Maternal B Cell and Antibody Responses to Zika Virus for Design of Immune Interventions in Pregnancy

Zika virus (ZIKV) re-emerged in the 2015-2016 epidemic in the Americas, when it was recognized that this mosquito transmitted virus can also be transmitted congenitally. One in 10 infants born to ZIKV-infected pregnancies presented with congenital defects including microcephaly, brain and ocular damage, neurodevelopmental delays, and mobility defects. In the past outbreak 11,000 children in Brazil were born with microcephaly and even more have lifelong disability. To prevent congenital transmission, a vaccine that is effective in pregnancy is urgently needed. Developing such an intervention requires an understanding of the targeted immune responses that mediates protection in pregnancy. Therefore, we investigated maternal B cell and antibody immunity to ZIKV.

First, we characterized ZIKV immunity in mothers with natural ZIKV infection and found that all mounted high ZIKV neutralizing antibody titers within 10 days of symptoms and maintained this throughout gestation. We then evaluated transplacental transfer of IgG and found that maternal ZIKV infection did not impair transfer of vaccine-elicited and flavivirus-neutralizing IgG, indicating that maternal immunization is a viable option to transfer immune protection to the newborn. However, a potential risk of transferred IgG is antibody-dependent enhancement of a heterologous flavivirus in early life. Consequently, we next tested the role of vertically transferred cross-reactive IgG in infant cord blood and found that transferred IgG may mediate in vitro flavivirus enhancement in absence of high ZIKV-neutralizing titers.

Next, we evaluated the role of IgM antibodies in the maternal immune response to ZIKV, as IgM do not cross the placenta during gestation and were found for an unusually long time in many cohorts. We found that plasma IgM contributes to early ZIKV neutralization across several ZIKV-infected pregnant women. We then isolated a potently ZIKV-neutralizing IgM monoclonal antibody, DH1017.IgM, which demonstrates 39-fold increased neutralization activity than a recombinant IgG with the same antigen binding sites, suggesting that the multivalency of the IgM may have a role in function. Potency of DH1017.IgM increased in a dose-dependent manner with complement. Structural studies of the epitope revealed a novel E dimer epitope on Domain II. This epitope can be bound by DH1017.Fab at multiple angles and is computationally predicted to have the capacity to be bound by all five IgM monomers simultaneously, defining a novel route of ZIKV neutralization. Importantly, the DH017.IgM protects mice from a lethal challenge of ZIKV and reduced in vitro enhancement otherwise observed with DH1017.IgG, suggesting that DH1017.IgM may be a suitable candidate for prophylactic intervention in pregnancy.

Altogether, we have identified key aspects of maternal immunity that will inform the development of ZIKV vaccines. First, we show that pregnant women can mount a robust B cell response against ZIKV, that is durable throughout pregnancy, and that protective levels of vaccine-elicited IgG can be transferred transplacentally to infants, suggesting that maternal immunization is a viable strategy to prevent infections in pregnancy and protect the newborn. However, with the transfer of flavivirus cross-reactive IgG, there is a need to monitor infants for enhanced flavivirus disease upon maternal immunization. Subsequently, we found that IgM antibodies in plasma contribute to ZIKV neutralization and identified a novel route for virus neutralization by DH1017.IgM that can be leveraged in immunogen design. Thus, vaccine design should assess if inclusion of multimeric immunogens that mimic the conformational surface of the virion supports development of IgM-mediated immunity.