Research ArticleANTIVIRAL IMMUNITY

Type I interferons instigate fetal demise after Zika virus infection

See allHide authors and affiliations

Science Immunology  05 Jan 2018:
Vol. 3, Issue 19, eaao1680
DOI: 10.1126/sciimmunol.aao1680
  • Fig. 1 Fetal IFN signaling leads to severe IUGR and resorption after ZIKV infection.

    Ifnar1−/− females mated to Ifnar1+/− males were infected intravaginally (Ivag) with 1.5 × 105 PFU Cambodian ZIKV on E5.5 or E8.5 and harvested on E17.5. (A) Schematic showing mating strategy. Fetal weights were measured (B), and fetuses were visually inspected (C). RNA was isolated from the placenta or resorbed conceptus (D) or fetus (E) to determine ZIKV levels. Relative ZIKV levels were determined by normalization to Hprt. Individual data points with mean are shown. For fetal weights, ZIKV E5.5 (n = 15 Ifnar1−/− and n = 17 Ifnar1+/− from four litters); ZIKV E8.5 (n = 21 Ifnar1−/− and n = 22 Ifnar1+/− from six litters); and uninfected (n = 13 Ifnar1−/− and n = 21 Ifnar1+/− from four litters). For ZIKV RNA, ZIKV E5.5 (n = 7 Ifnar1−/− and n = 7 Ifnar1+/− from three litters); ZIKV E8.5 (n = 9 to 10 Ifnar1−/− and n = 11 Ifnar1+/− from five litters); and uninfected (n = 11 from three litters with both genotypes pooled). Data are pooled from at least two independent experiments from each infection time point. Scale bars, 1 cm. *P < 0.05 and ** P < 0.01 by Tukey’s multiple comparison test.

  • Fig. 2 ZIKV-infected fetuses with intact IFN signaling are resorbed between E10.5 and E12.5.

    Ifnar1−/− females mated to Ifnar1+/− males were infected intravaginally with 1.5 × 105 PFU Cambodian ZIKV at E5.5 and harvested at indicated time points. (A) Representative images of three to five litters for infected and two to three litters for uninfected are shown per time point. Scale bars, 1 mm. (B) Crown-rump length was measured by tracing distance from crown of head to end of tail using ImageJ. Means with individual points are graphed. Data points shown represent the following: E9.5 uninfected Ifnar1−/− (n = 7) and Ifnar1+/− (n = 12 from two litters) and infected Ifnar1−/− (n = 9) and Ifnar1+/− (n = 10 from three litters); E10.5 uninfected Ifnar1−/− (n = 15) and Ifnar1+/− (n = 5 from three litters) and infected Ifnar1−/− (n = 24) and Ifnar1+/− (n = 17 from six litters); E11.5 uninfected Ifnar1−/− (n = 5) and Ifnar1+/− (n = 12 from three litters) and infected Ifnar1−/− (n = 12) and Ifnar1+/− (n = 11 from three litters); E12.5 uninfected Ifnar1−/− (n = 11) and Ifnar1+/− (n = 11 from three litters) and infected Ifnar1−/− (n = 19) and Ifnar1+/− (n = 19 from five litters). Data are pooled from at least two independent experiments from each infection time point. *P < 0.05 and **P < 0.0001 compared with all other groups by Tukey’s multiple comparison test. No significant differences were found between other groups.

  • Fig. 3 Placenta architecture of Ifnar1+/− fetuses is disrupted at E10.5.

    Ifnar1−/− females mated to Ifnar1+/− males were infected intravaginally with 1.5 × 105 PFU Cambodian ZIKV at E5.5 and harvested at E10.5. (A) Schematic of decidua and placenta architecture and cell types. (B) Placentas were fixed in PFA, and paraffin-embedded sections were stained by H&E. Whole placenta and decidua (top) or magnified labyrinth (bottom) are shown. Representative images of 10 placentas/deciduas per genotype from five litters were analyzed for infected, and four placentas/deciduas per genotype per time point from two litters were analyzed for uninfected. Labyrinth, L, and decidua, D, are labeled with respective letters. The asterisk indicates abnormal spheroid structure. Arrows indicate fetal RBCs. Scale bars, 100 μm (top) and 50 μm (bottom). (C) PFA-fixed frozen sections from infected littermates were costained for CK (red, trophoblasts), CD31 (green, blood vessels), and DAPI (blue). Representative images from at least three placentas per genotype from at least two litters are shown. Scale bars, 75 μm.

  • Fig. 4 Ifnar1+/− placentas show increased apoptosis and abnormal maternal-fetal blood barrier.

