Human serum/plasma lipoprotein analysis by NMR: Application to the study of diabetic dyslipidemia

The prevalence of type 2 diabetes mellitus (T2DM), a progressive metabolic disorder characterized by chronic hyperglycemia and the development of insulin resistance, has increased globally, with worrying statistics coming from children, adolescents, and young adults from developing countries like India. Here, we investigated unique circulating metabolic signatures associated with prediabetes and T2DM in an Indian cohort using NMR-based metabolomics.

The study subjects included healthy volunteers (N = 101), prediabetic subjects (N = 75), and T2DM patients (N = 108). Serum metabolic profiling was performed using ¹H NMR spectroscopy and major perturbed metabolites were identified by multivariate analysis and receiver operating characteristic (ROC) modules.

Of the 36 aqueous abundant metabolites, 24 showed a statistically significant difference between healthy volunteers, prediabetics, and established T2DM subjects. On performing multivariate ROC curve analysis with 5 commonly dysregulated metabolites (namely, glucose, pyroglutamate, o-phosphocholine, serine, and methionine) in prediabetes and T2DM, AUC values obtained were 0.96 (95 % confidence interval (CI) = 0.93, 0.98) for T2DM; and 0.88 (95 % CI = 0.81, 0.93) for prediabetic subjects, respectively.

We propose that the identified metabolite panel can be used in the future as a biomarker for clinical diagnosis, patient surveillance, and for predicting individuals at risk for developing diabetes.

Tilapia is one of the most common fish species that is intensively produced all over the world. However, significant measures at improving aquaculture health must be taken since disease outbreaks are often encountered in the rapidly developing aquaculture industry. Therefore, the objective of the study was designed to evaluate the metabolite changes in tilapia’ sera through ¹H NMR metabolomics in identifying the potential biomarkers responsible for immunomodulatory effect by the indigenous species of Malaysian microalgae Isochrysis galbana (IG). The results showed that IG-incorporated diet mainly at 5.0% has improved the immune response of innate immunity as observed in serum bactericidal activity (SBA) and serum lysozyme activity (SLA). The orthogonal partial least squares (OPLS) analysis indicated 5 important metabolites significantly upregulated namely as ethanol, lipoprotein, lipid, α-glucose and unsaturated fatty acid (UFA) in the 5.0% IG-incorporated diet compared to control. In conclusion, this study had successfully determined IG in improving aquaculture health through its potential use as an immune modulator. This work also demonstrated the effective use of metabolomics approach in the development of alternative nutritious diet from microalgae species to boost fish health in fulfilling the aquaculture's long-term goals.

Serum concentration of low-density lipoprotein cholesterol (LDL-C) is markedly reduced after a meal. Does postprandial cholesterol in LDL truly decline via clearance of LDL particles or is there simply a redistribution of cholesterol in LDL subclasses? Thus, we sought to evaluate whether postprandial decline of LDL-C reflects a reduction of LDL particle and to assess the correlation between proprotein convertase subtilisin/kexin type 9 (PCSK9) concentration and postprandial atherogenic lipoproteins profile.

Eighty-seven persons were enrolled in this study. We measured lipid profiles by enzymatic and nuclear magnetic resonance (NMR)-based methods and serum PCSK9 concentration by enzyme-linked immunosorbent assays before and after a meal. Plasma samples were collected after a 10-h fasting and 2 and 4 h post-meal.

Compared to the fasting status, there was significant postprandial decline of LDL-C measured enzymatically (LDL-Ce) at 2nd and 4th h [99.38 (80.43, 120.65) vs 95.51 (74.25, 117.17) vs 87.01 (69.99, 108.28) mg/dl, p < 0.000]. But there was no significant reduction in LDL particle and its cholesterol content (LDL-Cn) determined by NMR. Just the postprandial large LDL particle [186.45 (151.36, 229.42) vs 176.92 (147.43, 220.91) vs 181.77 (149.05, 224.17), p < 0.000] and its cholesterol content [19.10 (15.09, 22.37) vs 18.28 (14.59, 21.84) vs 17.79 (14.62, 22.14), p < 0.000] were greatly decreased at 2nd and 4th h compared to the fasting one. Interestingly, postprandial serum PCSK9 was decreased at 2nd and 4th h compared with fasting concentration [298.75 (233.25, 396.92) vs 257.34 (207.52, 342.36) vs 250.57 (215.02, 339.66) ng/ml, p < 0.000]. The postprandial percent decrease in serum PCSK9 at 4th h was positively correlated to the percent decline in postprandial LDL-Ce (r = 0.252, p = 0.019) but was independently associated with the percent increase in remnant cholesterol (r = 0.262, p = 0.016).

