Vaccine strains of Rift Valley fever virus exhibit attenuation at the maternal-fetal placental interface

Abstract

Rift Valley fever virus (RVFV) infection causes abortions in ruminant livestock and is associated with an increased likelihood of miscarriages in women. Using sheep and human placenta explant cultures, we sought to identify tissues at the maternal-fetal interface targeted by RVFV. Sheep villi and fetal membranes were highly permissive to RVFV infection resulting in markedly higher virus titers than human cultures. Sheep cultures were most permissive to wild-type RVFV and ΔNSm infection, while live-attenuated RVFV vaccines (LAVs; MP-12, ΔNSs, and ΔNSs/ΔNSm) exhibited reduced replication. The human fetal membrane restricted wild-type and LAV replication, and when infection occurred, it was prominent on the maternal-facing side. Type I and type III interferons were induced in human villi exposed to LAVs lacking the NSs protein. This study supports the use of sheep and human placenta explants to understand vertical transmission of RVFV in mammals and whether LAVs are attenuated at the maternal-fetal interface.IMPORTANCEA direct comparison of replication of Rift Valley fever virus (RVFV) in sheep and human placental explants reveals comparative efficiencies and permissivity to infection and replication. Vaccine strains of RVFV demonstrated reduced infection and replication capacity in the mammalian placenta. This study represents the first direct cross-host comparison of the vertical transmission capacity of this high-priority emerging mosquito-transmitted virus.

Keywords: Rift Valley fever; bunyavirus; human; phlebovirus; placenta; pregnancy; sheep; vertical transmission.

Fig 1

Sheep and human placental structure. (A) Diagram depicting the structures that make up the sheep placenta. Sheep contain 70–100 placentomes that cover the gestational sac. The placentome consists of the villous, allantoic membrane, chorionic membrane, and decidua. (B) Diagram depicting the structures that make up the human placenta. The human placenta consists of the fetal-derived villous and allantoic and chorionic membranes (fetal membranes), which is attached to the maternal decidua. (C) Table representing the sheep and human tissues collected for these studies.

Fig 2

Sheep placenta explants are permissive to infection with RVFV. (A) Representative image of a whole (top), separated (middle; formalin-fixed), and dissected (bottom) sheep placenta. 5 × 5 mm sections of the villi (bottom, left), chorionic membrane (bottom, middle), and allantoic membrane (bottom, right) were cultured in 24-well plates. (B) Diagram depicting the timeline for sheep placenta explant culture infections. Tissue dissections were inoculated with 1 × 10⁵ pfu RVFV ZH501, ΔNSs, ΔNSm, MP-12, and ΔNSs/ΔNSm for 1 hour (n = 12 each). Virus was removed and washed prior to the addition of culture growth medium. Culture supernatant was collected at 0, 24, 36, 48, and 72 hours post-infection (hpi). (C) q-RT-qPCR was performed to quantitate virus production over time. Dashed line = limit of detection (LOD). Statistical significance was determined by one-way ANOVA compared to ZH501 cultures. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, and ^****P < 0.0001. n.s. = not significant. Red, green, purple, and orange asterisks designate the P-value of ΔNSs, ΔNSm, MP-12, and ΔNSs/ΔNSm compared to ZH501, respectively.

Fig 3

ZH501 and ΔNSm undergo widespread infection in sheep placenta explants. Immunohistochemistry (IHC) images of sheep allantoic membrane (AM), chorionic membrane (CM), and villi (V) cultures infected with RVFV ZH501, ΔNSm, ΔNSs, MP-12, and ΔNSs/ΔNSm or uninfected controls. (A) Images were taken at 20× magnification by light (optical) microscopy. (B) 40× magnification images of sheep placenta cultures infected with MP-12. Red-brown = RVFV nucleoprotein. Blue staining = cell structural counterstain. Black arrowheads highlight regions infected with RVFV.

Fig 4

Human placenta explant cultures are permissive to RVFV strains. (A) Representative image of a whole (top) and dissected (bottom) human placenta and decidua. (B) Experimental timeline for human placenta explant culture infections. Tissue dissections were inoculated with 1 × 10⁵ pfu RVFV ZH501, ΔNSs, ΔNSm, MP-12, and ΔNSs/ΔNSm for 1 hour (n = 3–5 each). Virus was removed and washed prior to the addition of culture growth medium. Culture supernatant was collected at 0, 24, 36, 48, and 72 hpi. (C) q-RT-PCR was performed to quantitate virus production over time. Dashed line = LOD. Statistical significance was determined by one-way ANOVA compared to ZH501 cultures. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, and ^****P < 0.0001. n.s. = not significant. Red, green, purple, and orange asterisks designate the P-value of ΔNSs, ΔNSm, MP-12, and ΔNSs/ΔNSm compared to ZH501, respectively.

Fig 5

RVFV ZH501 and attenuated strains infect human placenta villous explants, and RVFV ZH501 infects the maternal-facing side of the fetal membrane. (A) Immunohistochemistry images of human villous cultures infected with RVFV ZH501, ΔNSm, ΔNSs, MP-12, and ΔNSs/ΔNSm or uninfected controls. Images were taken at 40× magnification by light microscopy. Green arrowheads highlight regions of syncytiotrophoblasts infected with RVFV. Blue arrowheads highlight intravillous cells infected with RVFV. (B) Immunohistochemistry images of ZH501 infected (left) or uninfected (right) fetal membrane cultures. Full tissue scans were performed at 20× magnification. Structures of the fetal membrane are highlighted by brackets. Structures on the fetal side include the amnion epithelial cells (AEC) and amniotic mesoderm (A Meso). The spongy layer (SL) separates the fetal and maternal sides. Structures on the maternal side include the chorionic mesoderm (C Meso) and the chorionic trophoblasts (C Tropho). Black arrowheads highlight punctate regions infected with RVFV. Red-brown = RVFV nucleoprotein. Blue staining = cell structural counterstain.

Fig 6

Attenuated RVFV LAVs induce type I and III IFNs in human placenta. Protein levels of IFN-α (A) or IFN-λ1 (B) were measured in culture supernatant collected at endpoint (48 or 72 hpi) by ELISA. Culture supernatant was collected from villous, decidua, or fetal membrane cultures inoculated with RVFV ZH501, ΔNSm, ΔNSs, MP-12, and ΔNSs/ΔNSm (n = 3–5). IFN-α LOD = 3.2 pg/mL. IFN-λ1 LOD = 13.72 pg/mL. Statistical significance was determined by two-way ANOVA compared to uninfected cultures. ^*P < 0.05; ^**P < 0.01. n.s. = not significant.

Fig 7

The rate of virus production is similar between hosts, despite delayed production in human cultures. A simple linear regression was performed to calculate the rate of virus production (slope) from virus growth curves for decidua, fetal membrane, and villous cultures starting at 24 hpi (sheep) and 36 hpi (humans). Positive slope = red shades; no slope = black. Individual slopes are depicted by white numbers.