Development of ENTV reverse genetics system and phenotypic evaluation of rescued virus reveals host-specific replication patterns in mosquitoes

Abstract

Entebbe bat virus (ENTV) is a bat-associated flavivirus with no known vector. Research into the biology of this virus, including assessment of the possibility that it may be vector-transmitted, is hindered by a lack of molecular tools and robust genetic systems. Therefore, we sequenced the complete 3' untranslated region, which was not previously available, and developed an infectious clone of ENTV to facilitate further investigation of the virus. Virus derived from the clone replicated similarly to the parental virus isolate in various vertebrate cells. Surprisingly, ENTV replicated to high titers in Aedes aegypti and Aedes albopictus mosquito cell lines, but there was no replication or infection in Culex tarsalis cells. In addition, phylogenetic and bioinformatics analyses strongly suggested that ENTV may be associated with a mosquito host. Given the bioinformatics support and efficient growth in Aedes cells, we orally exposed Ae. aegypti and Ae. albopictus to ENTV to evaluate infection. The ENTV blood-fed mosquitoes were all negative for infection; however, when ENTV was intrathoracically inoculated, bypassing the initial midgut infection and escape barriers, it replicated to high levels in the body, without dissemination of infectious virus into the saliva. These findings suggest that, despite demonstrating high molecular compatibility at the cellular level in Aedes mosquitoes, Ae. aegypti and Ae. albopictus are unlikely to serve as competent vectors for ENTV transmission due to strong midgut infection barriers. The clone presented in this manuscript should help to clarify the mechanisms for transmission and maintenance of ENTV, which remain poorly understood.

Figure 1.. Establishment of ENTV infectious clone.

A) A schematic image of the reverse genetics system. ENTV genome was split into four fragments and cloned into a CMV promoter-HDV ribozyme cassette. P0 virus was generated by transfecting Gibson assembly product directly to BHK cells. B) 500ng of mock plasmid (pCAG-GFP) or ENTV_IC was transfected to BHK cells, and pictures of CPE taken every 24 hours. C) Recovery of ENTV_IC from three independent transfections. Cell culture media was collected every 24 hours, frozen at −80°C, and titrated on BHK cells.

Figure 2.. ENTV_IC shows a comparable phenotype as ENTV

A) 5dpi plaques on BHK cells. B) Plaque diameters of ENTV and ENTV_IC (mean ± SD). No statistically significant difference was detected using Mann-Whitney’s rank test (n=24, p>0.05). C, D) Sub-confluent vertebrate (C) or invertebrate (D) cell monolayers were inoculated with ENTV or ENTV_IC at an MOI of 0.01 in biological triplicate (mean ± SD). Cell culture supernatant was collected at each time point, frozen at −80°C, and titrated on BHK cells.

Figure 3.. ENTV has characteristics consistent with mosquito-borne flaviviruses.

A) Various mosquito-borne, tick-borne, and no-known vector flaviviruses nucleotide sequences were aligned with ENTV by the MUSCLE algorithm. The phylogenetic tree, rooted with hepatitis C virus as an outgroup, was constructed using the UPGMA method with 500 bootstraps in Geneious Prime. B) CS1 and CS2 alignment of flaviviruses generated by ClustalW alignment in MEGA12. Consensus from MBFs is shown at the top and each virus nucleotides are shown if differ from the consensus. The virus abbreviation and the accession numbers are listed in Supplementary Table 2. C) Host prediction support for reservoir species, presence of arthropod vector, and arthropod vector species using Viral Host Predictor by CVR Bioinformatics (https://bioinformatics.cvr.ac.uk/software/viral-host-predictor/). The bars on the graph represent the median.

Figure 4.. ENTV can replicate in mosquitoes but cannot overcome the midgut and salivary gland barriers.

A) Ae. aegypti, Ae. albopictus, and Cx. quinquefasciatus were bloodfed with ENTV. At 7 dpi, mosquito whole bodies were homogenized and levels of ENTV RNA quantified (n=72, line shows median). B, C) Ae. aegypti and Ae. albopictus female were introthoracically inoculated with ENTV. B) Mosquitoes were collected 0, 7, and 14dpi, and ENTV RNA copy in the whole body was determined by qRT-PCR. C) Mosquitoes were salivated at 14dpi and ENTV RNA copy in the carcass, legs and wings, and saliva was determined by qRT-PCR. The dotted line shows the limit of detection.