Detection and Characterization of Phosphorylation, Glycosylation, and Fatty Acid Bound to Fetuin A in Human Blood

Abstract

The hepatokine fetuin A (Fet A) has been associated with diverse pathological states such as insulin resistance, type 2 diabetes, macrovascular disease, and systemic ectopic and vascular calcification. Fet A may also play a role in tumor growth and metastasis. The biological activity of Fet A may be affected by various modifications, including phosphorylation, O- and N-glycosylation and fatty acid binding. We developed an antibody-based assay for the detection of Fet A phosphorylated at serine 312. Fatty acid pattern was determined by gas chromatography. Using the antibody, we found that the phosphorylation was stable in human plasma or serum at room temperature for 8 h. We observed that Fet A is present in several glycosylation forms in human plasma, but the extent of Ser³¹² phosphorylation was not associated with glycosylation. The phosphorylation pattern did not change during an oral glucose tolerance test (0-120 min). We further found that human Fet A binds preferentially saturated fatty acids (>90%) at the expense of mono- and poly-unsaturated fatty acids. Our results indicate that different molecular species of Fet A are present in human plasma and that these different modifications may determine the different biological effects of Fet A.

Keywords: FAM20C; fatty acid bound to fetuin A; fatty liver; fetuin A; glycosylation; obesity; phosphorylation.

Figure 1

Enzymatic dephosphorylation and phosphorylation of Fet A. (A) Western blot analysis of time-dependent λ-phosphatase-catalyzed dephosphorylation of Fet A (red) pFet A (green) and overlay pFet A/Fet A (yellow/red). (B) Densitometric evaluation of Western blot data of pFet A (green) and pFet A/Fet A (orange) is shown. A representative Western blot is shown, three experiments have been performed. (C) Western blot analysis of time-dependent phosphorylation of commercial Fet A with FAM20C. (D) For comparison, the same Western blot was also developed by the conventional chemiluminescence method. (E) This figure shows the densitometric evaluation of the time-dependent phosphorylation of commercial Fet A. For detailed description of the procedure and the quantification see the Methods section. Bold arrows indicate molecular weight marker; AU, arbitrary units.

Figure 2

Stability of pSer³¹² Fet A in human blood samples collected in different routine blood tubes at Western blot analysis of blood samples from a healthy volunteer and the stability of pFet A was evaluated in sample material used in routine clinical chemistry. Blood samples were collected in tubes containing different additives i.e., heparin, EDTA, and EDTA mixture as described in the method section. Bold arrows indicate molecular weight marker (m.w.m).

Figure 3

Assay variations of two subjects with different BMI. Representative Western blots of Fet A, pFet A and overlay from two plasma samples from subjects with (A) BMI of 20 kg/m² and (B) BMI of 35 kg/m². To evaluate precision of the analysis 23 samples of each individual were run in one assay. For evaluation of the assay variation the last 3 lanes were not included for the calculation of variation because of unproper run of these samples. Western blot analysis and densitometric evaluations were performed as described in the Method section. m.w.m = molecular weight marker.

Figure 4

Deglycosylation of Fet A from control, prediabetic, and diabetic patients and effect on Ser³¹² Fet A phosphorylation. Serum sample from control, prediabetic and diabetic patients (two of each) were analyzed untreated or treated with the deglycosylation enzymes as indicated. Samples were analyzed for Fet A, pFet A, and overlay by Western blotting as described in the Methods section; bold arrows indicate molecular weight markers.

Figure 5

Analysis of plasma pSer³¹² pFet A and Fet A samples obtained from non-diabetic individuals (#1–35) Untreated (A) and completely deglycosylated (B) samples obtained from non-diabetic individuals (n = 35) were analyzed by Western blotting as described in the Method section; m.w.m = molecular weight marker.

Figure 6

Effect of short-term metabolic changes on plasma pSer³¹² pFet A and Fet A induced by an OGTT. Serum samples were obtained from 12 controls during an OGTT and sample taken at 0 min (basal) and 120 min were analyzed by Western blotting as described in the Method section. PG: Plasma glucose in mmol/L. Bold arrows indicate molecular weight markers.

Figure 7

Free fatty acid analysis of human serum and Fet A preparation. Free fatty acid pattern of a representative human serum, Fet A from Sigma-Aldrich, and rhFet A were analyzed as described in the Method section. (A) A gas chromatogram of the fatty acid pattern of a serum from a representative healthy individual is shown. (B,C) show chromatograms of the fatty acid pattern of commercial Fet A and of rhFet A, respectively. 16:0, 18:0 and 20:0 indicate the saturated fatty acids palmitate, stearate and arachinate; 18:1 indicates the monounsaturated fatty acid oleate and the poly-unsaturated fatty acids as indicated by 18:2 = linoleate, 20:4 = arachidonate and 22:6 = docosahexaenoic acid, respectively; IST = internal standard. Since the areas of the respective peaks are not readable in the original chromatograms they are listed below (arbitrary units): 16:0 559.8, 175.2, 38.95; 18:0: 295.8, 196.2, 44.75; 18:1: 696.8, 10.0, 4.47; 18:2: 281.2 6.65, 1.34; 20:0: 4.37, 4.75, 1.58; 20:4: 32.5, 3.29, 0.25; IS: 223.7, 447.7, 90.0 and 22:6: 17.9, 1.3, 0.71 for chromatogram A, B, and C, respectively.