Fetal and Maternal Innate Immunity Receptors Have Opposing Effects on the Severity of Experimental Malaria in Pregnancy: Beneficial Roles for Fetus-Derived Toll-Like Receptor 4 and Type I Interferon Receptor 1

Abstract

Malaria in pregnancy (MiP) is a distinctive clinical form of Plasmodium infection and is a cause of placental insufficiency leading to poor pregnancy outcomes. Maternal innate immunity responses play a decisive role in the development of placental inflammation, but the action of fetus-derived factors in MiP outcomes has been overlooked. We investigated the role of the Tlr4 and Ifnar1 genes, taking advantage of heterogenic mating strategies to dissect the effects mediated by maternally and fetally derived Toll-like receptor 4 (TLR4) or type I interferon receptor 1 (IFNAR1). Using a mouse infection system displaying severe MiP outcomes, we found that the expressions of TLR4 and IFNAR1 in the maternal compartment take part in deleterious MiP outcomes, but their fetal counterparts patently counteract these effects. We uncovered that fetal TLR4 contributes to the in vitro uptake of infected erythrocytes by trophoblasts and to the innate immune response in the placenta, offering robust protection of fetus viability, but had no sensible impact on the placental parasite burden. In contrast, we observed that the expression of IFNAR1 in the fetal compartment was associated with a reduced placental parasite burden but had little beneficial effect on fetus outcomes. Furthermore, the downregulation of Ifnar1 expression in infected placentas and in trophoblasts exposed to infected erythrocytes indicated that the interferon-IFNAR1 pathway is involved in the trophoblast response to infection. This work unravels that maternal and fetal counterparts of innate immune pathways drive opposing responses in murine placental malaria and implicates the activation of innate receptors in fetal trophoblast cells in the control of placental infection and in the protection of the fetus.

Keywords: IFNAR1; TLR4; fetal innate immunity; pregnancy malaria.

FIG 1

Parasitemia progression and survival in nonpregnant Tlr4^−/− and Ifnar1^−/− females. Adult nonpregnant females were infected i.p. with 10⁶ P. berghei NK65-infected erythrocytes, and peripheral blood parasitemia in the initial phase of infection was determined by FACS analysis using DRAQ5-labeled blood samples at the indicated days postinfection. Survival curves (A) and time course of parasitemia in nonpregnant Tlr4^−/− (B) and Ifnar1^−/− (C) females were compared with those of the wild type. Comparison of survival curves (A) using the log₁₀ rank (Mantel-Cox) test showed nonsignificant differences (P > 0.05). For peripheral blood parasitemia of Tlr4^−/− and Ifnar1^−/− mice (B and C), no significant differences were detected in comparisons against the wild type by using a Kruskal-Wallis test with Dunn's correction for multiple comparisons.

FIG 2

Effects of maternal versus fetal TLR4 on outcomes of malaria in pregnancy. Pregnant females of the indicated Tlr4 maternal/fetal genotype combinations were infected i.v. at G13 with 10⁶ IEs. (A) Maternal parasitemia at G16 to G18 was determined by FACS analysis using DRAQ5-labeled samples. No significant differences were detected between genotype combinations by using a Kruskal-Wallis test with Dunn's correction for multiple comparisons. Data for each genotype group are presented as individual values and means. (B) Stillbirth incidence evaluated at G18. Data are presented as percentages of females showing an abnormal stillbirth rate (shadowed area) for each genotype combination. Abnormal stillbirth in individual females was determined when the stillbirth rate was above the maximum rate observed for the uninfected wild-type controls (stillbirth rates for individual females are depicted in Fig. S2 in the supplemental material). Differences in abnormal stillbirth incidences were analyzed by χ² Fisher's exact test (*, P < 0.05; ***, P < 0.001). (C) Viable fetal weight is represented as a box and whisker plot, with whiskers extending to the minimum and maximum values obtained in each group. A linear mixed-effects model incorporating fetal genotype and maternal infection status was applied as described in Materials and Methods, showing that maternal infection status plays a significant role in the reduction of the weight of viable fetuses (P < 0.001), whereas fetal genotype does not (P = 0.65).

FIG 3

Fetal TLR4, placental infection, and trophoblast-infected erythrocyte interactions. (A) Pregnant females of the indicated maternal/fetal genotype combinations were infected, and placentas were collected at G18. The placental parasite burden was evaluated by quantification of P. berghei 18S rRNA levels in individual placentas by quantitative real-time PCR. ΔC_T was calculated by subtracting the C_T value of the target gene from that of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene. No statistical differences were detected by using a Kruskal-Wallis test with Dunn's correction for multiple comparisons. (B) Relative quantification of Tlr4 mRNA expression levels in placentas from noninfected (n = 6) and infected (n = 6) wild-type females by quantitative real-time PCR (*, P < 0.05). (C) Trophoblast primary cultures from wild-type and Tlr4^−/− placentas were prepared as described in Materials and Methods and exposed or not exposed to IEs (ratio, 1:1) during 4 or 6 h. P. berghei 18S rRNA levels in individual cultures were quantified by quantitative real-time PCR and are represented as fold increases relative to the values for wild-type cultures (**, P < 0.01). (D) Expression profiling of selected inflammation-related genes in infected placentas from wild-type (Tlr4^+/+), TLR4 KO (Tlr4^−/−), or heterogenic (Tlr4^+/−) matings. RNA pools of 10 placentas of each genotype class were analyzed by using the TaqMan Array Mouse Immune panel and quantified by quantitative real-time PCR. Quantification results are relative to those of wild-type uninfected placentas and represent the means of data from two independent experiments.

FIG 4

Maternal versus fetal IFNAR1 effects on outcomes of malaria in pregnancy. Maternal P. berghei NK65 parasitemia at G16 to G18 (A) and abnormal stillbirth incidences (B) and weights of viable fetuses (C) at G18 were analyzed for the indicated maternal-fetal Ifnar1 genotype combinations as described in the legend of Fig. 2. The presentation of data is analogous to that for Fig. 2 (stillbirth rates for individual females are depicted in Fig. S2 in the supplemental material). Detection of statistical differences between genotype combinations was done by using a Kruskal-Wallis test with Dunn's correction for multiple comparisons (A), χ² Fisher's exact test (B), or a linear mixed-effects model approach incorporating either fetal or maternal genotype as a fixed effect alongside a random effect for each mother (C). Using type III ANOVAs with Satterthwaite approximation for degrees of freedom, we find that fetuses in infected mothers who are Ifnar^−/−, irrespective of the fetal genotype, show a trend toward higher weight than those of their counterparts carried by infected Ifnar^+/+ mothers (P = 0.087) (C). *, P < 0.05; **, P < 0.01.

FIG 5

Fetal IFNAR1 in placental infections and in trophoblasts. (A) The placental parasite burden in individual placentas was estimated at G18 for the indicated maternal/fetal genotype combinations by quantifying P. berghei 18S rRNA levels as described in the legend of Fig. 2. (B) IFNAR1 surface expression in freshly isolated trophoblasts from noninfected placentas was detected by FACS analysis (empty histogram, staining with anti-IFNAR1 antibody; filled histogram, isotype control). (C and D) Relative quantification of Ifnar1 mRNA gene expression levels in infected (n = 6) versus noninfected (n = 6) placentas (C) and in primary trophoblast cultures from wild-type placentas exposed or not exposed to P. berghei NK65-infected erythrocytes during 4 h (D) by quantitative real-time PCR. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.