Review Article

Dysfunctional High-Density Lipoprotein: An Innovative Target for Proteomics and Lipidomics

Table 3

Contributions of proteomics and lipidomics to cardiovascular risk estimation.

Author [reference] Methodology Results

Vaisar et al. [66] 7 males with established CVD were compared with 6 healthy, age-matched subjects, whose HDL-C, HDL3, and HDL-associated proteins were studied. No significant differences in HDL-C concentration were found between groups (40 ± 11 mg/dL versus 45 ± 12 mg/dL, resp.). In individuals with CVD, the proteins most commonly found associated with HDL3 were Apo C-IV, PON-1, C3, Apo A-IV, and Apo E. HDL3 of control subjects were found to have increased levels of clusterin and vitronectin.

Tan et al. [222]40 subjects with established CVD were compared to 40 healthy subjects, who had their HDL3 and HDL2 studied quantitatively and qualitatively. No significant differences in HDL-C concentration were found between groups. However, in subjects with CVD, HDL3 were found to be rich in Apo E, Apo A-I, Apo A-IV, Apo L1, Serum Amyloid P component, PON-1, α-1B glycoprotein, and vitamin D-binding protein, along with low Rab levels. Likewise, HDL2 were found to have low levels of Apo A1, Apo E, PON-1, Apo L1, haptoglobin, serotransferrin, Rab7, and complement factor B, along with increased Serum Amyloid P component, α-1 antitrypsin, and acid ceramidase.

Yan et al. [223] Case-control study comprising 10 males with chronic heart disease versus 10 healthy subjects matched by age, Body Mass Index, and lipid profiles, who had their HDL composition studied for comparison. 12 HDL-associated proteins differed significantly between subjects with chronic heart disease and healthy individuals, most of which participate in lipid metabolism. Gene ontology analysis revealed proteins involved in inflammation and other immune responses (SAA, C5, histone H1, and fibrinogen beta chain) to be differentially upregulated, whereas proteins involved in lipid metabolism (Apo C-I, Apo C-II, and fatty acid-binding protein) were differentially downregulated. Further ELISA analysis supported these findings, confirming higher SAA and lower Apo C-I in subjects with chronic heart disease versus healthy subjects (126.5 ± 67.3 g/mg versus 68.7 ± 12.4 μg/mg, ; and 68.8 ± 14.4 μg/mg versus 81.1 ± 10.6 μg/mg, , resp.).

Lepedda et al. [224] The apolipoproteins of 79 patients undergoing carotid endarterectomy (due to stenosis >70%) were isolated and compared with those from 57 normolipemic subjects. Apo A-I, Apo C-II, Apo C-III, Apo E, Apo D, and SAA were found to be associated with HDL. Only SAA was found to display a significant differential distribution, being more abundant in the group undergoing carotid endarterectomy (). SAA may be a CVR marker reflecting HDL quality.

Holzer et al. [226] HDL was isolated from end-stage renal disease patients on maintenance hemodialysis () and healthy subjects (); proteomic techniques allowed identification of HDL-associated proteins in both groups. Patients on hemodialysis had lower levels of HDL-C (61 mg/dL versus 43 mg/dL, ). 35 HDL-associated proteins were identified, most abundantly Apo A-I and Apo A-II. SAA was found only in the HDL of patients on hemodialysis. 9 proteins were found to be significantly altered in this group, including SAA. In addition, HDL of this group displayed lower proportions of phospholipids and higher proportions of LPC.

Mangé et al. [227] A quantitative proteomic analysis was realized in 23 patients on hemodialysis and 23 age-matched control subjects. Individuals on hemodialysis showed significantly lower HDL-C and serotransferrin levels, along with increased expression of Apo C-II and Apo C-III (with greater Apo C-II/Apo C-III ratio), which may act as markers of HDL maturity.

Weichhart et al. [230] HDL was isolated from patients with end-stage renal disease and healthy subjects through sequential ultracentrifugation. Shotgun proteomics was used to identify HDL-associated proteins in a uremia-specific pattern. Gene ontology functional analysis showed that in the group with end-stage renal disease, HDL-associated proteins involved in lipid metabolism were disrupted (including Apo A-I, Apo E, Apo A-IV, PON-1, LCAT, and PLTP). Instead, their HDL were found to be rich in surfactant protein B, Apo C-II, SAA, and α-1-microglobulin, representing a possible explanation for the increased inflammation and cardiovascular mortality seen in uremia.

Yassine et al. [231] 11 subjects with DM2, 15 with DM2 plus established CVD, and 8 control subjects had their HDL isolated in order to determine relative ratios of oxidation of the M148 residue of Apo A-I. Patients with DM2 plus CVD displayed significantly lower levels of HDL-associated Apo A-I when compared to subjects with DM2 only (84 ± 39 versus 90 ± 40; ). Molecular methods allowed determination of a relative oxidation ratio of the M148 residue in Apo A-I. This ratio was significantly higher in the groups with DM2 and CVD (0.236 ± 0.084) and DM2 only (0.127 ± 0.037), in comparison to the control group (0.087 ± 0.02); and .

Jensen et al. [232]173,230 subjects from the Nurses' Health Study (NHS) and the Health Professionals Follow-Up Study (HPFS), who had their levels of Apo C-III-associated and non-ApoC-III-associated HDL quantified and evaluated in regard to CVR.HDL-C concentration was negatively correlated with CVR in both studies (IRR: 0.78; 95% IC: 0.63–0.96, ). Nevertheless, increased levels of non-Apo C-III-associated HDL were negatively associated with CVR (IRR: 0.66; 95% IC: 0.53–0.83, ), whereas increased levels of Apo C-III-associated HDL were positively associated with CVR (IRR: 1.18; 95% IC: 1.03–1.34, ).

Ståhlman et al. [225] Mass spectrometry was used to characterize the lipidome of 3 groups of women from the DIWA study: (a) control group; (b) DM2 + insulin resistance + dyslipidemia; (c) DM2 + insulin resistance + normolipemia. Smaller HDL particles were found in the dyslipidemic group, with increased LPC (13%) palmitate-rich triacylglycerols and diacylglycerols (77%) possibly reflecting enhanced CETP activity. The subjects also displayed a high Apo A-I/plasmalogen ratio compatible with oxidative stress seen in DM2.

Kostara et al. [244] Case-control study with 60 subjects with normal coronary arteries and 99 patients with established CVD grouped by severity of coronary artery stenosis (mild, moderate, and severe). Lipidomic analysis assessed patterns in the constitution of HDL in each group. HDL-C was significantly lower in the mild disease group versus severe disease group (43.6 ± 10.9 mg/dL versus 38.4 ± 6.8 mg/dL). Subjects with CVD had higher proportions of saturated fatty acids, phospholipids, triacylglycerides, and cholesteryl esters in HDL in comparison to controls, along with lower proportions of sphingomyelin and phosphatidylcholine. Likewise, subjects with mild disease had greater proportions of phosphatidylcholine, unsaturated fatty acids, omega-3 fatty acids, and sphingomyelin than subjects with severe disease.

Yetukuri et al. [240] Subjects from the Fibrate Intervention and Event Lowering in Diabetes (FIELD) substudy, whose changes in proteome and lipidome were evaluated after receiving (a) fenofibrate 200 mg daily or (b) placebo. No difference was found in HDL-C levels between groups (). HDL from the fenofibrate group had lower LPC and higher sphingomyelin and Apo A-II.