Marilen Benner

CHAPTER 8 224 dynamics; induction of clusters might occur at specific time points during gestation, which may vary between individuals. Although the presence of uterine B cells as being a part of healthy pregnancy has been disputed in the past, several studies now confirm their presence and they clearly warrant additional investigation (41-44). Where do these B cells come from? Circulatory B cells are decreased from mid-gestation to delivery, which might result from a gestational reduction in lymphopoiesis, as shown in mice (45- 50). This might not be the only explanation for reduced systemic B cells; evidence supporting a relocation of B cells to the uterus increases. Placental explants are able to attract B cells through CCL20 (placenta) - CCR6 (B cell) interaction (51). Also, data on decidua support the notion that B cells are retained within the uterus during gestation. Matching our observation that naïve B cells can be isolated from all decidual samples, an increased proportions of mature/naïve B cells were found within the intervillous space (51). However, probably not all of the observed decidual B cells are the result of pregnancy-induced homing to the uterus as menstrual blood samples also contain B cells. This hints towards a tissue-resident population independent of implantation. Furthermore, we observed that B cells of the decidua contain subsets with phenotypes distinct from peripheral blood. Either the adaptations to the uterine environment itself, or pregnancy induced interaction with B cells causes differential marker expression of local B cells. As such, primary human samples of the fetal-maternal interface cannot be replaced by investigation of systemic cells. A QUESTION OF TIMING Origin of immune cells, as well as gestational duration, strongly impacts immune phenotype and function. In our efforts to understand the source of pregnancy complications, we need to consider the characteristics of the immune cells at the respective time point involved in the physiological adaptation needed for a healthy continuation of pregnancy. Generating an overview of the immunological fingerprints associated with the different gestational time points, we assessed general lymphocyte composition from menstrual blood, to first and second trimester decidua, and up to decidua parietalis. Unsupervised-clustering offers analysis of flow cytometry data without the need for manual gating which depends on arbitrary, biased selection of cell populations to assign a classification. This confirmed that CD4 + T cells and NK cells are the most stratifying features of uterine lymphocytes allowing for assignment of an immune- profile to its respective gestational time point (Chapter 4) . Both NK and T cells frequencies strongly change upon onset of pregnancy. Of note, in MMC, a relatively large variation in NK and T cells frequencies was observed, and some donors presented with similar lymphocyte levels as observed in first trimester decidua samples. We can only speculate whether this lack of difference compared to the gestational profile is due to natural variation, or to an underlying