Multi-system dysregulation in placental malaria contributes to adverse perinatal outcomes in mice

Abstract

Sequestration of Plasmodium parasites in the placental vasculature contributes to increased morbidity and mortality in pregnant compared to non-pregnant patients in malaria-endemic regions. In this study, outbred pregnant CD1 mice with semi-allogeneic fetuses were infected with transgenic Plasmodium berghei or mock inoculated by mosquito bite at either embryonic day (E)6 (first trimester-equivalent) or 10 (second trimester-equivalent) and were compared to non-pregnant females. P. berghei-infected mosquitoes had greater biting avidity for E10 dams than uninfected mosquitoes, which was not apparent for E6 dams nor non-pregnant females. Infected E10 dams had greater numbers of parasites than E6 dams in the uterus and spleen, but not in the blood or liver. While parasites were found in placentas, no parasites were present in fetuses. Maternal infection at E6 caused greater maternal morbidity, with greater rates of fetal reabsorption and stillbirths than at E10. Infection at E10 caused adverse offspring outcomes, including growth restriction. To identify possible mechanisms of adverse offspring outcomes, E10 dams were euthanized during peak parasitemia (8 days postinfection [dpi]), and outcomes were compared to mock-infected dams. P. berghei caused significant systemic maternal immune activation with elevated circulating lymphocytes, eosinophils, and neutrophils and splenic cytokine concentrations. P. berghei infection at E10 increased corticosterone and decreased progesterone concentrations, which could contribute to adverse perinatal outcomes through immunomodulation. There were limited changes in the maternal fecal microbiome after P. berghei infection. Mosquito bite infection of outbred dams with P. berghei causes placental malaria and provides a novel, tractable model to investigate therapeutic treatments.

Keywords: Plasmodium; corticosterone; cytokines; microbiome; pregnancy; progesterone.

Fig 1

Mosquito bite Plasmodium berghei challenge results in infection within the placenta. (A) CD-1 dams were infected on embryonic day (E)6 or E10 by mosquitos infected with P. berghei (Pb) GFP-luc and were monitored until they succumbed to infection or delivered. A subset of dams was infected on E6 or E10 and was followed for 8 dpi, after which they were euthanized, and tissues were collected. (B) Blood-feeding avidity of mock- and Pb-infected female mosquitoes during blood meals for non-pregnant (NP) females (triangles) and pregnant dams infected at E6 (open circles) or E10 (closed circles). n = 3–18 dams/group across six experiments. (C) Parasitemia was measured by blood smear with Giemsa stain of non-pregnant or pregnant dams infected at E6 or E10. n = 12–18 dams/group across six experiments. (D) Representative luminescent images of the spleen, liver, placenta, uterus, and fetus (differentiated by the outline color) from dams infected at E6 or E10, captured by the Intra-Vital Imaging System (IVIS) at 8 dpi. (E) Tissue parasite burden quantified by luciferase radiance from the IVIS in tissues collected at 8 dpi. n = 3 dams/group. (F) Representative histology of Giemsa-stained placental sections from E6 and E10 dams collected at 8 dpi, labyrinth zone imaged at 100× magnification. Infected red blood cells (iRBCs) are indicated by arrows. Dotted outlines indicate maternal blood spaces (MBS), and solid outlines indicate fetal blood spaces (FBS). Data were analyzed using a two-way ANOVA with Tukey post hoc test. Asterisks indicate P < 0.05.

Fig 2

P. berghei infection during pregnancy causes maternal and perinatal morbidity and mortality in a gestational age-dependent manner. (A) Body mass change was calculated for dams infected at either embryonic day (E)6 or E10 across embryonic days. Black stippled lines indicate the day of P. berghei (Pb) infection at E6 or E10 or the average day of delivery. n = 3–18 dams/group across six experiments. (B) Kaplan-Meier survival curves for E6 and E10 Pb-infected dams across embryonic days. n = 7–12 dams/group across six experiments. (C) The percentage of dams with reabsorption was calculated as the percentage of pups that survived until E19. Data were analyzed across two experiments. (D) The percentage of stillborn pups was calculated by dividing the number of stillborn pups by the total litter size. Data were analyzed across two experiments. (E) Embryonic day of delivery of mock- or Pb-infected E10 dams. (F) Mass of fetuses from mock- or Pb-infected E10 dams collected by cesarean section at 8 days postinfection. Statistical analysis was performed using an unpaired t-test. n = 5–11 pups/group across two experiments. Asterisks indicate P < 0.05. (G) Representative histology of H&E-stained placental sections from E10 dams collected at 8 dpi. Labyrinth zone imaged at 2× magnification (scale bar, 500 µm) and 40× magnification (scale bar, 30 µm).

Fig 3

Malaria infection during pregnancy induces systemic inflammation and placental chemokine activation. (A) Lymphocyte, (B) eosinophil, (C) neutrophil, and (D) neutrophil/lymphocyte concentrations were analyzed from complete blood counts from E10 mock- and Pb-infected dams at 8 dpi. Statistical analysis was performed using an unpaired t-test. n = 5–6 per group across two experiments. A heatmap illustrating the log fold change in splenic (E) and placental (F) cytokine concentrations of Pb-infected E10 dams relative to the mock controls at 8 dpi. n = 5–9 per group across two experiments. (G) Corticosterone and (H) progesterone concentrations were measured in plasma from E10 dams at 8 days postinfection by ELISA. n = 8–16 dams/group across four experiments. Asterisks indicate P < 0.05.

Fig 4

P. berghei (Pb) infection during pregnancy causes minimal changes in the maternal fecal microbiome during peak parasitemia. (A) Alpha diversity measures of the observed number of species, or richness (left) and Shannon diversity (right), were calculated and compared between non-pregnant females and embryonic day (E)10 mock- and Pb-infected dams at 8 dpi. Pairwise comparisons were conducted using the Mann-Whitney test. (B) PCoA plot of weighted Bray-Curtis dissimilarity distance. Ellipses were constructed using the standard deviation of non-pregnant (NP) females (dashed line) versus pregnant dams (solid line). (C) Relative abundances agglomerated at the phylum level, separated by infection and pregnancy status. Comparison of the relative abundance of (D) Firmicutes and (E) Bacteroidota between NP females and E10 dams. Pairwise comparisons were conducted using the Mann-Whitney test. Spearman correlation analysis was performed between (F) Firmicutes and splenic G-CSF, (G) Mucispirillum schaedleri and placental IL-4, and (H) Lachnospiraceae and pup mass. A simple linear regression analysis was performed and is displayed. For all graphs, the color designates infection status (mock: gray; Pb: orange), and pregnancy status is categorized by the shape (non-pregnant females: triangles; E10-infected dams: circles). n = 5 dams/group across two experiments. Asterisks indicate P < 0.05.