Patent Application
20210018516
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.
Obesity is a global public health epidemic and an independent risk factor for the development of cardiovascular disease (CVD). The mechanisms linking these disease states however are not fully understood. Presence of other co-morbidities including insulin resistance (IR), hypertension and type 2 diabetes mellitus further increases CVD risk in obese individuals. Obesity often co-exists with the metabolic syndrome (MetS) which is diagnosed as the clustering of three or more of the following; abdominal obesity, hyperglycaemia, hypertension, hypertriglyceridemia, and reduced HDL-cholesterol (C)) according to the Adult Treatment Program III criteria. Approximately 10-40% of obese individuals present with no cardiometabolic abnormalities and are classified as metabolically healthy but obese (MHO) while those with more metabolic complications are classified as metabolically unhealthy obese (MUO).
Metabolic dyslipidemia is a classic hallmark of obesity with increased levels of small, dense low-density lipoprotein (LDL) particles and circulating triacylglyerides (TAG) and reduced levels of HDL-C. While HDL-C is inversely associated with CVD, raising HDL-C levels using cholesterol ester transfer protein (CETP) inhibitors failed to have clinical benefit in high risk patients. Furthermore genetically-defined low HDL-C has not been associated with CVD. These findings have questioned the role of HDL particles during CVD pathogenesis and have highlighted the need for a greater understanding of HDL particle biology, particularly in the disease setting.
Reliance on static HDL-C measurements to reflect HDL particle biology is particularly over-simplified. HDL particles are heterogeneous complex particles which differ according to size, density, charge, lipid and protein composition and in turn function, including anti-oxidant, anti-inflammatory, anti-coagulant, anti-thrombotic and cholesterol-efflux promoting functions. Small HDL particles mediate cholesterol efflux via the ATP-binding cassette (ABC), sub-family A, member 1 (ABCA1) transporter (ABCA1-dependent efflux), while larger particles mediate efflux via ABC sub-family G member 1 (ABCG1) and scavenger receptor class B member 1 (SR-BI) transporters (ABCA1-independent efflux). Measurement of HDL efflux capacity in turn is a better predictor of CVD than static HDL-C and similarly ABCA1-dependent efflux, but not HDL-C, inversely correlates with pulse wave velocity in humans.
There is a need for an improved method of diagnosing or prognosing a metabolic disorder in a subject, and which does not rely on static HDL-C measurements as a reflection of HDL particle biology.
According to a first aspect of the present invention, there is provided a method of diagnosing or prognosing a metabolic disorder in a subject, the method comprising the steps of:
Optionally, the determining step (a) comprises determining the quantitative or qualitative level of two or more biomarkers in the biological sample from the subject.
Further optionally, the determining step (a) comprises determining the quantitative or qualitative level of three, four, five, ten, twenty, twenty five, thirty, thirty five, forty, or forty four biomarkers in the biological sample from the subject.
Optionally, the determining step (a) comprises determining the quantitative or qualitative level of all of the biomarkers in the biological sample from the subject.
Optionally or additionally, the determining step (a) comprises determining the quantitative or qualitative level of each of the biomarkers in the biological sample from the subject.
Optionally, the or each biomarker is a gene. Further optionally, the or each biomarker is a nucleic acid. Still further optionally, the or each biomarker is a deoxyribonucleic acid.
Optionally, the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; 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; 13L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.
Optionally, the or each biomarker is a translation product of a gene.
Optionally, the or each biomarker is a translation product of a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; 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; 13L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.
Optionally, the or each biomarker is a protein. Further optionally, the or each biomarker is a peptide. Still further optionally, the or each biomarker is a polypeptide.
Optionally, the or each biomarker is a protein encoded by a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; 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; 13L145; KV110; SAMP; PROS; IGHG1; PHLD; IC1; LV104; ACTG; IGHV3-74; CBG; IGKV2D-29; PROC; and LV205.
Optionally, 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.
Optionally, 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.
Optionally, the determining step (a) comprises determining the quantitative or qualitative level of two or more subsets of one or more biomarkers in the biological sample from the subject.
Optionally, the determining step (a) comprises determining the quantitative or qualitative level of one or more of a first or second subset of one or more biomarkers in the biological sample from the subject.
Optionally, the first subset comprises one or more biomarkers selected from: CO4B; HEP2; IGKV2D-28; K7ERI9; APOC3; HV304; APOA1; APOA4; A2AP; and GELS.
Optionally, the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; APOC3; HV304; APOA1; APOA4; A2AP; and GELS.
Optionally, the or each biomarker is a translation product of a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; APOC3; HV304; APOA1; APOA4; A2AP; and GELS.
Optionally, the or each biomarker is a protein encoded by a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; APOC3; HV304; APOA1; APOA4; A2AP; and GELS.
Optionally, 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.
Optionally, the second subset comprises CO4B; HEP2; IGKV2D-28; K7ER19; 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.
Optionally, the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19;
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.
Optionally, the or each biomarker is a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; 30 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.
Optionally, the or each biomarker is a translation product of a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; 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.
Optionally, the or each biomarker is a protein encoded by a gene selected from: CO4B; HEP2; IGKV2D-28; K7ER19; 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.
Optionally, 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)].
