Preeclampsia is a pregnancy-induced hypertensive complication closely linked to the immune system. It severely impacts maternal and infant health and has an incidence rate of approximately 2-5%. It usually occurs after 20 weeks of pregnancy, so early screening and risk prediction are crucial for early intervention and protecting the health of mother and child. However, due to the incomplete understanding of the disease, there has been a lack of effective early screening methods.

Recently, BGI-Research collaborated in research into preeclampsia with Shenzhen Maternity & Child Healthcare Hospital, with their findings published in the journal Communications Biology. By using BGI's independently developed single-cell analysis technology, they constructed a peripheral immune cell atlas of patients with preeclampsia. The study compared the immune cells of healthy pregnant women with those of preeclampsia patients, and revealed the abnormal molecular characteristics of the patients' peripheral immune cells. Based on these findings, the research team was able to develop a new artificial intelligence algorithm for early screening of the disease.

The research, “Characterizing immune variation and diagnostic indicators of preeclampsia by single-cell RNA sequencing and machine learning,” was published in Communications Biology.

By using the atlas of peripheral immune cells constructed for patients in the early stages of preeclampsia, the research team were able to conduct a comprehensive comparison of the immune cell gene expression, functional pathways, cell differentiation, and interactions between these patients and healthy pregnant women.

The research revealed that among patients with preeclampsia, there is an increased proportion of regulatory T-cells. These cells help maintain immune tolerance by suppressing overactive immune responses, thereby preventing autoimmune diseases and playing a role in modulating lymphocyte-mediated immunity pathways. Additionally, the research revealed a decrease in natural killer cells, a type of cytotoxic lymphocyte critical to the innate immune system, which plays a major role in the host-rejection of both tumors and virally infected cells.

Meanwhile, the research team discovered two monocyte subsets potentially associated with the pathogenesis of preeclampsia. These subsets exhibit increased expression levels in pathways related to complement activation, immune response, angiogenesis, and coagulation. Additionally, the pro-inflammatory S100 family genes were found to be upregulated in these monocyte subsets. These findings suggest that these subgroups of monocytes may play a significant role in the inflammatory response associated with preeclampsia. This is crucial for understanding the pathological mechanisms of preeclampsia and has significant implications for potential future treatment approaches.

Schematic outline of the study design.

Furthermore, based on relevant differential characteristics, the research team utilized four cell-type-specific machine learning models. These models were constructed to distinguish preeclampsia from normal pregnancy and to provide insights into potential biomarkers for diagnosis at the single-cell level. 

These efforts collectively deepen our understanding of preeclampsia. Read the research article: https://www.nature.com/articles/s42003-023-05669-2#code-availability