Urinary Peptides Associated Closely with Gestational Diabetes Mellitus

Abstract

Gestational diabetes mellitus (GDM) is a common disease of pregnant women, which has a higher incidence in recent years. The purpose of this study is to explore urinary biomarkers that could predict and monitor gestational diabetes mellitus (GDM). Urine samples from 30 normal pregnant women and 78 GDM patients were collected and purified by weak cationic exchange magnetic beads (MB-WCX), then analyzed by matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF-MS). The urinary peptide signatures of the two groups were compared by BioExplorer software. The potential ability of the differently expressed peptides to distinguish GDM patients from normal pregnant women was evaluated by receiver operating characteristic (ROC) analysis. At last, the differently expressed peptides were identified by liquid chromatography tandem mass spectrometry (LC-MS). There were four differently expressed peptides (m/z 1000.5, 1117.5, 1142.9, and 2022.9) between two groups, which were identified as fragments of urinary albumin, α2-macroglobulin, human hemopexin, and α1-microglobulin, respectively. The diagnostic efficacy of m/z 1142.9 was better than the other peptides. The area under the curve (AUC) of the m/z 1142.9 was 0.690 (95% CI: 0.583-0.796). The discovery of urinary polypeptides provides the possibility for the early prediction of GDM and the monitoring of glucose metabolism in GDM patients by a noninvasive method.

Figure 1

Typical urinary sample mass spectra from MALDI-TOF-MS after being purified by MB-WCX: (a) one sample of a normal pregnant woman; (b) one sample of a GDM patient.

Figure 2

Urine polypeptides in the N group and the GDM group were analyzed. The samples in the N and GDM groups were urine samples of 30 normal pregnant women and 78 GDM patients, respectively. (a) The average peak area distributions of all polypeptide peaks were measured in two groups. Five urine polypeptides which were indicated by the arrow were statistically significant in the comparative analysis of two groups (P < 0.05). (b) In the comparative analysis of five polypeptides between the N and GDM groups, there were significant differences in the m/z 1000.5 and 1117.5 (left, P < 0.05). The distributions of two polypeptides in two groups are shown (middle). The peak value of two polypeptides in the GDM group was higher than that in the N group (right, P < 0.05). (c) There were significant differences in the m/z 1142.9, 2022.9, and 4636.5 (left, P < 0.05). The distributions of three polypeptides in two groups are shown (middle). The peak value of three polypeptides in the GDM group was lower than that in the N group (right, P < 0.05).

Figure 3

Comparison of the difference of urine peptides between the GM1 (n = 45) and GM2 (n = 33) groups. (a) The distribution of molecules with m/z 1000.5, 2022.9, and 4636.5 in two groups is shown. (b) There were significant differences between the GM1 and GM2 groups in three molecules with m/z 1000.5, 2022.9, and 4636.5 (P < 0.05). (c) The distribution of molecules with m/z 1078.6, 1117.5, 1142.9, 1289.6, and 1501.8 in two groups is shown. (d) There was no significant difference between the GM1 and GM2 groups in three molecules with m/z 1078.6, 1117.5, 1142.9, 1289.6, and 1501.8 (P > 0.05).

Figure 4

Trend analysis of differential polypeptide peaks and FPG. (a) Trends of FPG and the molecule with m/z 1000.5 in the N group (n = 30), GM1 group (n = 45), and GM2 group (n = 33). (b) Trends of 1/FPG and the molecules with m/z 2022.0 and 4636.5 in three groups.

Figure 5

ROC analysis. The ROC curve and the AUC were established according to the peak value of urine polypeptides of GDM patients (n = 78) and normal pregnant women (n = 30) with m/z 1000.5, 1117.5, 1142.9, 2022.9, and 4636.5 and comprehensive index.