    Ifnar1−/− females mated to Ifnar1+/− males were infected intravaginally with 1.5 × 105 PFU Cambodian ZIKV at E5.5 and harvested at E10.5. Paraffin-embedded sections were stained for Ki67 (A, dividing cells) or cleaved Casp3 (B, apoptotic cells) with DAB. From (A) and (B), images from labyrinth are shown. Scale bars, 50 μm. Representative images from at least three placentas/deciduas per condition from at least two litters are shown. (C) Placentas were fixed at least 24 hours in formaldehyde with 2.5% glutaraldehyde at 4°C. The labyrinth was dissected and processed for electron microscopy. Uninfected Ifnar1+/− labyrinth shows four layers of cells between frbc and mrbc: fetal endothelial cell (ec), two syncytiotrophoblast layers (ST-II, ST-I), and the sinusoidal trophoblast giant cell (stgc). Infected Ifnar1−/− placenta is similar to the uninfected control. The barrier between maternal and fetal blood of ZIKV-infected Ifnar1+/− placentas is highly abnormal with unfused cells (left). ZIKV-infected Ifnar1+/− shows multiple examples of mixing between maternal and fetal blood (right). Scale bars, 2 or 10 μm as labeled on the image. Multiple sections and planes from one placenta per condition were analyzed.

  • Fig. 5 ZIKV-infected Ifnar1+/− fetuses show up-regulated hypoxia response genes just before demise.

    Ifnar1−/− females mated to Ifnar1+/− males were infected intravaginally with 1.5 × 105 PFU Cambodian ZIKV at E5.5 and harvested at E10.5. RNA was extracted from fetuses, and expression of previously reported hypoxia-response genes Vegfa (A), Adm (B), Bnip3 (C), Pfkfb3 (D), and Glut1 (E) analyzed by reverse transcription qPCR. Data represent n = 9 fetuses per genotype from three litters from ZIKV-infected litters and n = 3 Ifnar1−/− and n = 5 Ifnar1+/− fetuses from two uninfected litters. Data are pooled from at least two independent experiments per group. *P < 0.05, **P < 0.01, and ***P < 0.0001 by Tukey’s multiple comparison test. Data were normalized to Hprt and represented as fold change over Ifnar−/− uninfected placentas.

  • Fig. 6 Poly(I:C) injection of pregnant dams leads to fetal resorption in a maternal-IFNAR–dependent fashion.

    WT females mated to WT males (A) or Ifnar1−/− females mated to Ifnar1+/− males (B) were injected with 200-μg HMW poly(I:C) (PIC) at E7.5. Representative images from mice harvested between E9.5 and E12.5 are shown. Scale bars, 1 mm. (C) Crown-rump length was measured using ImageJ. Mean with SD and individual data points are shown. Data points represent the following: WTxWT litters untreated E9.5 (n = 17 from two litters), PIC E9.5 (n = 33 from four litters), untreated E10.5 (n = 13 from two litters), and PIC E10.5 (n = 8 from one litter); additional five injected litters showed no fetal remnants at time of harvest at E10.5. For Ifnar1−/−xIfnar1+/ litters, untreated E10.5 (n = 15 Ifnar1−/− and n = 5 Ifnar1+/− from three litters), PIC E10.5 (n = 5 Ifnar1−/− and n = 11 Ifnar1+/− from two litters), untreated E12.5 (n = 11 Ifnar1−/− and n = 11 Ifnar1+/− from three litters), and PIC E12.5 (n = 13 Ifnar1−/− and n = 7 Ifnar1+/− from two litters). Uninfected measurements for Ifnar1−/−xIfnar1+/− litters are the same as those shown in Fig. 2B.

  • Fig. 7 IFN treatment of human midgestation villous explants induces syncytial knot formation.

    (A) Human midgestation (19 to 23 weeks) chorionic villi were isolated, placed in culture, and treated with 100 or 1000 U of recombinant human IFN-β or 1000 U of IFN-λ3 for ~16 to 20 hours. Villous explants were harvested, fixed in PFA, and stained for CK19 (green, trophoblasts) and actin (red). DAPI-stained nuclei are shown in blue and differential interference contrast (DIC) (bottom). Scale bars, 100 μm. Images are representative of villi isolated from four donors. Arrow indicates syncytial knot. (B) Three-dimensional image reconstruction of mock- or IFN-β (1000 U)–treated explants stained for CK19 (green) and actin (red). DAPI-stained nuclei are shown in blue. Scale bars, 20 μm. (C) Quantification of syncytial knot size using Imaris in villi treated with 10, 100, or 1000 U of IFN-β or 1000 U of IFN-λ3 for ~24 hours. Each symbol represents an individual villous from a total of three donors, and the black line represents the mean. ***P < 0.001 by Dunnett’s multiple comparison test to mock. ns, not significant. (D) Confocal micrographs of mock- or IFN-β (1000 U)–treated villi stained for actin (red, right). DIC is shown on the left. White box denotes zoomed area shown at the bottom left (mock) or right (IFN-β). Scale bars, 20 μm.

