Pigment Epithelium-Derived Factor, a Novel Decidual Natural Killer Cells-Derived Factor, Protects Decidual Stromal Cells Via Anti-Inflammation and Anti-Apoptosis in Early Pregnancy
Ji Zheng,Yanhong Li,Yifei Sang,Ling Xu,Xueling Jin,Yu Tao,Dajin Li,Meirong Du
DOI: https://doi.org/10.1093/humrep/deaa118
IF: 6.1
2020-01-01
Human Reproduction
Abstract:STUDY QUESTION: What is the role of pigment epithelium-derived factor (PEDF) from decidual natural killer (dNK) cells during early pregnancy? SUMMARY ANSWER: PEDF from dNK cells limits the lipopolysaccharide (LPS)-induced apoptosis and inflammation of decidual stromal cells (DSCs) to maintain DSCs homoeostasis and immune balance at the maternal-foetal interface during early pregnancy. WHAT IS KNOWN ALREADY: dNK cells, which secrete PEDF, play critical roles during pregnancy via a series of key regulators. PEDF, a multifunctional endogenous glycoprotein, exhibits a wide range of biological actions upon angiogenesis, inflammation, metabolic homoeostasis, immunomodulation etc., providing potential clinical applications. STUDY DESIGN, SIZE, DURATION: Natural killer (NK) cells from decidua and peripheral blood as well as DSCs isolated from normal pregnancy (NP) during the first trimester (6-10 weeks) and the matched patients suffering recurrent miscarriage (RM) were studied. RNA-sequencing analysis of dNK cells was performed to screen for potential key genes involved in RM. The expression of PEDF in dNK cells in NP and RM was examined. A coculture system with LPS-stimulated DSCs and NK cell supernatants derived from NP or RM was established to explore the regulatory mechanisms of PEDF at the maternal-foetal interface. PARTICIPANTS/MATERIALS, SETTING, METHODS: Peripheral blood and decidual tissues were obtained from women with NP (n = 61) and RM (n = 21). The expression levels of PEDF in NK cells and its receptor (PEDFR) on DSCs were analysed using flow cytometry, western blot and immunohistochemistry. Purified peripheral natural killer (pNK) cells were cocultured with DSCs or trophoblast cells or a combination of both cell types, and PEDF expression in pNK cells was then examined by flow cytometry. DSCs were treated with LPS, an outer-membrane component of Gram-negative bacteria, thereby mimicking an enhanced inflammatory status within decidua, and were cocultured with dNK cell supernatants from NP or RM. In the coculture system, plasmids expressing short hairpin RNA were used to silence PEDFR on DSCs and block the PEDF/PEDFR interaction. Inflammatory cytokines and apoptosis of DSCs treated as described above were assessed by flow cytometry. Western blotting was performed, and the specific signal pathway inhibitors were used to determine downstream PEDF/PEDFR signalling in early decidua. MAIN RESULTS AND THE ROLE OF CHANCE: Markedly higher RNA (P<0.001) and protein expression of PEDF (P<0.01) was detected in normal dNK cells when compared with pNK cells. Compared with pNK cells cultured alone, PEDF expression in pNK cells was elevated after coculture with DSCs (P<0.01) or trophoblast cells (P<0.001). The increased pro-inflammatory cytokine, tumour necrosis factor-alpha and apoptosis of DSCs following LPS stimulation were suppressed by recombinant human PEDF (P<0.001) or the supernatant of dNK cells derived from NP (P<0.001). However, these effects were somewhat abrogated when the PEDF/PEDFR interaction was blocked with PEDFR short hairpin sRNA (P<0.01). Furthermore, dNK cell-derived PEDF protected DSCs from LPS-induced inflammation via inhibition of nuclear factor kappa-B activation, while also protecting DSCs from LPS-induced apoptosis via promotion of extracellular signal-regulated kinase expression. Compared with NP, both significantly decreased PEDF RNA (P<0.001) and protein expression (P<0.001) in dNK cells, but not in pNK cells (P>0.05), were detected in women with RM. PEDFR on DSCs was also decreased within RM compared with that within NP (P<0.001). As a result, dNK cell-mediated anti-inflammation (P<0.01) and anti-apoptosis (P<0.05) for protection of LPS-treated DSCs was attenuated in patients suffering from RM. LARGE SCALE DATA: N/A. LIMITATIONS, REASONS FOR CAUTION: We cannot exclude the possibility that the differences in amounts of PEDF and its receptor in tissue from NP versus RM women could be caused by the miscarriage event in women with RM. Our experiments only involved human samples investigated in vitro. Experiments in animal models and human study cohorts are still needed to confirm these findings and further clarify the role of PEDF-PEDFR in NP and/or RM. WIDER IMPLICATIONS OF THE FINDINGS: To the best of our knowledge, this is the first study to demonstrate PEDF expression and function at the maternal-foetal interface in the first trimester, providing further evidence that PEDF exhibits functional diversity and has great potential for clinical application(s). The findings of selectively high expression of PEDF in normal dNK cells and the PEDF-mediated role of dNK cells during NP and RM help to further elucidate the immune mechanisms behind RM.