Dual miRNA targeting restricts host range and attenuates neurovirulence of flaviviruses

Abstract

Mosquito-borne flaviviruses are among the most significant arboviral pathogens worldwide. Vaccinations and mosquito population control programs remain the most reliable means for flavivirus disease prevention, and live attenuated viruses remain one of the most attractive flavivirus vaccine platforms. Some live attenuated viruses are capable of infecting principle mosquito vectors, as demonstrated in the laboratory, which in combination with their intrinsic genetic instability could potentially lead to a vaccine virus reversion back to wild-type in nature, followed by introduction and dissemination of potentially dangerous viral strains into new geographic locations. To mitigate this risk we developed a microRNA-targeting approach that selectively restricts replication of flavivirus in the mosquito host. Introduction of sequences complementary to a mosquito-specific mir-184 and mir-275 miRNAs individually or in combination into the 3'NCR and/or ORF region resulted in selective restriction of dengue type 4 virus (DEN4) replication in mosquito cell lines and adult Aedes mosquitos. Moreover a combined targeting of DEN4 genome with mosquito-specific and vertebrate CNS-specific mir-124 miRNA can silence viral replication in two evolutionally distant biological systems: mosquitoes and mouse brains. Thus, this approach can reinforce the safety of newly developed or existing vaccines for use in humans and could provide an additional level of biosafety for laboratories using viruses with altered pathogenic or transmissibility characteristics.

Fig 1. Schematic representation of viral genomes used in this study.

Positions of miRNA targets for brain-expressed mir-124 and mosquito-specific mir-1, mir-184, or mir-275 in the ORF and 3’NCR of DEN4 genome are indicated by blue and red boxes, respectively. Gray area represents duplicated, codon optimized DEN4 E/NS1 sequence (nts from 2130 to 2451 of DEN4 genome) encoding 98 amino acids from the C-terminal end of the DEN4 E protein and 7 amino acids from the N-terminal end of the NS1 protein. TM1 and TM2 are transmembrane helixes in the C-terminal anchor region of protein E. Red arrows indicate signalase cleavage sites. Asterisk in the table indicates that miRNA target sequence has been altered using synonymous codons.

Fig 2. Effect of a single copy of miRNA target in the 3’NCR on DEN4 replication in mosquito cells and live mosquitoes.

(A and B) Growth kinetics of D4s, D4-1s, D4-184s and D4-275s viruses in mosquito Aag2 (A) and C₇10 (B) cells. Cells were infected at an MOI of 0.01. Each time point represents an average of two replicates ± standard deviation (shown as brackets). The dashed line indicates the limit of virus detection [0.7 log₁₀ pfu/ml]. (C) A. aegypti mosquitoes were infected intrathoracically with 100 pfu of the indicated viruses and incubated for 7 days. Virus titer in each whole mosquito body suspension was determined in Vero cells. Each point represents the virus titer of an individual mosquito. Horizontal line represents the mean virus titer for all mosquitoes. The dashed line indicates the limit of virus detection [0.2 log₁₀ pfu/mosquito]. P-values were calculated using unpaired one-tailed Student's t-test and adjusted using Bonferroni correction method to account for multiple comparisons.

Fig 3. Effect of combined mir-184 and mir-275 co-targeting of DEN4 genome in the 3’NCR on virus replication in mosquito and Vero cells.

Confluent monolayers of Aag2 (A), C6/36 (B) or Vero (C) cells were infected with D4, D4-184, D4-275, D4-275x2 and D4-275-184 viruses at an MOI of 0.01. Each time point represents an average of two replicates ± standard deviation (shown as brackets). The dashed line indicates the limit of virus detection [0.7 log₁₀ pfu/ml].

Fig 4. Effect of combined mir-184 and mir-275 co-targeting of DEN4 genome in the 3’NCR on virus fitness in A. aegypti and A. albopictus.

A. aegypti ( A and C ) and A. albopictus ( B and D ) were orally infected with blood meals (BM) containing 7.1 to 7.4 log₁₀ pfu/ml of recombinant D4 or D4-275-184 virus, respectively, and processed at 14 dpi. (A and B) Each point represents virus titer in individual mosquito (n = 18 for A. aegypti exposed to D4, and n = 24 mosquitoes for each other virus). Horizontal line represents an average titer for all mosquitoes. The dashed line indicates the limit of the assay detection [0.2 log₁₀ pfu/mosquito]. P-values comparing mean titers of D4 and miRNA targeted viruses were calculated using unpaired one-tailed Student's t-test and adjusted using Bonferroni correction method to account for multiple comparisons. (C and D) Percentage of mosquitoes that became infected (blue) or developed a disseminated infection (orange) with each virus. Differences in infection and dissemination frequencies were compared between D4 and miRNA targeted viruses using one-tailed Fisher’s exact test. P-values were adjusted using Bonferroni correction method to account for multiple comparisons.

Fig 5. miRNA targeting of the DEN4 genome within the ORF and 3’NCR greatly attenuates virus replication in Aag2 and C6/36 cells but not in Vero cells.

Confluent monolayers of Aag2 (A), C6/36 (B) and Vero (C) cells were infected in duplicate with either D4, D4-E, D4-E*, D4-275-184, D4-E-NCR1, and D4-E-NCR2 viruses at an MOI of 0.01. Each time point represents an average titer for two replicates ± standard deviation (shown as brackets). The dashed line indicates the limit of virus detection [0.7 log₁₀ pfu/ml].

Fig 6. Effect of mir-184 and mir-275 co-targeting of DEN4 genome in the ORF and 3’NCR on virus fitness in A. aegypti mosquitoes.

A. aegypti were presented with blood meals containing approximately 7 log₁₀ pfu/mL of indicated recombinant viruses and processed after 14 days. (A) Each point represents virus titer in an individual mosquito (n = 24 mosquitoes per group). Horizontal line represents the mean virus titer for all mosquitoes in the group. The dashed line indicates the limit of assay detection [0.2 log₁₀ pfu/mosquito]. P-values used to compare mean titers of D4 and miRNA targeted viruses were calculated using unpaired one-tailed Student's t-test and adjusted using Bonferroni correction method to account for multiple comparisons. (B) Percentage of mosquitoes that became infected (blue) or developed a disseminated infection (orange) with each virus. Differences in infection and dissemination frequencies were compared between D4 and miRNA targeted viruses using one-tailed Fisher’s exact test, and P-values were adjusted using Bonferroni correction method to account for multiple comparisons.

Fig 7. Effect of multiple mosquito- and brain-specific miRNA target insertions on DEN4 neurovirulence and replication in the brain of sucking mice.

(A) Replication kinetics of D4, D4-E, D4-E**, D4-184, D4-E-NCR1, and D4-E-NCR2 viruses in the brains of mice. Three-day-old mice were inoculated IC with 3 log₁₀ pfu of the indicated virus, and brains were collected from three mice in each group following the course of infection. Each time point represents an average titer for two replicates ± standard deviation (shown as brackets). The dashed line indicates the limit of virus detection [1.7 log₁₀ pfu/ml]. Brain collection of mice infected with D4 and D4-184 was not performed at 14 dpi due to earlier paralysis of the animals. (B) Survival curves of sucking mice infected IC with 10³ pfu of recombinant DEN4 viruses (litter of eight 3-day-old mice per virus). Mice were monitored daily for morbidity for 21 days. We selected a 3 log₁₀ pfu dose of virus inoculation since parental DEN4 strain 814669 has an IC LD₅₀ of 407 pfu as estimated previously for this age of mice [48].