  • Fig. 8 ISGs are induced in IFN-β–treated villous explants.

    Human midgestation (19 to 23 weeks) chorionic villi were isolated, placed in culture, and treated with 1000 U of recombinant human IFN-β or 1000 U of IFN-λ3 for ~24 hours, and RNA was extracted. (A) MA plots generated in R after DeSeq2 analysis demonstrating the differential expression between IFN-β–treated (left) or IFN-λ–treated (middle) villi relative to mock-treated controls and between IFN-β− and IFN-λ–treated villi (right). Data are plotted as log2 fold changes (y axis) and mean expression (x axis). Red symbols denote transcripts whose expression was differentially expressed at P < 0.05. (B) Graph demonstrating the number of transcripts up-regulated (in green) or down-regulated (in red) after IFN-β or IFN-λ treatment of villi. (C) Venn diagram denoting the overlap of transcripts (12 in total) between villi treated with IFN-β and those treated with IFN-λ. (D) Heat map (based on log reads per kilobase of transcript per million mapped reads values) of known ISGs from mock-, IFN-β–, or IFN-λ–treated villi. (E and F) Volcano plots of villi treated with IFN-β (E) or IFN-λ (F) denoting specific ISGs differentially expressed by treatment. Red circles denote ISGs and purple circles denote non-ISG transcripts. For RNA-seq experiments, data represent villi isolated from three independent placental preparations.

Supplementary Materials

  • immunology.sciencemag.org/cgi/content/full/3/19/eaao1680/DC1

    Methods

    Fig. S1. ISG expression is elevated in Ifnar1+/− placentas and fetuses but not in Irf3−/−Irf7−/− after ZIKV infection.

    Fig. S2. Ifnar1−/− placentas harbor more infectious ZIKV compared with Ifnar1+/− littermates.

    Fig. S3. Global gene expression analysis reveals elevated ISG levels in infected Ifnar1+/− placentas.

    Fig. S4. ZIKV infection of the maternal-fetal interface is restricted to the decidua.

    Fig. S5. Placental architecture of Ifnar1+/− fetuses is normal at E9.5 but disrupted at E11.5 and E12.5.

    Fig. S6. Labyrinth of Ifnar1+/− placenta exhibits decreased fetal endothelial cells.

    Fig. S7. Ifnar1+/− but not Ifnar1−/− fetuses are resorbed after subcutaneous infection with Brazilian ZIKV.

    Table S1. Primers for mouse qPCR

    Table S2. Top differentially regulated genes and pathways in ZIKV-infected Ifnar1+/− versus Ifnar1−/− placentas.

    Table S3. Individual values included in all graphs

    References (5661)

  • Supplementary Materials

    Supplementary Material for:

    Type I interferons instigate fetal demise after Zika virus infection

    Laura J. Yockey, Kellie A. Jurado, Nitin Arora, Alon Millet, Tasfia Rakib, Kristin M. Milano, Andrew K. Hastings, Erol Fikrig, Yong Kong, Tamas L. Horvath, Scott Weatherbee, Harvey J. Kliman, Carolyn B. Coyne, Akiko Iwasaki*

    *Corresponding author. Email: akiko.iwasaki{at}yale.edu

    Published 5 January 2018, Sci. Immunol. 2, eaao1680 (2017)
    DOI: 10.1126/sciimmunol.aao1680

    This PDF file includes:

    • Methods
    • Fig. S1. ISG expression is elevated in Ifnar1+/− placentas and fetuses but not in Irf3−/−Irf7−/− after ZIKV infection.
    • Fig. S2. Ifnar1−/− placentas harbor more infectious ZIKV compared with Ifnar1+/− littermates.
    • Fig. S3. Global gene expression analysis reveals elevated ISG levels in infected Ifnar1+/− placentas.
    • Fig. S4. ZIKV infection of the maternal-fetal interface is restricted to the decidua.
    • Fig. S5. Placental architecture of Ifnar1+/− fetuses is normal at E9.5 but disrupted at E11.5 and E12.5.
    • Fig. S6. Labyrinth of Ifnar1+/− placenta exhibits decreased fetal endothelial cells.
    • Fig. S7. Ifnar1+/− but not Ifnar1−/− fetuses are resorbed after subcutaneous infection with Brazilian ZIKV.
    • Table S1. Primers for mouse qPCR.
    • Table S2. Top differentially regulated genes and pathways in ZIKV-infected Ifnar1+/− versus Ifnar1−/− placentas.
    • Legend for table S3
    • References (56—61)

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Table S3 (Microsoft Excel format). Individual values included in all graphs.

    Files in this Data Supplement:

Stay Connected to Science Immunology

Navigate This Article