CD8 and CD4 Positive NKT Subpopulations and Immune-Checkpoint Pathways in Early-Onset Preeclampsia and Healthy Pregnancy

Abstract

Although many studies have investigated the clinical aspect of early-onset preeclampsia, our knowledge about the immunological consequences of improper placenta development is scarce. The maternal immunotolerance against the fetus is greatly influenced by the Th1 predominance developed by the mother's immune system. Thirty-two early-onset preeclamptic and fifty-one healthy pregnant women with appropriately matched gestational age were involved in our study. Mononuclear cells were separated from peripheral venous blood and the frequency of CD8⁺, CD4⁺, double positive (DP), and double negative (DN) NKT cell subpopulations was determined using multicolor flow cytometry. Following the characterization, the expression levels of different immune checkpoint receptors and ligands were also defined. Soluble CD226 levels were quantified by ELISA. Novel and significant differences were revealed among the ratios of the investigated NKT subsets and in the expression patterns of PD-1, LAG-3, TIGIT and CD226 receptors. Further differences were determined in the expression of CD112, PD-1, LAG-3 and CD226 MFI values between the early-onset preeclamptic and the healthy pregnant groups. Our results suggest that the investigated NKT subpopulations act differently in the altered immune condition characteristic of early-onset preeclampsia and indicate that the different subsets may contribute to the compensation or maintenance of Th1 predominance.

Keywords: LAG-3; NKT; PD-1; TIGIT; immune checkpoint; preeclampsia.

Figure 1

Differentiation of the NKT subpopulations using flow cytometric analyses. Gating strategy for flow cytometry analysis. The selection method to gate the four investigated NKT cell subpopulations.

Figure 2

The frequency of different NKT subpopulations in EOP and healthy pregnancy. The frequency of the CD8⁺, CD4⁺, DP, DN NKT cells in the lymphogate (A) and the CD3⁺CD56⁺ NKT gate (B) in EOP patients and healthy pregnant women. The solid bars represent medians of 51 and 32 determinations, respectively. The boxes indicate the interquartile ranges, and the whiskers show the most extreme observations. The middle square within the box represents the mean value. Significant differences with p-values < 0.01 *** and <0.03 ** are indicated.

Figure 3

PD-1 and PD-L1 expression by different NKT cell subpopulations in EOP and healthy pregnancy. PD-1 receptor expression (A) and PD-L1 ligand expression (B) by the CD8 and CD4 positive and negative NKT cell subpopulations in EOP patients and healthy pregnant women. The solid bars represent medians of 18 and 17 determinations, respectively. The boxes indicate the interquartile ranges, and the whiskers show the most extreme observations. The middle square within the box represents the mean value. Significant differences with p-values < 0.01 *** and <0.03 ** are indicated. Representative FACS plots show the PD-1 surface marker (C) and PD-L1 surface molecule (D) expression by cells in the lymphocyte gate. Fluorescent minus one (FMO) control was used to determine PD-1 and PD-L1 expression.

Figure 4

LAG-3 and Gal-3 expression by different NKT cell subpopulations in EOP and healthy pregnancy. LAG-3 receptor (A) and Gal-3 ligand expression (B) by the CD8 and CD4 surface molecule positive and negative NKT cell subpopulations in EOP patients and healthy pregnant women. The solid bars represent medians of 37, 28, 16 and 15 determinations, respectively. The boxes indicate the interquartile ranges, and the whiskers show the most extreme observations. The middle square within the box represents the mean value. Significant differences with p-values < 0.01 *** and <0.03 ** are indicated. Representative FACS plots show the expression of LAG-3 (C) and Gal-3 molecule (D) by cells in the lymphocyte gate. FMO control was used to determine LAG-3 and Gal-3 positivity.

Figure 5

TIGIT and CD226 receptor expression by different NKT cell subpopulations in EOP and healthy pregnancy. TIGIT (A) and CD226 receptor expression (B) by the CD8 and CD4 surface molecule positive and negative NKT cell subpopulations in EOP patients and healthy pregnant women. The solid bars represent medians of 39, 40, 39 and 28 determinations, respectively. The boxes indicate the interquartile ranges, and the whiskers show the most extreme observations. The middle square within the box represents the mean value. Significant differences with p-values < 0.01 *** and <0.03 ** are indicated. Representative FACS plots show the TIGIT (C) and CD226 surface molecule (D) expression by cells in the lymphocyte gate. FMO control was used to determine TIGIT and CD226 expression.

Figure 6

CD112 and CD155 ligand expression by different NKT cell subpopulations in EOP and healthy pregnancy. CD112 (A) and CD155 ligand expression (B) by the CD8 and CD4 surface molecule positive and negative NKT cell subpopulations in EOP patients and healthy pregnant women. The solid bars represent medians of 16, 16, 17 and 17 determinations, respectively. The boxes indicate the interquartile ranges, and the whiskers show the most extreme observations. The middle square within the box represents the mean value. Significant differences with p-values < 0.01 ***, <0.03 ** and <0.05 * are indicated. Representative FACS plots show the CD112 surface marker (C) and CD155 surface molecule (D) expression by cells in the lymphocyte gate. FMO control was used to determine CD112 and CD155 expression.

Figure 7

Relative TIGIT and CD226 expression by different NKT cell subpopulations in EOP and healthy pregnancy. Mean fluorescent intensity of the TIGIT (A) and the CD226 (B) receptors by the CD8 and CD4 surface molecule positive and negative NKT cell subpopulations in EOP patients and healthy pregnant women. The solid bars represent medians of 39, 40, 39 and 28 determinations, respectively. The boxes indicate the interquartile ranges, and the whiskers show the most extreme observations. The middle square within the box represents the mean value. Significant differences with p-values < 0.01 *** and <0.03 ** are indicated.

Figure 8

Relationship between the surface expression or the relative expression of CD226 and the level of soluble CD226 in NKT subpopulations in EOP and healthy pregnancy. The serum concentration of CD226 (A) molecule in EOP patients and healthy control women. The solid bars represent medians, the boxes indicate the interquartile ranges and the lines show the most extreme observations. Linear regression analyses between the CD226 surface (B) or relative expression (C) with the sCD226 level in different NKT subsets in EOP and healthy control women. p values and coefficients of determination (R²) were calculated in R.