Marilen Benner

THREE TYPES OF TREG CONTROL PLACENTAL INFLAMMATION 61 3 EVTs and decidual macrophages increase Treg proportions during co-culture Previous studies demonstrated that EVTs directly increased FOXP3 levels in CD4 + CD25 HI Tregs during co-culture (5, 6) and that decidual macrophages favored Treg differentiation (39). To investigate the capacity of EVTs and decidual macrophages to increase the three Treg types, EVTs and decidual macrophages isolated from first-trimester placental tissue were co- cultured with naive CD4 + pTs for 3 days as described previously (6). Co-culture of EVTs or decidual macrophages with CD4 + pTs significantly increased the proportion of FOXP3 + and HELIOS + Tregs (Figure 5A). EVTs, but not decidual macrophages, also increased the proportion of PD1 HI Tregs, while neither EVTs nor decidual macrophages changed the TIGIT + population (Figure 5A). To deter mine whether EVTs and decidual macrophages increased Treg proportions through cell-cell contact or by secretion of soluble factors, EVTs or decidual macrophages and CD4 + pTs were co-cultured in a transwell system. Separation of EVTs or decidual macrophages from CD4 + pTs by a transwell membrane resulted in a small but not significant decrease in the induction of FOXP3 + and HELIOS + cells by EVTs and decidual macrophages (Figures 5B and 5C), suggesting that both cell-cell contact and soluble factors may play a role in the induction of FOXP3 + cells by EVTs and decidual macrophages. In contrast, when EVTs and CD4 + pTs were separated by a transwell mem brane, the increase in the proportion of PD1 HI cells was abrogated, demonstrating that cell-cell contact is required here (Figure 5B). To further investigate the mechanisms by which EVTs and decidual macrophages increase Treg proportions, additional cell cultures were established where EVTs or decidual macrophages were co-incubated with CD4 + pTs in the presence of IgG controls and blocking antibodies for a panel of co-inhibitory molecules and anti-inflammatory cytokines. The increase in FOXP3 + cells was not reversed by addition of any of the blocking antibodies tested in EVT co-cultures (TCR coreceptor CD3; HLA-C; HLA-G receptor ILT2; HLA-G; coinhibitory molecule PDL1; and the immune modulatory cytokine TGF β ) (Figure 5D) and decidual macrophage co-cultures (ILT2, HLA-DR, PDL1, TGF β , IL-10 receptor (IL-10R)) (Figure 5E). Interestingly, addition of HLA-C and CD3 antibodies signifi cantly inhibited the induction of PD1 HI cells by EVTs (Figure 5F), while blocking HLA-G interactions resulted in a small but not significant reduction. Blocking ILT2, PDL1, and TGF β in the EVTs and CD4 + T cell co-cultures did not significantly affect induction of PD1 HI cells by EVTs (Figure 5F). The question of whether placental viral infections alter Treg induction or Treg stability and thereby exacerbate placental inflammation is clinically important. EVTs were infected with human cytomegaloviruses (HCMVs), the most common pathogen to infect the placenta and a major cause of congenital disease (40). Interestingly, no differences were observed in the capacity of healthy or HCMV-infected EVTs to increase CD25 HI FOXP3 + and PD1 HI Treg proportions (Figure 5G). Collectively, these results suggest that Tregs can locally be induced by EVTs and decidual macrophages and that antigen-specificity may be involved.