Urinary biomarkers for the diagnosis of cervical cancer by quantitative label-free mass spectrometry analysis

Abstract

Due to the invasive procedure associated with Pap smears for diagnosing cervical cancer and the conservative culture of developing countries, identifying less invasive biomarkers is of great interest. Quantitative label-free mass spectrometry was performed to identify potential biomarkers in the urine samples of patients with cervical cancer. This technique was used to study the differential expression of urinary proteomes between normal individuals and cancer patients. The alterations in the levels of urinary proteomes in normal and cancer patients were analyzed by Progenesis label-free software and the results revealed that 60 proteins were upregulated while 73 proteins were downregulated in patients with cervical cancer. This method could enrich high molecular weight proteins from 100 kDa. The protein-protein interactions were obtained by Search Tool for the Retrieval of Interacting Genes/Proteins analysis and predicted the biological pathways involving various functions including cell-cell adhesion, blood coagulation, metabolic processes, stress response and the regulation of morphogenesis. Two notable upregulated urinary proteins were leucine-rich α-2-glycoprotein (LRG1) and isoform-1 of multimerin-1 (MMRN1), while the 3 notable downregulated proteins were S100 calcium-binding protein A8 (S100A8), serpin B3 (SERPINB3) and cluster of differentiation-44 antigen (CD44). The validation of these 5 proteins was performed by western blot analysis and the biomarker sensitivity of these proteins was analyzed individually and in combination with receiver operator characteristic curve (ROC) analysis. Quantitative mass spectrometry analysis may allow for the identification of urinary proteins of high molecular weight. The proteins MMRN1 and LRG1 were presented, for the first time, to be highly expressed urinary proteins in cervical cancer. ROC analysis revealed that LRG1 and SERPINB3 could be individually used, and these 5 proteins could also be combined, to detect the occurrence of cervical cancer.

Keywords: S100A8; cervical cancer; cluster of differentiation 44 antigen; label-free; leucine-rich α-2-glycoprotein; multimerin-1; serpin B3; urinary biomarkers.

Figure 1.

Protein expression. (A) Venn diagram comparing the total differential protein expression from cervical cancer between identified proteins by LC-MS/MS and the Human Protein Atlas Database. Differentially expressed urinary proteins from normal individuals and patients with cervical cancer were analyzed by PANTHER classification based on (B and D) molecular function and (C and E) biological processes: (B and C) 60 upregulated proteins and (D and E) 73 downregulated proteins. LC, liquid chromatography; MS, mass spectrometry.

Figure 2.

Search Tool for Retrieval of Interacting Genes/Proteins analysis. A network of interacting proteins: (A) 60 upregulated urinary proteins and (B) 73 downregulated urinary proteins from patients with cervical cancer.

Figure 3.

Western blot analysis. (A) Levels of LRG1, MMRN1, CD44, SERPINB3 and S100A8 in the urine of normal (n=8) individuals and patients with different stages (C1-C2=stage I; C3-C11=stage II; C12-C15=stage III; C16-C18=stage IV) of cervical cancer were studied by western blot analysis. (B) The loading controls were subsequently prepared by staining the membrane with Coomassie blue following western blotting. LRG1, leucine-rich α-2-glycoprotein; MMRN1, multimerin-1; SERPINB3, serpin B3; CD44, cluster of differentiation 44 antigen.

Figure 4.

ROC curve of the 5 individual potential biomarkers (LRG1, MMRN1, CD44, S100A8 and SERPINB3) was calculated to determine the area under the curve, sensitivity (95% CI) and specificity (95% CI). The ROC for the 5-biomarker panel was analyzed by combining the 5 potential biomarkers. ROC, receiver operating characteristic; LRG1, leucine-rich α-2-glycoprotein; MMRN1, multimerin-1; SERPINB3, serpin B3; CD44, cluster of differentiation 44 antigen; CI, confidence interval; AUC, area under the curve.