Zika Virus Infection in Mice Causes Panuveitis with Shedding of Virus in Tears

Abstract

Zika virus (ZIKV) is an emerging flavivirus that causes congenital abnormalities and Guillain-Barré syndrome. ZIKV infection also results in severe eye disease characterized by optic neuritis, chorioretinal atrophy, and blindness in newborns and conjunctivitis and uveitis in adults. We evaluated ZIKV infection of the eye by using recently developed mouse models of pathogenesis. ZIKV-inoculated mice developed conjunctivitis, panuveitis, and infection of the cornea, iris, optic nerve, and ganglion and bipolar cells in the retina. This phenotype was independent of the entry receptors Axl or Mertk, given that Axl(-/-), Mertk(-/-), and Axl(-/-)Mertk(-/-) double knockout mice sustained levels of infection similar to those of control animals. We also detected abundant viral RNA in tears, suggesting that virus might be secreted from lacrimal glands or shed from the cornea. This model provides a foundation for studying ZIKV-induced ocular disease, defining mechanisms of viral persistence, and developing therapeutic approaches for viral infections of the eye.

Figure 1. Viral burden in the eyes, tears, and lacrimal gland of ZIKV-infected animals

Four-week-old mice were infected with 10³ FFU of ZIKV Brazil (Paraiba 2015). Viral burden was measured by qRT-PCR assay. (A) Viral burden in the eyes of WT mice on days 2 and 6 after infection with ZIKV Brazil (Paraiba 2015) or French (H/PF/2013). Mice were treated with 2 mg of an anti-Ifnar1 or control mAb one day prior to infection.(B) Viral burden in the eyes of Ifnar1^−/− mice on days 6, 7, and 8 after infection with ZIKV Brazil (Paraiba 2015). (C) Viral burden in the eyes and brain of Ifnar1^−/− mice on day 28 after infection with ZIKV Brazil (Paraiba 2015). (D) Viral RNA in tears of mock- and ZIKV-infected Ifnar1^−/− mice on day 7 after infection with ZIKV Brazil (Paraiba 2015). (E) Viral RNA in the lacrimal gland of mock- and ZIKV-infected Ifnar1^−/− mice on day 7 after infection with ZIKV Brazil (Paraiba 2015). Symbols are derived from individual animals and pooled from 2 or 3 independent experiments. Bars indicate the mean of 5 to 13 mice per group. Dotted lines represent the limit of sensitivity of the assay.

Figure 2. Survival and viral burden in AG129 mice inoculated with tears and eye homogenate derived infected Ifnar1^−/− mice

Four-week-old AG129 mice were inoculated via an intraperitoneal route with 10⁴ FFU of parental ZIKV Brazil (Paraiba 2015) or eye-derived virus obtained from Ifnar1^−/− mice. (A) Kaplan-Meier survival curve in mice inoculated with day 7 or 28 eye homogenates from Ifnar1^−/− mice, or 10 µl of tears obtained 7 days after ZIKV infection of Ifnar1^−/− mice. (B) A representative photograph demonstrating gross ocular pathology and exudate in mice inoculated with 10⁴ FFU of parental ZIKV Brazil (Paraiba 2015) or eye-derived virus obtained from Ifnar1^−/− mice. (C) Representative hematoxylin and eosin stained eye sections from AG129 mice infected with parental and eye-derived ZIKV. Black arrowheads indicate inflammatory cell infiltrates in the posterior region of the eye. Scale bars = 100 µm for upper panels and 75 µm for lower panels. (D) Viral burden assessed by qRT-PCR assay in the spleen, brain, and eyes of AG129 mice inoculated with 10⁴ FFU of parental ZIKV Brazil (Paraiba 2015) or eye-derived virus obtained from Ifnar1^−/− mice. Symbols are derived from individual animals and pooled from 2 or 3 independent experiments. Bars indicate the mean of 2 to 5 to mice per group. Dotted lines represent the limit of sensitivity of the assay. See also Figure S1.

Figure 3. Viral burden in WT, Axl^−/−, Mertk^−/− or Axl^−/− Mertk^−/− DKO mice

Four-to-six-week-old WT, Axl^−/−, Mertk^−/− or Axl^−/− Mertk^−/− DKO mice were treated with 2 mg of anti-Ifnar1 mAb one day prior to subcutaneous infection with 10³ FFU of ZIKV Brazil (Paraiba 2015). On day 6 after infection, viral burden was measured by qRT-PCR in the eye (A), brain (B), serum (C) and spleen (D). Symbols are derived from individual animals and pooled from 2 independent experiments. Bars indicate the mean of 8 to 14 mice per group. Dotted lines represent the limit of sensitivity of the assay. Data were analyzed by Kruskal-Wallis one-way ANOVA with a Dunn’s multiple comparison test (P > 0.1 for comparison of all genotypes to the WT in all tissues).

