Neuroinvasiveness of the MR766 strain of Zika virus in IFNAR-/- mice maps to prM residues conserved amongst African genotype viruses

Zika virus (ZIKV) strains are classified into the African and Asian genotypes. The higher virulence of the African MR766 strain, which has been used extensively in ZIKV research, in adult IFNα/β receptor knockout (IFNAR-/-) mice is widely viewed as an artifact associated with mouse adaptation due to at least 146 passages in wild-type suckling mouse brains. To gain insights into the molecular determinants of MR766’s virulence, a series of genes from MR766 were swapped with those from the Asian genotype PRVABC59 isolate, which is less virulent in IFNAR-/- mice. MR766 causes 100% lethal infection in IFNAR-/- mice, but when the prM gene of MR766 was replaced with that of PRVABC59, the chimera MR/PR(prM) showed 0% lethal infection. The reduced virulence was associated with reduced neuroinvasiveness, with MR766 brain titers ≈3 logs higher than those of MR/PR(prM) after subcutaneous infection, but was not significantly different in brain titers of MR766 and MR/PR(prM) after intracranial inoculation. MR/PR(prM) also showed reduced transcytosis when compared with MR766 in vitro. The high neuroinvasiveness of MR766 in IFNAR-/- mice could be linked to the 10 amino acids that differ between the prM proteins of MR766 and PRVABC59, with 5 of these changes affecting positive charge and hydrophobicity on the exposed surface of the prM protein. These 10 amino acids are highly conserved amongst African ZIKV isolates, irrespective of suckling mouse passage, arguing that the high virulence of MR766 in adult IFNAR-/- mice is not the result of mouse adaptation.

La Crosse Virus Shows Strain-Specific Differences in Pathogenesis

La Crosse virus (LACV) is the leading cause of pediatric viral encephalitis in North America, and is an important public health pathogen. Historically, studies involving LACV pathogenesis have focused on lineage I strains, but no former work has explored the pathogenesis between or within lineages. Given the absence of LACV disease in endemic regions where a robust entomological risk exists, we hypothesize that some LACV strains are attenuated and demonstrate reduced neuroinvasiveness. Herein, we compared four viral strains representing all three lineages to determine differences in neurovirulence or neuroinvasiveness using three murine models. A representative strain from lineage I was shown to be the most lethal, causing >50% mortality in each of the three mouse studies. However, other strains only presented excessive mortality (>50%) within the suckling mouse neurovirulence model. Neurovirulence was comparable among strains, but viruses differed in their neuroinvasive capacities. Our studies also showed that viruses within lineage III vary in pathogenesis with contemporaneous strains, showing reduced neuroinvasiveness compared to an ancestral strain from the same U.S. state (i.e., Connecticut). These findings demonstrate that LACV strains differ markedly in pathogenesis, and that strain selection is important for assessing vaccine and therapeutic efficacies.

Ultrastructural Features of Human Liver Specimens from Patients Who Died of Dengue Hemorrhagic Fever

Recent advances in electron microscopy and tomography have revealed distinct virus-induced endoplasmic reticulum (ER) structures unique for dengue virus (DV) and other flaviviruses in cell culture models, including hepatocytes. These altered ultrastructures serve as sites for viral replication. In this study, we used transmission electron microscopy to investigate whether such structures were present in the liver of fatal dengue hemorrhagic fever (DHF) autopsy cases. In parallel, electron microscopic examination of suckling mouse brains experimentally infected with DV was performed as an in vivo model of acute DV infection. Typical features of ER changes containing abundance of replicative virions were observed in neurons and microglia of DV-infected suckling mouse brains (SMB). This indicated that the in vivo DV infection could induce similar viral replication structures as previously described in the in vitro DV-infected cell model. Nevertheless, liver tissues from autopsy of patients who died of DHF showed scant changes of ER membrane structures and rare particles of virions in hepatocytes, despite overwhelming evidence for the presence of viral antigens and RNA–indicating active virus replication. Instead hepatocytes contained an abundance of steatotic vesicles and structural damages. This lack of structural changes indicative of virus replication in human hepatocytes is discussed.