Metabolomics and proteomics analyses reveal the role of the glycerophospholipid metabolism pathway in unexplained recurrent spontaneous abortion

Abstract

Background: Unexplained recurrent spontaneous abortion (URSA) is a complex pregnancy complication with a high miscarriage rate. Incomprehensive understanding of the molecular mechanism in URSA also leads to a lack of effective treatment methods. Hence, the current study aimed to explore the underlying pathogenesis of URSA applying metabonomic and bioinformatics analysis.

Methods: The decidual tissues of eight URSA samples and eight normal pregnancy (normal control, NC) samples were collected for liquid chromatography-mass spectrometry (LC-MS) analysis using the Progenesis QI metabolomics software. The orthogonal partial least squares discrimination analysis (OPLS-DA) and the Human Metabolome Database (HMDB) were employed for differential metabolite analysis and pathway enrichment analysis, respectively. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway topological analysis was performed to rank the importance of pathways involved in URSA, and differential proteins were identified based on fold change difference. Finally, a metabolic network was visualized by the Cytoscape tool.

Results: After LC-MS analysis and quality control, samples in the same group showed high consistency and reliability. Differential metabolites between NC and URSA groups were mainly enriched to five biological processes, with glycerophospholipid metabolism pathway containing the greatest number of differential metabolites. KEGG enrichment analysis showed significant differences in glycerophospholipid metabolism, bile secretion, and choline metabolism pathways, with glycerophospholipid metabolism showing a higher pathway importance. Proteome and metabolome analysis revealed a total of 65 overlapping pathways involved in the differential proteins and differential metabolites, and finally PLD1, CHPT1 and PLA2G2A were identified as the key genes in glycerophospholipid metabolism pathway.

Conclusion: LC-MS analysis revealed that glycerophospholipid metabolism pathway and its three key genes were crucially involved in URSA progression, providing novel insights into the treatment strategy of URSA.

Keywords: Bioinformatics technology; Glycerophospholipid metabolism; Liquid chromatography-mass spectrometry; Orthogonal partial least squares discrimination analysis; Unexplained recurrent spontaneous abortion.

Figure 1. Sample reliability detection for metabolomics analysis.

(A) Heatmap of the sample correlation. (B) Principal components analysis (PCA) for outlier. (C) OPLS-DA for differential metabolites analysis. (D) Permutation testing for the reliability of OPLS-DA.

Figure 2. Functional enrichment analysis of metabolites.

(A) KEGG enrichment analysis of differential metabolites. (B) The number of metabolites in different pathways. (C) Pathway difference analysis between NC and URSA by using hypergeometric test. (D) Pathway importance score analysis.

Figure 3. Differential protein analysis between URSA and NC groups.

(A) Volcano plot of differential proteins. (B) GO enrichment analysis of differential proteins. (C) KEGG enrichment analysis of differential proteins.

Figure 4. Conjoint analysis of differential metabolites and proteins.

(A) Venn plot of differential metabolites and proteins. (B) The top10 pathways of differential metabolites and proteins enriched. (C) The KEGG enrichment analysis of different proteins/metabolites in each group.

Figure 5. Metabolic network analysis.

(A) Correlation heatmap of differential metabolites and proteins. (B) The metabolic network plot. (C) ssGSEA for glycerophospholipid metabolism pathway difference.

Figure 6. Expression validation and functional assessment of key genes.

(A) Based on qRT-PCR to verify the mRNA expression levels of decidua markers PRL and IGFBP1. (B) Based on qRT-PCR to verify differential expression of key genes (PLD1, CHPT1, and PLA2G2A) in control and decidualization groups. (C) CCK-8-based assay to validate the effect on proliferation of decidualization induced T-hESCs after silencing CHPT1. (D) Flow cytometry to assess the effect on the apoptotic capacity of decidualization induced T-hESCs after silencing CHPT1. Control group refers to T-hESCs that have not undergone decidualization induction; decidualization group refers to the induction of decidualization of T-hESCs in vitro. All experiments were three independent replications. ** indicates p < 0.01 and **** indicates p < 0.0001.