0094] Lipoprotein separation and sub-particle functionality A sub-cohort of age and sex-matched individuals were selected from NW (n=12) and obese (n=17) groups (Table 4A) and serum lipoproteins were separated by FPLC (FIG. 2A). Cholesterol on LDL fraction 21 was significantly increased in the obese group compared to NW, with a trend towards reduced cholesterol on larger HDL fractions (fractions 30-36), with no difference in cholesterol levels on smaller HDL fractions (FIG. 2A). We subsequently assessed the efflux capacity of FPLC fractions containing large (fraction 32), medium (fraction 34) and small (fraction 37) HDL particles. Cholesterol concentration was reduced on fractions 32&34, with no difference in fraction 37 (FIG. 2B), in obese group compared to NW. The efflux function of larger HDL particles was significantly reduced (FIG. 2C), with no difference in efflux to medium and small HDL fractions (FIGS. 2D&E). SAA levels were significantly increased on obese-HDL fraction 37 as detected by ELISA compared to NW (FIG. 2F).

TABLE-US-00008 TABLE 4A Characteristics of age and sex matched sub-cohort NW Obese n = 12 n = 17 P* Age (years) 41.25 2.53 41.76 2.16 0.879 Gender (M, F %) 50, 50 53, 47 0.587 BMI (kg/m.sup.2) 23.64 0.68 47.60 3.05 <0.001 SBP (mmHG) 124.25 3.78 132.59 2.50 0.065 DBP (mmHG) 75.42 3.40 84.18 1.98 0.028 TAG (mmol/L) 1.02 0.09 1.72 0.24 0.013 Cholesterol (mmol/L) 4.43 0.19 4.75 0.26 0.366 HDL (mmol/L) 1.50 0.10 1.11 0.06 0.001 LDL (mmol/L) 2.46 0.24 2.87 0.20 0.189 HbA1c (mmol/mol) 29.34 1.69 44.70 5.21 0.023 Glucose (mmol/L) 4.70 0.09 6.12 0.51 0.028 Insulin (mmol/L) 7.38 3.18 29.33 4.15 0.001 HOMA-IR 1.43 1.41 8.23 1.50 0.001 High sensitivity CRP (mg/L) 0.65 0.10 9.00 2.23 0.003 n, number of participants. *Differences across groups were assessed using an Independent Samples T-Test n = 6 (NW), n = 15 (Obese), n = 9 (NW), n = 12 (Obese)

Example 3

HDL Proteomic Profile Modulated in Obesity

[0095] The obese group was sub-divided into MHO (n=7) and MUO (n=10) sub-groups (clinical parameters outlined in Table 4B) and proteomics was carried on HDL fraction 38 to determine whether proteins pertaining to other HDL functions, beyond efflux capacity, were altered dependent upon metabolic health status. HDL proteomics was performed on FPLC fraction 38 where no significant difference in HDL-C was evident across groups (FIG. 8A). 146 proteins were identified on the HDL particles and gene ontology (GO) analysis was performed within Cytoscape across the combined cohort. The primary pathways identified included high density lipoprotein particle remodelling, acute inflammatory response, protein activation cascade and reverse cholesterol transport (FIG. 3A and FIG. 8B). Given ApoA-I is the primary protein on HDL we correlated levels of ApoA-I with the proteomic data-set across the combined cohort (FIG. 3B). ApoA-I positively correlated with ApoC-III, Alpha-2-antitrypsin, clusterin, ApoC-II, ApoC-I and ApoA-II and negatively correlated with complement factor 4B (CO4B), complement factor 1 (CF1) inter-alpha-trypsin inhibitor and C-reactive protein.

