Efficient transplacental IgG transfer in women infected with Zika virus during pregnancy

Abstract

Zika virus (ZIKV) is a newly-identified infectious cause of congenital disease. Transplacental transfer of maternal IgG to the fetus plays an important role in preventing many neonatal infections. However, antibody transfer may also have negative consequences, such as mediating enhancement of flavivirus infections in early life, or trafficking of virus immune complexes to the fetal compartment. ZIKV infection produces placental pathology which could lead to impaired IgG transfer efficiency as occurs in other maternal infections, such as HIV-1 and malaria. In this study, we asked whether ZIKV infection during pregnancy impairs transplacental transfer of IgG. We enrolled pregnant women with fever or rash in a prospective cohort in Vitoria, Brazil during the recent ZIKV epidemic. ZIKV and dengue virus (DENV)-specific IgG, ZIKV and DENV neutralizing antibodies, and routine vaccine antigen-specific IgG were measured in maternal samples collected around delivery and 20 paired cord blood samples. We concluded that 8 of these mothers were infected with ZIKV during pregnancy and 12 were ZIKV-uninfected. The magnitude of flavivirus-specific IgG, neutralizing antibody, and vaccine-elicited IgG were highly correlated between maternal plasma and infant cord blood in both ZIKV-infected and -uninfected mother-infant pairs. Moreover, there was no difference in the magnitude of plasma flavivirus-specific IgG levels between mothers and infants regardless of ZIKV infection status. Our data suggests that maternal ZIKV infection during pregnancy does not impair the efficiency of placental transfer of flavivirus-specific, functional, and vaccine-elicited IgG. These findings have implications for the neonatal outomes of maternal ZIKV infection and optimal administration of antibody-based ZIKV vaccines and therapeutics.

Fig 1. Algorithm used to categorize flavivirus exposure history according to ZIKV and DENV focus-reduction neutralization-50 titers (FRNT-50).

Maternal and cord blood plasma were tested by FRNT50 against 5 viruses (ZIKV and 4 DENV serotypes) and FRNT-50 titers used to infer flavivirus exposure history. All samples were anti-flavivirus IgM negative, reducing the likelihood of cross-reactivity resulting from recent infections.

Fig 2. Histology of the placenta from a ZIKV-infected pregnant mother.

Placental tissue from subject B1_0004 was stained with hematoxylin and eosin. Lymphocytes and macrophages are present in the chorionic villi (A-100X, B-400X). The arrow indicates inflammatory cells within a villus.

Fig 3. ZIKV binding and neutralizing IgG responses persist throughout pregnancy.

A. Maternal plasma collected serially from three women diagnosed with ZIKV during pregnancy was assessed for IgG binding responses to ZIKV via virion capture ELISA. The estimated dilution at 50% of maximal binding (ED₅₀) was calculated from serial dilutions of maternal plasma. Black filled points indicate time points when ZIKV viremia was detected by RT-PCR. B. ZIKV focus reduction neutralization titer throughout pregnancy.

Fig 4. Efficient transplacental transfer of flavivirus-specific IgG.

Plasma antibody binding to ZIKV, DENV1, DENV2, DENV3, and DENV4 was measured by virion capture ELISA using serial dilutions of maternal plasma and infant cord blood collected at delivery. The dilution at 50% of maximal binding (ED50) was calculated and the infant ED50 was assessed as a percentage of the maternal ED50 to yield percent transfer. Dotted line indicates 100% transfer and the solid line indicates the median. No significant differences in percent transfer were found in comparing ZIKV-infected and uninfected women for the all viruses tested by Exact Wilcoxon Rank Sum Test; Bonferroni adjusted P > 0.05 for all viruses tested.

Fig 5. Maternal ZIKV infection does not disrupt transplacental transfer of DENV neutralizing IgG.

Kendall Tau correlation of focus neutralization reduction titer-50 (FRNT-50) for maternal plasma and infant cord blood, separated by maternal ZIKV serostatus. Panels indicate correlation of maternal and infant neutralizing titers by flavivirus: ZIKV (A), DENV1 (B), DENV2 (C), DENV3 (D) and DENV4 (E). All correlations are P<0.05, except DENV1 and ZIKV in ZIKV-infected mothers where P<0.09 and P<0.45 respectively.

Fig 6. Strong correlation of maternal and infant vaccine-elicited IgG levels in ZIKV-infected mothers indicates efficient transplacental transfer during maternal ZIKV infection.

IgG response to vaccine antigens in infant cord blood plasma and maternal plasma collected at delivery were measured by a binding antibody multiplex assay. Concentrations of vaccine-elicited IgG responses were calculated from reference sera standards as International Units (IU)/mL or μg/mL. ZIKV-infected (n = 8) and uninfected (n = 12) subjects are indicated in red and black respectively, and dotted lines denote WHO established protective IgG levels. Kendall Tau correlations were performed for each ZIKV infection group, with p<0.05 for all.