Optionally, the diagnosing or prognosing step (b) comprises comparing the quantitative or qualitative level of the or each biomarker in the biological sample from the subject with the quantitative or qualitative level of the or each respective biomarker in a normal sample.
Optionally, the normal sample is a biological sample from a subject not suffering from a metabolic disorder.
Optionally, a quantitative or qualitative level of the or each biomarker in the biological sample from the subject greater than the quantitative or qualitative level of the or each respective biomarker in a normal sample is indicative of the quantitative or qualitative level of the metabolic disorder.
Optionally, a quantitative or qualitative level of the or each biomarker in the biological sample from the subject greater than the quantitative or qualitative level of the or each respective biomarker in a normal sample is indicative of the quantitative or qualitative presence of the metabolic disorder.
Optionally, the determining step (a) comprises determining the quantitative or qualitative level of all of the biomarkers in one or more of the first or second subsets.
Optionally, the determining step (a) comprises determining the quantitative or qualitative level of each of the biomarkers in one or more of the first or second subsets.
Optionally, the diagnosing or prognosing step (b) comprises comparing the quantitative or qualitative level of the or each biomarker in the or each subset in the biological sample from the subject with the quantitative or qualitative level of the or each respective biomarker in a normal sample.
Optionally, the normal sample is a biological sample from a subject not suffering from a metabolic disorder.
Optionally, a quantitative or qualitative level of the or each biomarker in the or each subset in the biological sample from the subject greater than the quantitative or qualitative level of the or each respective biomarker in a normal sample is indicative of the quantitative or qualitative level of the metabolic disorder.
Optionally, a quantitative or qualitative level of the or each biomarker in the or each subset in the biological sample from the subject greater than the quantitative or qualitative level of the or each respective biomarker in a normal sample is indicative of the quantitative or qualitative presence of the metabolic disorder.
Optionally, the biological sample is selected from whole blood, serum, plasma, urine, interstitial fluid, peritoneal fluid, cervical swab, tears, saliva, buccal swab, skin, brain tissue, and cerebrospinal fluid.
Optionally, the biological sample comprises liporotein. Further optionally, the biological sample comprises high-density lipoprotein (HDL). Still further optionally, the biological sample comprises high-density lipoprotein (HDL) particles.
Optionally, the metabolic disorder is selected from one or more of acid-base imbalance; metabolic brain diseases; calcium metabolism disorders; DNA repair-deficiency disorders; glucose metabolism disorders; hyperlactatemia; iron metabolism disorders; lipid metabolism disorders; malabsorption syndromes; metabolic syndrome X; inborn error of metabolism; mitochondrial diseases; phosphorus metabolism disorders; porphyrias; and proteostasis deficiency.
Optionally, the glucose metabolism disorder is selected from one or more of diabetes mellitus (Type I or Type II); lactose intolerance; fructose malabsorption; galactosemia; and glycogen storage disease.
Optionally, the lipid metabolism disorder is selected from one or more of Gaucher's Disease (Type I, Type II, or Type III); Neimann-Pick Disease; Tay-Sachs Disease; Fabry's Disease; Sitosterolemia; Wolman's Disease; Refsum's Disease; and Cerebrotendinous Xanthomatosis.
Optionally, the malabsorption syndrome is selected from one or more of abetalipoproteinaemia; and coeliac disease.
Optionally, 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)).
Optionally, the metabolic disorder is cardiovascular disease.
Optionally, there is provided 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:
Optionally, there is provided a method of diagnosing or prognosing obesity, optionally abdominal obesity, in a subject; the method comprising the steps of:
CBG; IGKV2D-29; PROC; and LV205.
Optionally, there is provided a method of diagnosing or prognosing cardiovascular disease in a subject; the method comprising the steps of:
Optionally, there is provided 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:
Optionally, there is provided a method of diagnosing or prognosing obesity, optionally abdominal obesity, in a subject, irrespective of the weight of the subject; the method comprising the steps of:
Optionally, there is provided a method of diagnosing or prognosing cardiovascular disease in a subject, irrespective of the weight of the subject; the method comprising the steps of:
Reference will be made to the accompanying drawings in which:
Reference will now be made to the following non-limiting examples:
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.
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/m2, while the inclusion criteria for the normal weight (NW) control subjects were: age (20 to 70 years old), BMI<30 kg/m2, 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 Program—Adult 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.
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 μmol/min/L.
Fast Protein Liquid Chromatography (FPLC): Lipoproteins from frozen serum samples (150 μL) 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.
Cholesterol efflux capacity (CEC): J774 macrophages were labeled for 24h with 3H-cholesterol (1 μCi/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.
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 (NH4HCO3, 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.
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.
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 mean±SEM. 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.
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 (
Serum PON1 activity was significantly reduced in the obese group relative to NW (
4.70 ± 0.11
2.68 ± 0.14
a,b,cMean values with unlike superscript letters are significantly different between groups (P < 0.05)
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 (
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 (
a,bMean values with unlike superscript letters are significantly different between groups (P < 0.05)
Levels of 49 proteins were significantly different between MUO-HDL and NW-HDL (
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) (
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 (
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.
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.
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.
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 NW—again 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.
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.
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.
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.
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.
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.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1804734.0 | Mar 2018 | GB | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2019/057347 | 3/22/2019 | WO | 00 |