TABLE-US-00009 TABLE 4B Characteristics of age and sex matched sub-cohort, split by MHO and MUO NW MHO MUO n = 12 n = 7 n = 10 P* Age (years) 41.25 2.53.sup. 41.86 3.55.sup. 41.70 2.88.sup. 0.937 Gender (M, F %) 50, 50 57, 43 50, 50 0.947 BMI (kg/m.sup.2) 23.64 0.68.sup.a 50.60 6.00.sup.b 45.40 3.22.sup.b <0.001 SBP (mmHG) 124.25 3.78.sup.a 125.43 2.79.sup.ab 137.60 2.92.sup.b 0.017 DBP (mmHG) 75.42 3.40.sup.a 80.86 2.02.sup.ab 86.50 2.91.sup.b 0.043 TAG (mmol/L) 1.02 0.09.sup.a 1.01 0.16.sup.a 2.21 0.31.sup.b <0.001 Cholesterol (mmol/L) 4.43 0.19 4.32 0.12.sup. 5.050 0.42.sup. 0.187 HDL (mmol/L) 1.50 0.10.sup.a 1.24 0.10.sup.ab 1.03 0.06.sup.b 0.002 LDL (mmol/L) 2.46 0.24 2.68 0.19.sup. 3.01 0.31.sup. 0.315 HbA1c (mmol/mol) 29.34 1.69.sup.a 38.41 2.09.sup.ab 49.10 8.64.sup.b 0.036 Glucose (mmol/L) 4.70 0.09.sup.a 5.11 0.18.sup.ab 6.83 0.79.sup.b 0.008 Insulin (mmol/L) 7.38 3.18.sup.a 28.64 5.54.sup.ab 28.90 6.24.sup.b 0.026 HOMA-IR 1.43 1.41.sup.a 6.49 3.27.sup.ab 9.46 7.19.sup.b 0.006 High sensitivity CRP (mg/L) 0.65 0.10.sup.a 8.60 4.12.sup.ab 9.29 2.74.sup.b 0.026 n, number of participants *Differences across groups were assessed using a one-way ANOVA with Bonferroni Correction .sup.a,bMean values with unlike superscript letters are significantly different between groups (P < 0.05) n = 6 (NW), n = 7 (MHO), n = 8 (MHO), n = 9 (NW), n = 5 (MHO), n = 7 (MHO)

[0096] Levels of 49 proteins were significantly different between MUO-HDL and NW-HDL (FIG. 4A) with enrichment of complement factors (CF1, CO2, CO4B, CO6), acute phase response proteins (CRP, Serum amyloid P component (SAMP)) and coagulation factors (coagulation factor X and heparin co-factor 2 (Hep2)) and inter-alpha-trypsin inhibitors H1 (ITIH1), ITIH2 and ITIH4 on MUO-HDL relative to NW-HDL. By contrast significantly reduced levels of a range of immunoglobulins, apolipoproteins, PON1, gelsolin, alpha-2-antitrypsin and ceruloplasmin were observed on MUO-HDL relative to NW-HDL (FIG. 4B-E). GO analysis of these differentially expressed proteins identified four major hubs; reverse cholesterol transport, regulation of protein processing (complement activation, acute phase response and negative regulation of blood coagulation), protein activation cascade and phagocytosis (FIGS. 9A&B).

Example 4

Differences in Proteomic Composition Between MHO and MUO Groups

[0097] The effect of metabolic health on HDL proteomic composition was subsequently assessed. A smaller number of proteins were identified as being significantly different between MHO and MUO groups (n=14) (FIG. 5A). Coagulation factor X, angiotensinogen and kallikrein were enriched on MUO-HDL compared to MHO-HDL, while gelsolin, ceruloplasmin and a range of immunoglobulins were enriched on MHO-HDL compared to MUO-HDL (FIG. 5B-D).

