Reduction of the rate of poliovirus protein synthesis through large-scale codon deoptimization causes attenuation of viral virulence by lowering specific infectivity

Abstract

Exploring the utility of de novo gene synthesis with the aim of designing stably attenuated polioviruses (PV), we followed two strategies to construct PV variants containing synthetic replacements of the capsid coding sequences either by deoptimizing synonymous codon usage (PV-AB) or by maximizing synonymous codon position changes of the existing wild-type (wt) poliovirus codons (PV-SD). Despite 934 nucleotide changes in the capsid coding region, PV-SD RNA produced virus with wild-type characteristics. In contrast, no viable virus was recovered from PV-AB RNA carrying 680 silent mutations, due to a reduction of genome translation and replication below a critical level. After subcloning of smaller portions of the AB capsid coding sequence into the wt background, several viable viruses were obtained with a wide range of phenotypes corresponding to their efficiency of directing genome translation. Surprisingly, when inoculated with equal infectious doses (PFU), even the most replication-deficient viruses appeared to be as pathogenic in PV-sensitive CD155tg (transgenic) mice as the PV(M) wild type. However, infection with equal amounts of virus particles revealed a neuroattenuated phenotype over 100-fold. Direct analysis indicated a striking reduction of the specific infectivity of PV-AB-type virus particles. Due to the distribution effect of many silent mutations over large genome segments, codon-deoptimized viruses should have genetically stable phenotypes, and they may prove suitable as attenuated substrates for the production of poliovirus vaccines.

FIG. 1.

Codon use statistics in synthetic P1 capsid designs. PV-SD maintains nearly identical codon frequencies compared to wt, while maximizing codon positional changes within the sequence. In PV-AB capsids, the use of nonpreferred codons was maximized. The lengths of the bars and the numbers behind each bar indicate the occurrence of each codon in the sequence. As a reference, the normal human synonymous codon frequencies(Freq. [as a percentage of 100]) for each amino acid are given in the third column.

FIG. 2.

Codon-deoptimized virus phenotypes. (A) Overview of virus constructs used in this study. (B) One-step growth kinetics in HeLa cell monolayers. (C to H) Plaque phenotypes of codon-deoptimized viruses after 48 h (C to F) or 72 h (G and H) of incubation; stained with anti 3D^pol antibody to visualize infected cells. (C) PV(M), (D) PV-SD, (E) PV-AB, (F) PV-AB^755-1513, (G and H) PV-AB^2470-2954. Cleared plaque areas are outlined by a rim of infected cells (C and D). (H) No plaques are apparent with PV-AB^2470-2954 after subsequent crystal violet staining of the well shown in panel G. (I and J) Microphotographs of the edge of an immunostained plaque produced by PV(M) (I) or an infected focus produced by PV-AB^2470-2954 (J) after 48 h of infection.

FIG. 3.

Codon deoptimization leads to a reduction of specific infectivity. (A) Agarose gel electrophoresis of virion genomic RNA isolated from purified virus particles of PV(M) (lane 1), PV-AB^755-1513 (lane 2), and PV-AB^2470-2954 (lane 3). (B) Silver-stained SDS-PAGE protein gel of purified PV(M) (lane 1), PV-AB^755-1513 (lane 2), and PV-AB^2470-2954 (lane 3) virus particles. The three larger of the four capsid capsid proteins (VP1, VP2, and VP3) are shown, demonstrating the purity and relative amounts of virus preparations. (C) Development of a virus capture ELISA using a poliovirus receptor-alkaline phosphatase (CD155-AP) fusion protein probe. Virus-specific antibodies were used to coat ELISA plates, and samples containing an unknown virus concentration were applied followed by detection with CD155-AP. Virus concentrations were calculated using a standard curve prepared in parallel with known amounts of purified wt virus (E). (D) The amounts of purified virus and extracted virion RNA were spectrophotometrically quantified, and the number of particles or genome equivalents (1 genome = 1 virion) was calculated. In addition, virion concentrations were determined by ELISA. The infectious titer of each virus was determined by plaque/infected-focus assay, and the specific infectivity was calculated as PFU/particle or FFU/particle.

FIG. 4.

The PV-AB phenotype is determined at the level of genome translation. (A) A standard in vitro translation in HeLa S10 extract, in the presence of exogenously added amino acids and tRNAs reveals no differences in translation capacities of codon-deoptimized genomes compared to the PV(M) wt. Shown is an autoradiograph of [³⁵S]methionine-labeled translation products resolved on a 12.5% SDS-PAGE gel. The identity of an aberrant band (*) is not known. (B) In vitro translation in nondialyzed HeLa S10 extract without the addition of exogenous amino acids and tRNA and in the presence of competing cellular mRNAs uncovers a defect in translation capacities of codon-deoptimized PV genomes. Shown is a Western blot of poliovirus 2C reactive translation products (2C^ATPase, 2BC, and P2) resolved on a 10% SDS-PAGE gel. The relative amounts of the 2BC translation products are expressed below each lane as percentages of the wt band.

FIG. 5.

Analysis of in vivo translation using dicistronic reporter replicons confirms the detrimental effect of codon deoptimization on PV translation. (A) Schematic of dicistronic replicons. Various P1 capsid coding sequences were inserted upstream of the firefly luciferase gene (F-Luc). Determination of changing levels of F-Luc expression relative to an internal control (R-Luc) allows for the quantification of ribosome transit through the P1 capsid region. (B) Replicon RNAs were transfected into HeLa cells and incubated for 7 h in the presence of 2 mM guanidine-hydrochloride to block RNA replication. The relative rate of translation through the P1 region was inversely proportional to the extent of codon deoptimization. While the capsid coding sequences of two viable virus constructs, PV-AB^2470-2954 and PV-AB^2954-3386, allow between 60 and 80% of wt translation, translation efficiency below 20% is associated with the lethal phenotypes observed with the PV-AB, PV-AB^2470-3386, and PV-AB^1513-2470 genomes. Each value represents the average of six determinations from three independent experiments.