Enhanced prevalence of plasmatic soluble MHC class I chain-related molecule in vascular pregnancy diseases

Abstract

The major histocompatibility complex class I related chain (MIC) is a stress-inducible protein modulating the function of immune natural killer (NK) cells, a major leukocyte subset involved in proper trophoblast invasion and spiral artery remodeling. Aim of the study was to evaluate whether upregulation of soluble MIC (sMIC) may reflect immune disorders associated to vascular pregnancy diseases (VPD). sMIC was more frequently detected in the plasma of women with a diagnostic of VPD (32%) than in normal term-matched pregnancies (1.6%, P < 0.0001), with highest prevalence in intrauterine fetal death (IUDF, 44%) and vascular intrauterine growth restriction (IUGR, 39%). sMIC levels were higher in preeclampsia (PE) than in IUFD (P < 0.01) and vascular IUGR (P < 0.05). sMIC detection was associated with bilateral early diastolic uterine notches (P = 0.037), thrombocytopenia (P = 0.03), and high proteinuria (P = 0.03) in PE and with the vascular etiology of IUGR (P = 0.0038). Incubation of sMIC-positive PE plasma resulted in downregulation of NKG2D expression and NK cell-mediated IFN-γ production in vitro. Our work thus suggests that detection of sMIC molecule in maternal plasma may constitute a hallmark of altered maternal immune functions that contributes to vascular disorders that complicate pregnancy, notably by impairing NK-cell mediated production of IFN-γ, an essential cytokine favoring vascular modeling.

Figure 1

(a) Frequency of bilateral early diastolic uterine notch, in sMIC-positive and -negative subgroups of preeclamptic patients. (b) Frequency of thrombocytopenia in sMIC-positive and -negative subgroups of preeclamptic patients. (c) Median and interquartile range of proteinuria per day in sMIC-positive and -negative preeclampsia subgroups.

Figure 2

Differential effect of plasma isolated from sMIC-positive VPD patients or term-matched sMIC-negative plasma of women undergoing normal pregnancy (NP) on percentage (Figure 2(a)) and mean fluorescence intensity (mfi, Figure 2(b)) of NK cell surface NKG2D expression within PBMC. For each experiment, PBMC isolated from healthy nonpregnant control donors (NP) was cultured for 48 hours before flow cytometry analysis in media containing either 20% sMIC-positive plasma from women experiencing vascular pregnancy diseases (black) or 20% sMIC-negative plasma from normal pregnancies (NP, gray). Gestational age at plasma sampling of VPD cases was matched to that of NP plasma used as control in 8 independent experiments. Plasma-induced modifications of NKG2D cell surface expression are illustrated as variation in the % of NKG2D positive CD3⁻CD56⁺ NK cells found within PBMC (a) and mean fluorescence intensity of NKG2D staining in NKG2D positive CD3⁻CD56⁺ NK cells (b). (c) Flow cytometry plots illustrate detection of NK cells and gating of NKG2D expression within CD3⁻CD56⁺ NK cells in 3 representative experiments. Overlay of NKG2D downregulation resulting from incubation of NK cells with sMIC-positive VPD plasma (black) versus sMIC-negative plasma from control normal undergoing pregnancies (NP Figure 2(c), right panel). Wilcoxon matched pairs tests were used to compare the two groups.

Figure 3

Quantification of INF-γ secretion by the NK-92 NK cell line after 72 hours incubation in media containing 20% plasma from representative sMIC-positive vascular pregnancy diseases (VPD, sMIC⁺) and gestational age-matched sMIC-negative plasma normal pregnancies (NP, sMIC⁻). (a) ELISA detection of INF-γ levels in supernatants of NK-92 cells cultured in 20% plasma. (b) Real-time PCR transcriptional analysis of INF-γ mRNA levels in NK-92 cells. mRNA transcript levels are represented as relative copy numbers per 10⁶ GAPDH transcripts. Results are expressed as median and 25–75 interquartile ranges (n = 4).