Chemical modifications of antisense morpholino oligomers enhance their efficacy against Ebola virus infection

Abstract

Phosphorodiamidate morpholino oligomers (PMOs) are uncharged nucleic acid-like molecules designed to inactivate the expression of specific genes via the antisense-based steric hindrance of mRNA translation. PMOs have been successful at knocking out viral gene expression and replication in the case of acute viral infections in animal models and have been well tolerated in human clinical trials. We propose that antisense PMOs represent a promising class of therapeutic agents that may be useful for combating filoviral infections. We have previously shown that mice treated with a PMO whose sequence is complementary to a region spanning the start codon of VP24 mRNA were protected against lethal Ebola virus challenge. In the present study, we report on the abilities of two additional VP24-specific PMOs to reduce the cell-free translation of a VP24 reporter, to inhibit the in vitro replication of Ebola virus, and to protect mice against lethal challenge when the PMOs are delivered prior to infection. Additionally, structure-activity relationship evaluations were conducted to assess the enhancement of antiviral efficacy associated with PMO chemical modifications that included conjugation with peptides of various lengths and compositions, positioning of conjugated peptides to either the 5' or the 3' terminus, and the conferring of charge modifications by the addition of piperazine moieties. Conjugation with arginine-rich peptides greatly enhanced the antiviral efficacy of VP24-specific PMOs in infected cells and mice during lethal Ebola virus challenge.

FIG. 1.

Structures of a PMO, PMO⁺, and PPMO conjugated to either the 5′ or the 3′ PMO terminus. The PPMOs are shown as (RXR)₄XB conjugates and are representative of PPMOs conjugated to other peptides used in the experiments described herein.

FIG. 2.

Targeting the EBOV VP24 gene. (A) Schematic diagram of the negative-sense EBOV genome showing relative locations for sequences homologous to VP24-5′Term, VP24-AUG, and VP24-AUG+4. (B) Partial sequence for EBOV VP24 mRNA showing regions complementary to the PMO sequences. The VP24 start codon is indicated.

FIG. 3.

Sequence-specific inhibition of EBOV VP24 gene targets in a cell-free translation assay. A plasmid containing ∼150 nucleotides encompassing the VP24-5′Term PMO target area from EBOV VP24 was fused to firefly luciferase and was used to generate RNA. Inhibition of the VP24 RNA target was assessed on the basis of the level of luciferase expression (relative light units) by using in vitro translation reaction mixtures containing different concentrations of the PMO. The PMOs used in the assay included scrambled PMO, unmodified VP24-5′Term, (RXR)₄XB-conjugated VP24-5′Term, or VP24-5′Term containing piperazine modifications to the PMO backbone at the indicated positions relative to the 5′ end. All treatments were conducted in triplicate in 96-well plates. The luciferase signal produced by each reaction mixture was quantified with a plate reader, normalized to the mean of the reaction mixtures for the water control included on each plate, and expressed as percent inhibition of luciferase translation. The data are shown as means of three replicates per data point, with error bars representing standard deviations.

FIG. 4.

Inhibition of ZEBOV-GFP replication by PMOs in vitro. Vero E6 cells were pretreated for 2 h with 0 to 10 μM of either unconjugated or peptide-conjugated VP24-5′Term PMO before infection (MOI, 1) with ZEBOV-GFP (33). After 48 h, the cells were fixed in 10% formalin and examined by fluorescence microscopy. Representative photos of cells treated with VP24 PMOs with various delivery chemistries are shown. The width of the field represented is 880 μm. (A) Vero E6 cells were treated with PMO and mock infected or infected with ZEBOV-GFP without PMO treatment. (B) Vero E6 cells were treated with 1, 5, or 10 μM of (RXR)₄XB-conjugated (at either the 5′ or the 3′ PMO terminus) VP24-5′Term or unconjugated VP24-5′Term and infected with ZEBOV-GFP. (C) GFP fluorescence was measured by the use of Discovery-1 automated microscopy (Molecular Devices Corp.) by measuring nine individual spots per well. The percent inhibition achieved by the treatments was calculated by dividing the average GFP fluorescence from treatment wells by the average GFP fluorescence from the control wells that contained medium only. The experiments were performed at least three times; the results of a single and representative experiment are shown. Peptides to which VP24-5′Term was conjugated are indicated on the x axis along with the PMO terminus (5′ or 3′) to which each was conjugated. PMOs were tested at concentrations of 10, 5, and 1 μM, with the displayed wedges indicating the relative concentration.

FIG. 5.

Dose-dependent protective efficacy of arginine-rich, peptide-tagged VP24 PMOs against lethal EBOV infection of mice. Groups of mice (n = 20 to 30) were pretreated twice at 4 h and 24 h prior to EBOV infection with either 500-μg, 5-μg, or 1-μg doses of VP24-5′Term or VP24-AUG PMOs delivered i.p. PMOs were tagged with arginine-rich peptides on either the 5′ or the 3′ end, as indicated in the keys associated with the individual graphs. All animals in these studies were challenged i.p. with 1,000 PFU of mouse-adapted EBOV and were monitored for illness and survival for at least 28 days. PBS, phosphate-buffered saline.

FIG. 6.

Various protective efficacies associated with multiple arginine peptide-tagged VP24 PMOs against lethal EBOV challenge in mice. Groups of mice (n = 20 to 30) were pretreated with 500-μg doses (delivered i.p.) of the VP24-5′Term PMO at 4 h and 24 h prior to i.p. EBOV challenge. PMOs were conjugated to peptides (RX)_nB, where n is equal to two, four, six, or eight arginine-6-aminohexanoic acid repeats, and the peptides were positioned at either the 5′ or the 3′ end of the PMO. Mice were challenged with 1,000 PFU of mouse-adapted EBOV and monitored for illness and survival for at least 28 days.

FIG. 7.

Efficient prediction of in vivo protective efficacy by activity of PMOs by in vitro translation and ZEBOV-GFP screening assays. (A) To determine the relationship between in vivo protection and in vitro inhibition of viral replication, for each PMO in Table 1, the EC₅₀ derived from the in vitro translation assay was plotted versus the percent survival associated with in vivo challenge. (B) Likewise, to evaluate the relationship between in vivo protection and PMO-mediated inhibition of VP24 translation, percent survival was plotted versus percent inhibition from data presented for each PMO presented in Table 1. NHP, nonhuman primates.