Dysregulation of decidual NK cell proliferation by impaired decidual cells: a potential contributor to excessive trophoblast invasion in placenta accreta spectrum

Abstract

Aberrant interactions among decidual stromal cells, decidual natural killer (dNK) cells, and trophoblasts are implicated in placenta accreta spectrum (PAS) pathogenesis, though the underlying mechanisms remain unclear. This study investigates the relationship between defective decidualization of endometrial stromal cells and dysregulated dNK cell proliferation, which may contribute to excessive trophoblast invasion and the development of PAS. We established an in vitro system that mimics the decidual microenvironment to investigate these interactions. Maternal decidua-derived mesenchymal stem cells (MD-MSCs) from healthy pregnancies and PAS patients (PA-MSCs) were isolated and induced to undergo decidualization using hormonal and chemical stimuli. Peripheral natural killer (pNK) cells were then co-cultured with these MSCs to generate dNK-like cells. Cellular interactions among MSCs, dNK-like cells, and trophoblasts were evaluated using an in vitro co-culture system. Decidualization defects in PA-MSCs were marked by reduced morphological changes and dysregulated expression of decidual markers, potentially associated with estrogen receptor (ER) overexpression. Furthermore, both PA-MSCs and normal MD-MSCs similarly regulated trophoblast invasion, suggesting an indirect impact of impaired decidual cells on trophoblast behavior. Interestingly, decidualized MD-MSCs (De-MD-MSCs) showed the potential to induce the conversion of pNK cells into dNK-like cells, which displayed reduced cytotoxicity on trophoblasts and elevated KIR2DL4 expression. These dNK-like cells exhibited increased proliferation when co-cultured with PA-MSCs, enhancing trophoblast invasion and spiral artery remodeling. Conditioned medium derived from PA-MSCs-induced dNK-like cells demonstrated a higher capacity to promote trophoblast invasion in a dose-dependent manner. The abnormal proliferation of dNK cells induced by impaired decidual cells may contribute to the pathogenesis of PAS, providing valuable insights into its mechanisms and potential therapeutic interventions.

Keywords: decidual natural killer cell; decidualization; immune tolerace; placenta accreta spectrum (PAS); trophoblast invasion.

FIGURE 1

Characterization of MD-MSCs for decidualization potential through morphological changes and identification of decidual cell markers. (A) Schematic representation of the workflow for MD-MSCs isolation and culture for subsequent analysis. (B) (a) Representative images of normal MD-MSCs stained with WGA under E2/P4 and cAMP/MPA treatments. (b) Quantification of changes in circularity and roundness in normal MD-MSCs following decidualization treatment, assessed using ImageJ software. (C) (a) Venn diagrams display differentially upregulated genes’ overlap in both E2/P4 and cAMP/MPA treatment groups. (b) A selection of 6431 upregulated genes was subsequently overlaid with two published databases to identify 33 potential decidual cell markers. 9 annotated genes are selected from the upregulated genes in decidual cells within the curated database TISSUES (BTO:0002770). (D) Heatmap presenting the expression levels of the 33 annotated upregulated genes during decidualization in MD-MSCs and two reference databases. Values are expressed as z-scores, which have been normalized to the gene expression levels. The asterisk key “*” signifies the annotation of 9 genes for potential validation as decidual cell markers. (E) Validation of 9 selected genes through qPCR analysis. Prl, Igfbp1, and Scara5 are identified as potential marker genes for De-MD-MSCs. Values are normalized to (F) The protein expression levels of PRL, IGFBP1, and SCARA5 of MD-MSCs under E2/P4 and cAMP/MPA decidualization treatment. Values are expressed as mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001. Prl: Prolactin; Igfbp1: Insulin-like growth factor binding protein 1; Scara5: Scavenger receptor class A type 5.

FIGURE 2

PA-MSCs demonstrate decidualization defects with reduced morphological changes and decidual cell marker expression levels. (A) (a) Representative images and quantification of (b) circularity and roundness changes in PA-MSCs stained with WGA under E2/P4 and cAMP/MPA treatments compared with MD-MSCs. (B) The mRNA expression levels of three potential decidual cell markers, Prl, Igfbp1, and Scara5, in PA-MSCs compared with MD-MSCs. (C) The protein expression levels of three potential decidual cell markers, PRL, IGFBP1, and SCARA5, in PA-MSCs compared with MD-MSCs. (D) Immunohistochemistry images showing the differential expression levels of (a) PRL, (b) IGFBP1, and (c) SCARA5 in decidual cells between normal and PAS patients. Values are expressed as mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001. PRL: Prolactin; IGFBP1: Insulin-like growth factor binding protein 1; SCARA5: Scavenger receptor class A type 5.