Postprandial decline of LDL-C determined enzymatically was not confirmed by NMR-based methods. Indeed, there exists cholesterol redistribution in LDL subclasses following a meal. The decrease of postprandial PCSK9 may be secondary to the increase in intrahepatic lipids following food intake.

Determining the particle size and number of lipoprotein components found in blood plasma (HDL, LDL and VLDL) has become an important clinical tool in diagnosing risk of cardiovascular disease. Proton (¹H) NMR spectroscopy methods to quantify lipoprotein particle subclasses have been advancing since NMR lineshape analysis of plasma samples was first proposed in the 1990’s. NMR methods, including a more recent DOSY-based diffusion spectroscopy test, provide the foundation for the advanced lipoprotein tests, including Lipoprotein® and Liposcale® analyses available for clinical use to determine particle size and number. At the time of this submission, no NMR studies exist which explore physical parameters of individual lipoprotein fractions when they are deformed by pressure. This study reports ¹H NMR frequency shifts and T₂* measurements for the broad methyl peak attributed to terminal methyls (cholesteryl positions 26, 27 and terminal acyl methyl groups) in three primary lipoprotein fractions as a function of hydraulic pressure. This terminal CH₃ resonance shifted linearly upfield as a function of pressure for HDL and VLDL (observed slopes of −0.014 Hz/bar). The LDL terminal CH₃ resonance shows segmented behavior, with a shallow slope between 0–900 bar (−0.008 hz/bar) and a slope similar to HDL and VDL across the range from 1000 to 2400 bar (slope −0.016 Hz/bar). ¹H T₂* values measured for VLDL and HDL dropped linearly with increasing pressure. ¹H T₂* values for LDL demonstrated segmented behavior as a function of pressure. The unique behavior observed for LDL terminal CH₃ frequency and ¹H T₂* trends suggests an approximate pressure at which phase transition occurs.

Increased concentrations of calculated remnant cholesterol in triglyceride-rich lipoproteins are observationally and genetically, causally associated with increased risk of ischemic heart disease; however, when measured directly, the fraction of plasma cholesterol present in remnant particles is unclear. We tested the hypothesis that a major fraction of plasma cholesterol is present in remnant lipoproteins in individuals in the general population.

We examined 9293 individuals from the Copenhagen General Population Study using nuclear magnetic resonance spectroscopy measurements of total cholesterol, free- and esterified cholesterol, triglycerides, phospholipids, and particle concentration. Fourteen subclasses of decreasing size and their lipid constituents were analysed: six subclasses were very low-density lipoprotein (VLDL), one intermediate-density lipoprotein (IDL), three low-density lipoprotein (LDL), and four subclasses were high-density lipoprotein (HDL). Remnant lipoproteins were VLDL and IDL combined.

Mean nonfasting cholesterol concentration was 1.84 mmol/L (72 mg/dL) for remnants, 2.01 mmol/L (78 mg/dL) for LDL, and 1.83 mmol/L (71 mg/dL) for HDL, equivalent to remnants containing 32% of plasma total cholesterol. Of 14 lipoprotein subclasses, large LDL and IDL were the ones containing most of plasma cholesterol. The plasma concentration of remnant cholesterol was from ∼1.4 mmol/L (54 mg/dL) at age 20 to ∼1.9 mmol/L (74 mg/dL) at age 60. Corresponding values for LDL cholesterol were from ∼1.5 mmol/L (58 mg/dL) to ∼2.1 mmol/L (81 mg/dL).

Using direct measurements, one third of total cholesterol in plasma was present in remnant lipoproteins, that is, in the triglyceride-rich lipoproteins IDL and VLDL.