Macrophage Migration Inhibitory Factor (MIF) Prevents Maternal Death, but Contributes to Poor Fetal Outcome During Congenital Toxoplasmosis

Figures

• 

  FIGURE 1. (A) Change in body weight in C57BL/6 MIF-/- and C57BL/6 WT mice orally infected with 10 cysts of strain ME49 of Toxoplasma gondii. Animals were monitored for 21 days, and the results are presented as the mean of data obtained from 5 animals in each group. Asterisks indicate significant differences between MIF-/- and WT mice (Mann–Whitney test; ∗P < 0.05; ∗∗P < 0.01). (B) Kaplan–Meier survival curve of infected/pregnant mice. MIF-/- and WT females (n = 5 animals per group) were orally infected with five cysts of T. gondii (ME49 strain) on the first day of pregnancy and monitored for 80 days after infection. Data were analyzed with a log-rank test. Differences were considered significant if P < 0.05 (∗).

• 

  FIGURE 2. Histopathological analysis of implantation sites from pregnant C57BL/6 MIF-/- and C57BL/6 WT mice infected or not by Toxoplasma gondii. Paraffin-embedded 4-μm tissue sections were stained with hematoxylin-eosin. Pregnant/non-infected WT (A) and MIF-/- (B) controls; pregnant/infected WT (C,E) and MIF-/- (D,F) females. E, embryo; D, decidual zone; M, myometrium; N, decidual cells in necrosis; H, hemorrhagic areas. Bar scale = 100 μM. The inserts in the figures show increased magnification of the selected area (original magnification in the insert is 10x).

• 

  FIGURE 3. Cytometric bead assay (CBA) for cytokine measurements in serum samples of pregnant and non-pregnant C57BL/6 MIF-/- and C57BL/6 WT mice orally infected by Toxoplasma gondii. Serum samples were collected within 8 days of infection/pregnancy. Non-infected pregnant and non-pregnant mice were used as controls. Data indicate levels of IFN-γ (A), TNF-α (B), IL-6 (C), IL-4 (D), IL-2 (E), IL-10 (F), and IL-17 (G), presented as mean and SEM from serum samples of five animals per group and are expressed in pg/mL. #Significant differences between MIF-/- and WT mice in the same experimental condition (#P < 0.05, ##P < 0.01, ###P < 0.001). ∗Comparisons between multiple experimental conditions within the same group (NP/NI, non-pregnant/non-infected; NP/I, non-pregnant/infected; P/NI, pregnant/non-infected; P/I, pregnant/infected) are indicated by bars (∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001). Data were compared by the Kruskal–Wallis test and Dunn multiple comparison post-test.

• 

  FIGURE 4. Cytometric bead assay (CBA) for cytokine measurements in uterus tissues from pregnant and non-pregnant C57BL/6 MIF-/- and C57BL/6 WT mice orally infected with Toxoplasma gondii. Uterus tissue samples were collected within 8 days of infection/pregnancy. Uterus tissues from non-infected pregnant and non-pregnant mice were used as controls (n = 5 animals per group). The uterus tissue homogenates were prepared in a concentration of 200 mg/mL. Data are expressed as mean and SEM of the amount of cytokines (pg) present in 5 mg of tissue (pg/5 mg of tissue). (A) IFN-γ, (B) TNF-α, (C) IL-6, (D) IL-4, (E) IL-2, (F) IL-10, and (G) IL-17. #Significant differences between MIF-/- and WT mice in the same experimental condition (#P < 0.05). ∗Comparisons between multiple experimental conditions within the same group (NP/NI, non-pregnant/non-infected; NP/I, non-pregnant/infected; P/NI, pregnant/non-infected; P/I, pregnant/infected) are indicated by bars (∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001). Data were compared by the Kruskal–Wallis test and Dunn’s multiple comparison post-test.

• 

  FIGURE 5. Indoleamine 2,3-dioxigenase (IDO) expression in uterus homogenates from pregnant and non-pregnant C57BL/6 MIF-/- and C57BL/6 WT mice orally infected with Toxoplasma gondii. Uterus tissue samples were collected within 8 days of infection/pregnancy. Uterus tissues from non-infected pregnant and non-pregnant mice were used as controls. (A) IDO densitometric analyses are expressed as the relative density of the ratio between IDO and β-actin bands. Data are presented as mean and SEM of uterus samples (n = 5 animals per group). #Significant differences between MIF-/- and WT mice in the same experimental condition (### P < 0.001). ∗Comparisons between multiple experimental conditions within the same group (NP/NI, non-pregnant/non-infected; NP/I, non-pregnant/infected; P/NI, pregnant/non-infected; P/I, pregnant/infected) are indicated by bars (∗P < 0.05). Data were compared by the Kruskal–Wallis test and Dunns’ multiple comparison post-test. (B) Representative cropped blots for IDO (43 kDa) and β-actin (42 kDa). All sample in the same gel were run simultaneously. Blots were probed with mouse anti-IDO and then stripped and reprobed with mouse anti β-actin, as a load control.

