A combination analysis based on bioinformatics tools reveals new signature genes related to maternal obesity and fetal programming

• Published in: Frontiers in medicine Vol. 11; p. 1434105
• Main Authors: Liu, Chunhong, Lu, Yulan, Huang, Chunchuan, Zeng, Yonglong, Zheng, Yuye, Wang, Chunfang, Huang, Huatuo
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 04.09.2024
• Subjects: bioinformatics; fetal programming; Medicine; obesity; placenta; pregnancy; placenta; fetal programming; obesity; bioinformatics; pregnancy
• ISSN: 2296-858X; 2296-858X
• DOI: 10.3389/fmed.2024.1434105

Maternal obesity significantly influences fetal development and health later in life; however, the molecular mechanisms behind it remain unclear. This study aims to investigate signature genes related to maternal obesity and fetal programming based on a genomic-wide transcriptional placental study using a combination of different bioinformatics tools. The dataset (GSE128381) was obtained from Gene Expression Omnibus (GEO). The data of 100 normal body mass index (BMI) and 27 obese mothers were included in the analysis. Differentially expressed genes (DEGs) were evaluated by limma package. Thereafter, functional enrichment analysis was implemented. Then, weighted gene co-expression network analysis (WGCNA) and the least absolute shrinkage and selection operator (LASSO) analysis were used to further screening of signature genes. Simple linear regression analysis was used to assess the correlation between signature genes and newborn birth weight. Gene set enrichment analysis (GSEA) was implemented to study signaling pathways related to signature genes. The expression of the signature genes was also explored in 48 overweight mothers in the same dataset. A total of 167 DEGs were obtained, of which 122 were up-regulated while 45 were down-regulated. The dataset was then clustered into 11 modules by WGCNA, and the MEbrown was found as the most significant module related to maternal obesity and fetal programming (cor = 0.2, = 0.03). The LASSO analysis showed that , , , , and are signature genes related to maternal obesity and fetal programming, which were increased in the placenta of obese mothers compared to those with normal BMI. The area under the curve (AUC) of the signature genes in the receiver operating characteristic curve (ROC) was 0.709, 0.660, 0.674, 0.667, and 0.717, respectively. Simple linear regression analysis showed that HOXB5 was associated with newborn birth weight. GSEA analysis revealed that these signature genes positively participate in various signaling pathways/functions in the placenta. , , , , and are novel signature genes related to maternal obesity and fetal programming, of which is implicated in newborn birth weight.