Example 5

Metabolic HDL Index (MHI) Score and Correlation to the MetS

[0098] The HDL proteomic signature of NW and MUO groups could stratify individuals into their respective groups with 92% and 90% accuracy respectively. The MHO group by contrast exhibited greater variability in their HDL proteome with n=2 clustered with NW, n=3 clustered with MUO and n=2 falling into their own grouping. A scoring algorithm was generated based on significantly different proteins between NW and MUO groups. MHI decreased incrementally in MHO and MUO groups compared to NW (FIG. 6A). Regression analysis was performed on the data-set generating the MHI and the list of contributing proteins is ranked in order of significance (Table 5). The top ten most significantly correlated proteins to MHI included CO6, CO4B, Hep2, protein IGKV2D-28, ApoC-I, ApoC-III, Ig heavy chain V-III region 23, ApoA-I, ApoA-IV, alpha-2-antiplasmin and gelsolin. The MHI score significantly correlated with BMI, glucose, systolic and diastolic blood pressure, HDL-C, fasting insulin and HOMA-IR with a trend towards negative correlation to TAG (p=0.056) (FIG. 6B and FIG. 10). A heat-map was generated based on the ranked MHI scores and clustered one MHO subject (BMI=30, MHI=17.7) into the NW group and another two MHO subjects (BMI=52, MHI=25.4 and BMI=57, MHI=22.8) into the MUO group (FIG. 6C). Of interest one NW subject exhibited a MHI score (7.5) more akin to the MHO group. These preliminary findings suggest that MHI could provide a novel continuous index upon which to measure metabolic health.

TABLE-US-00010 TABLE 5 Linear Regression of MHI Score to 44 Proteins Gene R R.sup.2 P Complement C4-B CO4B 0.797 0.635 <0.001 Heparin cofactor 2 HEP2 0.784 0.615 <0.001 Protein IGKV2D-28 IGKV2D-28 0.740 0.548 <0.001 Apolipoprotein C-I K7ERI9 0.735 0.540 <0.001 Apolipoprotein C-III APOC-III 0.797 0.635 <0.001 Ig heavy chain V-III region 23 HV304 0.733 0.537 <0.001 Apolipoprotein A-I APOA-I 0.777 0.604 <0.001 Apolipoprotein A-IV APOA-IV 0.690 0.476 <0.001 Alpha-2-antiplasmin A2AP 0.694 0.482 <0.001 Gelsolin GELS 0.682 0.465 <0.001 Complement factor I G3XAM2 0.636 0.404 <0.001 Protein IGHV4-31 IGHV4-31 0.649 0.421 <0.001 Ceruloplasmin CERU 0.633 0.401 <0.001 Serum paraoxonase/arylesterase 1 PON1 0.610 0.372 <0.001 Apolipoprotein D APOD 0.669 0.448 <0.001 Ig lambda-3 chain C regions IGLC3 0.576 0.332 0.001 Immunoglobulin lambda variable 2-8 IGLV2-8 0.574 0.329 0.001 Complement C2 CO2 0.572 0.327 0.001 Inter-alpha-trypsin inhibitor heavy chain H4 ITIH4 0.585 0.342 0.001 Ig kappa chain C region IGKC 0.582 0.339 0.001 Inter-alpha-trypsin inhibitor heavy chain H2 Q5T985 0.543 0.295 0.002 Ig gamma-2 chain C region IGHG2 0.522 0.272 0.002 Uncharacterized protein B4E1Z4 0.530 0.281 0.003 Protein IGKV1-17 IGKV1-17 0.540 0.292 0.004 C-reactive protein CRP 0.518 0.268 0.004 Leucine-rich alpha-2-glycoprotein A2GL 0.505 0.255 0.005 Complement component C6 CO6 0.516 0.266 0.005 Coagulation factor X FA10 0.495 0.245 0.006 Protein IGKV1-33 IGKV1D-33 0.500 0.250 0.006 Inter-alpha-trypsin inhibitor heavy chain H1 ITIH1 0.495 0.245 0.006 Sex hormone-binding globulin I3L145 0.478 0.228 0.009 Ig heavy chain V-I region 5 KV110 0.503 0.253 0.009 Serum amyloid P-component SAMP 0.465 0.216 0.011 Vitamin K-dependent protein S PROS 0.508 0.258 0.013 Ig gamma-1 chain C region IGHG1 0.455 0.207 0.013 Phosphatidylinositol-glycan-specific phospholipase D PHLD 0.463 0.214 0.015 Plasma protease C1 inhibitor IC1 0.415 0.172 0.025 Ig lambda chain V-I region 51 LV104 0.421 0.177 0.032 Actin, cytoplasmic 2 ACTG 0.415 0.172 0.032 Protein IGHV3-74 IGHV3-74 0.366 0.134 0.051 Corticosteroid-binding globulin CBG 0.356 0.127 0.058 Protein IGKV2D-29 IGKV2D-29 0.373 0.139 0.067 Vitamin K-dependent protein C PROC 0.317 0.100 0.094 Ig lambda chain V-II region BUR LV205 0.286 0.082 0.140