Figure 4. Histology and viral burden in the eye after congenital and early postnatal infection with ZIKV

In panels (A–E), pregnant WT mice treated with 2 mg of an anti-Ifnar1 mAb or Ifnar1^−/− mice were infected subcutaneously with 10³ FFU of ZIKV Brazil (Paraiba 2015 WT), except where the French Polynesian strain is indicated in panel (E). (A) Viral burden was assessed by qRT-PCR on E13.5 in WT fetuses or E18.5 in Ifnar1^+/− fetuses following E6.5 and 12.5 infection, respectively. (B)Representative hematoxylin and eosin stained eye sections from mock- and ZIKV-infected WT fetuses on E13.5. (C) A representative photograph demonstrating the average size of mock- and ZIKV-infected Ifnar1^+/− fetuses on E18.5. (D) Representative hematoxylin and eosin stained eye sections from mock- and ZIKV-infected Ifnar1^+/− fetuses on E18.5. The retinal detachment in the mock-infected sample is a commonly seen artifact of sectioning. (E) Viral burden in the eyes of WT pups on postnatal day 8 after in utero ZIKV infection of anti-Ifnar1 mAb-treated WT pregnant dams at the indicated gestational date with the indicated strain. (F–H) WT pups were inoculated subcutaneously on postnatal day 8 with 10³ FFU of ZIKV Brazil (Paraiba 2015 WT) and harvested 6 days later. (F) Viral burden was assessed by qRT-PCR in the spleen, brain, and eye. (G) Activated caspase-3 staining of mock- (left panel) and ZIKV-infected (right panel) WT pup brains including the visual pathway (optic chiasm, optic tract, and lateral geniculate nucleus (LGN)). The image was generated by combining images of several coronal sections in the same animal into a single merged figure. (H) Activated caspase-3 staining of a ZIKV-infected WT pup brain highlighting apoptosis in the visual cortex (red box) at higher magnification (right panel). Results in A–H are from at least 2–3 independent experiments with 7 to 16 animals for viral burden analysis and 2 to 4 mice for histological analysis. Scale bars = 25 µm for (B), 60 µm for (D), 500 µm for (G) and 1 mm (H). Dotted lines represent the limit of sensitivity of the assay.

Figure 5. Histology of the eye in adult Ifnar1^−/− mice after infection with ZIKV

Four-to-eight-week-old Ifnar1^−/− mice were inoculated subcutaneously with 10³ FFU of ZIKV Brazil (Paraiba 2015). On days 6 or 8 after infection, eyes were harvested. (A) Anatomy of an uninfected eye, demonstrating anterior (cornea, lens, iris, and ciliary body) and posterior (sclera, retina, retinal pigment epithelium/choroid, and optic nerve) structures. (B) TUNEL staining of the neurosensory retina of a mock (upper panel) or ZIKV-infected animal on day 6 (lower panel). Regions shown in higher magnification are indicated by a yellow box and arrow. Scale bars = 100 µm. (C) Hematoxylin and eosin stained eye sections from mock (left panels) and ZIKV-infected animals on day 8 (right panels). Regions shown in higher magnification are indicated by a box and displayed in the upper and lower panels. Black arrowheads indicate inflammatory cell infiltrates in the anterior (upper panels) and posterior (lower panels) chambers of the eye. Scale bars = 250 µm (middle panels) and 25 µm (upper and lower panels). The data and images in this Figure are representative of two independent experiments with 2 to 4 animals per group. See also Figure S2 and S3.

Figure 6. Viral burden and localization in the anterior and posterior regions of the eye

Four-to-eight-week-old Ifnar1^−/− mice were inoculated subcutaneously with 10³ FFU of ZIKV Brazil (Paraiba 2015). Eyes were harvested for microdissection and quantitation of ZIKV RNA by qRT-PCR (day 7) or for ZIKV RNA detection by ISH (day 6 or 8). (A) Viral burden in the microdissected cornea, iris, lens, retina, retinal pigment epithelium/choroid, and optic nerve on day 7 after infection. Symbols are derived from individual animals and pooled from two independent experiments. Bars indicate the mean of 8 to 14 mice per group. Dotted lines represent the limit of sensitivity of the assay. Data were analyzed by Kruskal-Wallis one-way ANOVA with a Dunn’s multiple comparison test (P > 0.05 only for samples from optic nerve compared to RPE/choroid complex). (B and C) RNA ISH using a ZIKV-specific probe to stain eye sections of mock- (B) or ZIKV-infected (C) Ifnar1^−/− mice. Red boxes indicate regions shown at higher magnification in adjacent panels. High magnification views in (B) indicate regions of the cornea (right panel) and retina (bottom panel). High magnification views in (C) show regions of the retina (right panel), optic nerve (middle panel), and cornea (lower panel). ISH data are representative of two independent experiments with at least two animals per group. Scale bar equals 200 µm.