FIGURE 3

PA-MSCs and MD-MSCs have similar effects on trophoblast invasion through an MD-MSC-based co-culture system. (A) Schematic diagram of the co-culture system and timeline for the assessment of trophoblast invasion. (B) The effects of MD-MSCs and PA-MSCs under different decidualization treatments were compared to assess their impact on trophoblast invasion. Values are expressed as mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001, n.s.: not significant. (C) (a) The heatmap shows the potential molecules that are involved in the dynamic regulation of trophoblast invasion, annotated through the GEO database. Genes that negatively regulate trophoblast invasion were marked in red. Values are expressed as z-scores, which have been normalized to the gene expression levels. (b) Predicted networks of protein-protein interaction among annotated proteins through STRING analysis. The line between two nodes typically denotes the interaction between those two proteins, with the line’s colors reflecting the available resources of interactions.

FIGURE 4

MD-MSCs can potentially convert dNK-like cells from pNK cells. (A) A schematic diagram and timeline depict the assessment of immune suppression in dNK-like cells using a co-culture system. (B) Evaluation of cytotoxicity involves three different MD-MSC-induced dNK-like cells interacting with trophoblasts. (C) (a) Flow cytometry analysis and (b) quantitative statistics represent surface KIR2DL4 expression levels on different subtypes of NK cells after co-culture with trophoblasts for 12 h. HEK293T serves as the positive control for KIR2DL4. Values are normalized to the IgG control and expressed as mean ± SEM. Significance levels are denoted as *p < 0.05, **p < 0.01, and ***p < 0.001, with all comparisons made against the pNK cell group. (D) (a) The heatmap illustrates three potential secreted proteins encoding genes that regulate the conversion of pNK cells into dNK-like cells. Values are expressed as z-scores, which have been normalized to the gene expression levels. (b) Predicted networks of protein-protein interaction among annotated proteins through STRING analysis. The line between two nodes typically represents the interaction between those two proteins, with the colors of the line reflecting the available resources for interactions. pNK cell: peripheral NK cells; dNK1 cells: MD-MSC-induced dNK-like cells; dNK2 cells: E2/P4-treated De-MD-MSC-induced dNK-like cells; dNK3 cells: cAMP/MPA-treated De-MD-MSC-induced dNK-like cells.

FIGURE 5

PA-MSCs demonstrate a greater capacity to enhance dNK-like cell proliferation than normal MD-MSCs. (A) A schematic diagram of the co-culture system for the dNK-like cell proliferation assay using MD-MSCs and PA-MSCs. (B) (a) The proliferation of dNK-like cells was measured at specific time points: 1, 4, 7, and 10 days under various conditions, including co-culture with MD-MSCs, PA-MSCs, and their corresponding decidual counterparts. (b) The effects of MD-MSCs, PA-MSCs, and their decidual counterparts on dNK-like cell proliferation were quantified at day 10. Values are expressed as mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001. (C) (a) The heat map illustrates the expression patterns of annotated secreted proteins responsible for stimulating dNK-like cell proliferation. Values are expressed as z-scores, which have been normalized to the gene expression levels. (b) Predicted networks of protein-protein interaction among annotated proteins through STRING analysis. The line between two nodes typically denotes the interaction between those two proteins, with the colors of the line reflecting the available resources of interactions. The thickness of the line is about the strength of the interactions. (D) Immunofluorescence staining of dNK cells in (a) a normal placenta and (b) a PAS placenta. The white-lined sections represent the decidual membrane. dNK cells are stained with CD56 (red) and CD9 (green). pNK cell: peripheral NK cells; dNK1 cells: MD-MSC-induced dNK-like cells; dNK2 cells: E2/P4-treated De-MD-MSC-induced dNK-like cells; dNK3 cells: cAMP/MPA-treated De-MD-MSC-induced dNK-like cells.

FIGURE 6

Enhanced trophoblast invasion was observed under E2/P4 treated PA-induced dNK-like cells with a dose-dependent correlation. (A) A schematic diagram and timeline depict the assessment of trophoblast invasion under MD-MSC or PA-induced dNK2 cell-derived conditioned medium. (B) Dose-dependent correlation in three subtypes of NK cells. Values are expressed as mean ± SEM. Significance levels are denoted as *p < 0.05, **p < 0.01, and ***p < 0.001, with all comparisons made against the pNK cell group. pNK cell: peripheral NK cells; dNK2 cells: E2/P4-treated De-MD-MSC-induced dNK-like cells.

FIGURE 7

Schematic representation of the underlying mechanisms of PAS. The high proliferation of dNK cells induced by defective decidual cells may dysregulate trophoblast invasion, leading to PAS. CTB: cytotrophoblast; SCT: syncytiotrophoblast; ESC: endometrial stromal cell; DC: decidual cell; EVT: extravillous trophoblast; dNK: decidual natural killer cell.