Placental gene-expression profiles of intrahepatic cholestasis of pregnancy reveal involvement of multiple molecular pathways in blood vessel formation and inflammation

Abstract

Background: Intrahepatic cholestasis of pregnancy (ICP) is a pregnancy-associated liver disease with potentially deleterious consequences for the fetus, particularly when maternal serum bile-acid concentration >40 μM. However, the etiology and pathogenesis of ICP remain elusive. To reveal the underlying molecular mechanisms for the association of maternal serum bile-acid level and fetal outcome in ICP patients, DNA microarray was applied to characterize the whole-genome expression profiles of placentas from healthy women and women diagnosed with ICP.

Methods: Thirty pregnant women recruited in this study were categorized evenly into three groups: healthy group; mild ICP, with serum bile-acid concentration ranging from 10-40 μM; and severe ICP, with bile-acid concentration >40 μM. Gene Ontology analysis in combination with construction of gene-interaction and gene co-expression networks were applied to identify the core regulatory genes associated with ICP pathogenesis, which were further validated by quantitative real-time PCR and histological staining.

Results: The core regulatory genes were mainly involved in immune response, VEGF signaling pathway and G-protein-coupled receptor signaling, implying essential roles of immune response, vasculogenesis and angiogenesis in ICP pathogenesis. This implication was supported by the observed aggregated immune-cell infiltration and deficient blood vessel formation in ICP placentas.

Conclusions: Our study provides a system-level insight into the placental gene-expression profiles of women with mild or severe ICP, and reveals multiple molecular pathways in immune response and blood vessel formation that might contribute to ICP pathogenesis.

Figures

Figure 1

Hierarchical clustering of differentially expressed genes in placentas from healthy pregnancies and women with ICP. (A) Clinical features of healthy women and women with mild or severe ICP. 10 pregnant women were recruited for each group, and data are shown as mean ± s.d. (n = 10). (B) Hierarchical clustering of genes whose expression was altered by more than 1.5-fold in ICP placentas comparing to those in healthy placentas. Red and green represent individual genes that were differentially up-regulated or down-regulated, respectively.

Figure 2

Significantly enriched Gene Ontology (GO) terms associated with ICP pathogenesis. The primary functions of the sorted differentially expressed genes in ICP placentas were assessed by GO analysis. Ten significantly enriched GO terms in both mild ICP and severe ICP (Intersection, A) and five significantly enriched GO terms in mild ICP only (B) and severe ICP only (C) are shown respectively. The vertical axis is the GO category and the horizontal axis is the enrichment value of GO.

Figure 3

Gene-interaction network. Genes associated with the significantly enriched GO terms were analyzed by gene-interaction network, which was built according to the relationships among genes, proteins, and compounds in the KEGG database. Circles, squares, and triangles represent, respectively, genes differentially regulated in both mild ICP and severe ICP (Intersection), mild ICP only, and severe ICP only. Red and green represent genes that were differentially up-regulated or down-regulated, respectively. Detailed gene-gene relationships with the degree of each gene are shown in Additional file 3: Spreadsheet S1.

Figure 4

Validation of the identified core regulatory genes associated with ICP pathogenesis by qPCR. Genes with expression patterns in placental tissues from healthy group, mild ICP, and severe ICP (10 pregnancies per group) consistent with those in the microarray data are shown: Genes involved in (A) immune response, (B) VEGF signaling and (C) G-protein-coupled receptor (GPCR) signaling. Relative expression of each target gene was normalized to GAPDH, and expression levels in the healthy group were defined as 1. All reactions were performed in triplicate, and data are presented as mean ± s.d. (n = 10). *p < 0.05; **p < 0.01; ***p < 0.001; NS, not significant. Primer pairs used were listed in Additional file 1: Table S1.

Figure 5

Immune cell infiltration in placental tissues from healthy group, mild ICP, and severe ICP. (A) Representative images of immunohistochemistry staining for the leukocyte marker CD45 in placental tissues from healthy group, mild ICP, and severe ICP (10 pregnancies per group). Brown staining (arrows) indicates reactivity. Scale bars represent 100 μm. (B) Quantification of leukocytes (CD45 positive), T cell (CD3 positive) and B cell (CD19 positive) in placental tissues from healthy group, mild ICP, and severe ICP (10 pregnancies per group). Data are presented as mean ± s.d. (n = 10). **p < 0.01; ***p < 0.001.

Figure 6

Histological staining of placental tissues from healthy group, mild ICP, and severe ICP. (A) Representative hematoxylin and eosin staining of placental tissues from three groups of pregnancies (10 pregnancies per group). Scale bar represents 200 μm. (B) Quantification of capillaries per villus of placental tissues from three groups of pregnancies (10 pregnancies per group). Data are presented as mean ± s.d. (n = 10). *p < 0.05; **p < 0.01.