• 

  FIGURE 6. (A) Mast cell (MC) quantification in uterus/implantation sites from non-pregnant and pregnant females of C57BL/6 MIF-/- and C57BL/6 WT mice infected or not by Toxoplasma gondii (n = 5 animals per group). Paraffin-embedded 4-μm tissue sections were stained with toluidine blue for MC quantification. MCs were counted in 25-microscopic fields per tissue section, and each animal was analyzed in two tissue sections. #Significant differences between MIF-/- and WT mice in the same experimental condition (#P < 0.05). ∗Comparisons between multiple experimental conditions within the same group (NP/NI, non-pregnant/non-infected; NP/I, non-pregnant/infected; P/NI, pregnant/non-infected; P/I, pregnant/infected) are indicated by bars (∗P < 0.05). Data were compared by the Kruskal–Wallis test and Dunn’s multiple comparison post-test. (B–E) Representative images from uterus tissues of pregnant/non-infected WT (B) and MIF-/- (C) females; and pregnant/infected WT (D) and MIF-/- (E) females. Arrows indicate granulated and degranulated MCs. Bar scale = 50 μM. Correlation between IDO expression and total MCs counted in the uterus of pregnant (F) and non-pregnant (G) C57BL/6 MIF-/- and WT females.

• 

  FIGURE 7. (A) ELISA for measurement of MIF in uterus homogenates of pregnant and non-pregnant C57BL/6 WT mice orally infected by Toxoplasma gondii (n = 5 animals per group). Uterus tissue samples were collected within 8 days of infection/pregnancy. Uterus tissues from non-infected pregnant and non-pregnant mice were used as controls. The homogenates were prepared in a concentration of 200 mg/mL. Data are expressed as the amount of cytokines (pg) present in 5 mg of tissue (pg/5 mg of tissue). NP/NI, non-pregnant/non-infected; NP/I, non-pregnant/infected; P/NI, pregnant/non-infected; P/I, pregnant/infected. (B) Relative density of CD74 expression (full-length isoforms) in the uterus homogenates from pregnant and non-pregnant MIF-/- and WT females infected or not by T. gondii. Densitometric analyses are expressed as relative density of the ratio between CD74 and β-actin bands. Data are presented as mean and SEM of uterus samples from five animals per group. #Significant differences between MIF-/- and WT in the same experimental condition (##P < 0.01). ∗Comparisons between multiple experimental conditions within the same group (NP/NI, non-pregnant/non-infected; NP/I, non-pregnant/infected;…

• 

  FIGURE 8. Parasite burden in the (A) uterus, (B) liver, and (C) spleen from non-pregnant and pregnant WT and MIF-/- mice quantified by real-time PCR. Females were orally infected with five cysts of Toxoplasma gondii and euthanized with 8 days of infection/pregnancy. Data are expressed as mean and SEM values obtained from five animals per group. Data were compared by the Kruskal–Wallis test and Dunn’s multiple comparison post-test. #Significant differences between MIF-/- and WT in the same experimental condition (# P < 0.01). ∗Comparisons between multiple experimental conditions within the same group (NP/I, non-pregnant/infected; P/I, pregnant/infected) are indicated by bars (∗P < 0.05).

• 

  FIGURE 9. Proposed scheme for the immune response mechanism to congenital toxoplasmosis in C57BL/6 MIF-/- females. C57BL/6 MIF-/- pregnant females (8 dop) infected by T. gondii presented upregulation in MIF receptor (CD74), as a possible compensatory mechanism for MIF absence. Additionally, these mice presented increased MC recruitment at the maternal fetal interface that can be associated with increased IDO expression at this interface. IDO upregulation is probably associated with increased IL-4 expression in these tissues. Likewise, C57BL/6 MIF-/- females presented a decrease in systemic proinflammatory cytokines. Altogether, these data clarify that the effects on pregnancy outcome are less harmful in C57BL/6 MIF-/- females but still indicate T. gondii proliferation and early death in these animals.

• 

  FIGURE 10. Experimental design: (A) Non-pregnant mice from both genetic backgrounds (WT and MIF-/-) were infected with 10 cysts of the ME-49 strain and were followed to verify the change in the body weight during 21 days. (B) Pregnant mice from both genetic backgrounds were infected with five cysts of the ME-49 strain on the day of detection of vaginal plug (vp), that is the first day of pregnancy (1st dop). Mice were followed during 80 days to survival analyses. (C) Pregnant and non-pregnant mice from both genetic backgrounds were infected with five cysts of the ME-49 strain on 1st dop. Animals were euthanized on day 8 post-infection. Blood sample was collected for cytokines analyzes. Liver, spleen and uterus were processed for parasitism analyses by PCR. Uteri have also been anatomically analyzed for evaluation of implantation sites (presence of reabsorption, hemorrhage and necrosis). Alternatively, uteri were used for cytokine measurements, protein analyzes by western blotting (IDO and CD74), histopathological evaluation and mast cell quantification.