[0099] HDL particle functionality has emerged as a novel target and more important determinant of cardiovascular risk than static HDL-C levels. Pathway analysis of the HDL proteome has identified HDL particle remodelling, acute inflammatory response, protein activation cascades, and reverse cholesterol transport as the major pathways associated with the particles which mirror the assigned cardio-protective functions of HDL. The alignment of HDL protein pathways with particle functions suggests that protein composition of HDL is not only specific, but is fundamental, to biological effects. The present invention has identified important changes in the network of proteins associating with HDL in obese subjects compared to NW controls (49 out of 146 proteins significantly changed) with enrichment of pro-inflammatory acute phase proteins and loss of anti-oxidant/anti-inflammatory proteins on obese particles. These findings indicate that HDL particles become metabolically activated during obesity with decreased cardio-protective potential. Further to this, the present invention demonstrates reduced PON-1 activity and reduced ABCA1-independent efflux capacity of serum in obese subjects compared to NW controls. Increasing the quality, as opposed to the quantity, of HDL particles in turn might be more beneficial in the setting of obesity.

[0100] While obesity increases the risk of CVD, this risk is enhanced with concurrent presentation of the MetS. The present invention therefore explores whether metabolic health status is an important pre-requisite for the preservation of healthy HDL particles in the obese state. Chronic inflammation is a classic hallmark of obesity that contributes to development of insulin resistance and likely is the causal link for enhanced cardiovascular risk. Without being bound by theory, the inventors therefore speculated that the sub-acute chronic inflammation observed in MUO may exaggerate HDL dysfunction and proteomic composition.

[0101] The inventors have evaluated total, ABCA1-independent and ABCA1-dependent efflux capacity of serum to delineate the ability of total, large and small HDL particles to support efflux respectively in NW, MHO and MUO groups; and demonstrate reduced total efflux capacity of serum from obese individuals compared to NW control, which was attributable to a specific reduction in ABCA1-independent efflux and not ABCA1-dependent efflux. FPLC analysis demonstrated reduced cholesterol within larger HDL fractions from obese individuals compared to NW, indicative that the number of larger HDL particles, the main acceptors via ABCA1-independent pathways, is reduced. Indeed, normalisation of results to HDL-C input demonstrates that HDL-C is an important determinant of reduced ABCA1-independent efflux in obese subjects. No significant difference in HDL efflux capacity was evident between MHO and MUO sub-groups.

[0102] PON1 is an important anti-oxidant protein that is primarily carried on HDL in serum and reduced levels are associated with increased CVD risk. The inventors hence measured serum PON1 activity as a surrogate for measuring the anti-oxidant capacity of HDL particles. Serum PON1 activity was significantly reduced in the obese cohort compared to NWagain no significant difference was observed between MHO and MUO sub-groups. Lack of difference in HDL functionality between MHO and MUO groups suggests that the obese phenotype alone is sufficient to drive HDL dysfunction or that stratification of obese individuals based on presence or absence of the MetS is not sensitive enough to distinguish between cohorts.

[0103] The efflux capacity of isolated HDL-fractions was evaluated ex vivo and demonstrated reduced efflux to the larger HDL fraction in the obese group compared to NW, with no difference in efflux to small/medium fractions, which was consistent with findings in ApoB-depleted serum. Levels of pro-inflammatory SAA1 were enriched on smaller obese-HDL particles compared to NW-HDL indicative of a pro-inflammatory particle, an effect that is also evident in patients with type 1 diabetes. FPLC analysis demonstrated reduced cholesterol on larger HDL fractions (fractions 30-36), with preservation of cholesterol on smaller HDL fractions (fractions 36-40), and increased LDL-C levels in the obese cohort compared to NW. Proteomic profiling of HDL particles was performed on HDL fraction 38, where HDL-C was equivalent between groups to avoid introduction of a potential systematic error. Furthermore, the obese group was sub-divided into MHO and MUO groupings to establish whether HDL proteomics was more sensitive to detect differences between these groups than HDL function assays.

