Vaginal Exposure to Zika Virus during Pregnancy Leads to Fetal Brain Infection

Abstract

Zika virus (ZIKV) can be transmitted sexually between humans. However, it is unknown whether ZIKV replicates in the vagina and impacts the unborn fetus. Here, we establish a mouse model of vaginal ZIKV infection and demonstrate that, unlike other routes, ZIKV replicates within the genital mucosa even in wild-type (WT) mice. Mice lacking RNA sensors or transcription factors IRF3 and IRF7 resulted in higher levels of local viral replication. Furthermore, mice lacking the type I interferon (IFN) receptor (IFNAR) became viremic and died of infection after a high-dose vaginal ZIKV challenge. Notably, vaginal infection of pregnant dams during early pregnancy led to fetal growth restriction and infection of the fetal brain in WT mice. This was exacerbated in mice deficient in IFN pathways, leading to abortion. Our study highlights the vaginal tract as a highly susceptible site of ZIKV replication and illustrates the dire disease consequences during pregnancy.

Keywords: antiviral defense; female reproductive tract; flavivirus; innate immunity; mucosal immunity; pattern recognition receptors; sexually transmitted infections; type I interferons.

Figure 1. Vaginal Infection of Virgin Female Wild-Type Mice Leads to Viral Replication in the Vaginal Mucosa

(A and B) Seven- to 22-week-old Depo-Provera-treated virgin female mice of the indicated genotypes were infected intravaginally with 2.5 × 10⁴ PFU of Cambodian ZIKV. (A) On the indicated days post-infection, vaginal washes were collected and viral load was analyzed by RT-qPCR. (B) Mice were weighed daily and represented as a percentage of starting weight. (C) Eight-week-old WT mice were infected with 2.5 × 10⁴ PFU of ZIKV intravaginally. Vaginal washes were collected at 6 hr and 24 hr post-infection and viral load was analyzed by RT-qPCR. (*p < 0.05). (D and E) WT mice were infected with 1.5 × 10⁵ PFU of ZIKV intraperitoneally (ip) or ivag and harvested on 1 and 3 dpi. RNA was isolated from the spleen (D) and vagina (E), and ZIKV levels were analyzed by RT-qPCR and normalized to Hprt. (F and G) On 2 dpi (F), and 5 dpi (G), ZIKV levels in whole blood were analyzed by RT-qPCR. ZIKV levels from the blood are normalized to Hprt. For (A, B, F, G), results shown are mean ± SEM involving uninfected (n = 7), WT(n = 9), Mx1 (n = 5), Rag2^−/− (n = 8), Ifnar1^−/− (n = 7), Tlr7^−/− Mavs^−/−Mx1 (n = 9), and Irf3^−/− Irf7^−/− (n = 4) mice pooled from two independent replicates. For (C), individual values from (n = 5) WT mice collected at indicated time points are shown. For(D, E), results shown are mean ± SEM involving uninfected (n = 4–5), IP infected (n = 6), ivag infected (n = 6) per time point pooled from two independent replicates. The gray dashed line represents the limit of detection. See also Figure S1 and Table S1.

Figure 2. Vaginal Infection with High Doses of ZIKV Is Lethal in IFNAR-Deficient Mice

(A–C) Four- to 5-week-old Depo-Provera-treated virgin female mice of the indicated genotypes were infected intravaginally with 5.2 × 10⁵ PFU of ZIKV. Mice were monitored for weight loss until 8 dpi (A) and survival until 14 dpi (B). All Ifnar1^−/− mice developed hind limb paralysis and were euthanized after developing hind limb paralysis. Graphs are pooled from 2 independent replicates involving uninfected (n = 4), WT (n = 6), and Ifnar1^−/−(n = 6) mice. (C) Organs of euthanized Ifnar1^−/− mice were harvested and levels of ZIKV RNA was analyzed by RT-pPCR. Results shown are mean ± standard error of the mean. Bars represent the mean of six Ifnar1^−/− mice. The gray dashed line represents limit of detection.

Figure 3. ZIKV Replicates in the Vaginal Mucosa of Pregnant Mice and Infects the Placenta in Mice Lacking the IFN Pathway

(A) Mice of indicated genotypes were infected intravaginally on E4.5orE8.5with 1.5 × 10⁵ PFU of Cambodian ZIKV. On the indicated days post-infection, vaginal washes were collected and viral load was analyzed by RT-qPCR. (B and C) On 2 dpi (B), and 5 dpi (C), ZIKV levels in whole blood were analyzed by RT-qPCR. ZIKV RNA from the blood are normalized to Hprt. (D) On E18.5, placenta and fetal bodies were analyzed for presence of ZIKV using RT-pPCR. ZIKV RNA was normalized to Hprt. Resorbed fetuses from Ifnar1^−/− dams infected E4.5 were harvested on E12.5. For (A–C) results shown are the mean ± SEM involving uninfected (n = 4), WT E4.5 (n = 6), WT E8.5 (n = 5), Irf3^−/− Irf7^−/− E4.5 (n = 3), Irf3^−/−Irf7^−/− E8.5 (n = 3), Ifnar1^−/− E4.5 (n = 3), and Ifnar1^−/− E8.5 (n = 3) pregnant dams. For (D), 2–4 pregnant dams per group were infected, and n = 3–8 placentas or fetuses were analyzed. The gray dashed line represents limit of detection.

Figure 4. Vaginal Infection of Pregnant Female Mice Leads to Fetal Growth Impairment and Fetal Resorption

(A–C) Mice of indicated genotypes were infected intravaginally on E4.5 or E8.5 with 1.5 × 10⁵ PFU of Cambodian ZIKV intravaginally. On E18.5, fetal weight was measured. Representative pictures of fetuses from each indicated genotype and day of infection are shown. All were harvested at E18.5 unless otherwise indicated. (**p = < 0.001, *p = < 0.05, scale bar, 1cm). For the Ifnar1^−/− dams mated with WT sire infected on E4.5, the upper panel shows the dissected aborted fetus, and the lower panel shows the entire fetal sac. Upper panel bars indicate the mean ± SEM of the weights of 7–34 pups from 1–4 pregnant dams per genotype for uninfected mice and 16–48 pups from 2–6 pregnant dams per genotype per time point infected.

Figure 5. Vaginal Infection of Pregnant Female WT Mice Leads to Fetal Brain Infection

(A–F) Immuno-EM depicting ZIKV infected cells (anti-flavivirus NS1 antibody) in the cortex and cerebellum of WT fetal brain from WT dams infected at E4.5 (A–C) and E8.5 (D–F) sacrificed on E18.5. Arrows indicate the ZIKV virions within the infected cells. Arrowheads indicate immunolabeling of NS1. The cells depicted in are neurons (A–D), glial cell (E), and endothelial cells (F). These figures are representative of 2–4 brains examined for each group.