[0104] A remarkable difference in the HDL proteome was evident between age- and sex-matched NW and MUO groups. Indeed, blinded analysis of the proteomics data could accurately separate the NW (91.7%, 11/12) and MUO (90%, 9/10) groups. MUO HDL particles were enriched for complement factor I, complement C4B, C2 and C6, C-reactive protein, serum amyloid P-component, heparin cofactor 2 (Hep2) and coagulation factor X. By contrast, Apo-AI, AIV, CI, CIII and D, paraoxonase, ceruloplasmin, sex hormone binding globulin (SHBG), cortiocosteroid binding globulin, and alpha-2-antiplasmin were all significantly reduced on MUO-HDL compared to NW-HDL. Previous plasma proteomics investigating the effects of weight-loss in obese individuals on the plasma proteome and demonstrated a specific reversal in some of the parameters observed in our study with down regulation of CRP and Hep2 and upregulation of SHBG. The inventors have observed a significant reduction in ApoC-III on MUO-HDL particles that strongly correlated with ApoA-I (r=0.89), despite plasma levels of ApoC-III usually being elevated with the MetS and CVD. ApoC-III prevents efficient catabolism of triglyceride-rich lipoprotein particles and is hence associated with hypertriglyceridemia. Sequestering of ApoC-III from ApoB/chylomicrons onto HDL improves triglyceride clearance in circulation and hence the re-direction of ApoC-III from HDL onto other triglyceride-rich lipoproteins within the obese cohort could partially mediate hypertriglyceridemia. Enrichment of MUO particles with pro-inflammatory proteins and loss of anti-inflammatory/anti-oxidant proteins indicates presence of a metabolically activated HDL particle within the MUO setting.

[0105] HDL proteomic analysis within the MHO sub-group revealed a greater diversity in expression with prediction tools placing 2/7 individuals in the NW group, 3/7 within the MUO group and 2/7 within their own grouping. These results are consistent with the growing evidence that many MHO individuals eventually progress into the MUO category and indeed HDL proteomics was able to identify a number of individuals who are likely at higher risk due to their metabolic HDL profile. A number of significantly different proteins (n=14) were noted between MHO and MUO HDL proteomes; coagulation FX, kallikrein and angiotensinogen were upregulated on MUO-HDL compared to MHO-HDL and are involved in the coagulation cascade, fibrinolysis and control of blood pressure. By contrast gelsolin, ceruloplasmin and a range of immunoglobulins were increased on MHO-HDL compared to MUO-HDL.

[0106] Given the accuracy of proteomic data to predict grouping of NW and MUO individuals, the present invention relates to a scoring algorithm to generate a metabolic HDL index (MHI). The MHI score positively correlated with HDL-C and ABCA1-independent efflux and negatively correlated with hypertension, hyperglycaemia, BMI, fasting insulin and HOMA-IR. Interestingly when the total cohort was unclustered and re-grouped based on MHI, we identified one MHO individual (BMI=30.2) who exhibited a MHI score akin to the NW group, while two MHO individuals (BMI=52 and 57) exhibited a MHI score that would re-classify them as MUO. We also identified one NW subject with a MHI score that aligned with the MHO group. These preliminary results suggest that HDL proteomic analysis could provide more sensitive stratification of high-risk lean and obese individuals than currently used guidelines but this remains to be validated.

[0107] The present invention has established a metabolic HDL index score based on HDL proteomic composition that correlates with metabolic health status and may provide a useful tool for more accurate stratification of high-risk individuals and subsequent assignment to more aggressive interventions as warranted.