The intraleader AUG nucleotide sequence context is important for equine arteritis virus replication

Abstract

The 5(-terminal leader sequence of the equine arteritis virus (EAV) genome contains an open reading frame (ORF) with an AUG codon in a suboptimal context for initiation of protein synthesis. To investigate the significance of this intraleader ORF (ILO), an expression plasmid was generated carrying a DNA copy of the subgenomic mRNA7 behind a T7 promoter. Capped RNA transcribed from this construct was shown to direct, in an in vitro translation system, the synthesis of leader peptide as well as N protein. Site-directed mutations aimed to either optimize or weaken the sequence context of the ILO start codon affected leader peptide synthesis as predicted; no peptide was detected when the initiation codon was incapacitated. Translation of the downstream N gene was inversely affected by leader peptide production, consistent with a ribosomal leaky scanning mechanism. To investigate the role of the leader peptide in the EAV replication life cycle we generated, using an infectious EAV cDNA clone, two mutant viruses in one of which the ILO start codon was in an optimal Kozak context for translation initiation while in the other the codon was again incapacitated. Surprisingly, both mutant viruses were equally viable and exhibited similar phenotypes in BHK-21 cells. However, their replication kinetics and viral yields were reduced relative to that of the wild-type parental virus, as were their plaque sizes. Importantly, the mutations introduced into the viruses appeared to be rapidly and precisely repaired upon passaging. Already after one viral passage a significant fraction of the viruses had regained the wild-type sequence as well as its phenotype. The results demonstrate that EAV replication is not dependent on the synthesis of the intraleader peptide. Rather, the leader peptide does not seem to have any function in the EAV life cycle. As we discuss, the available data indicate that the ILO 5( nucleotide sequence per se, not its functioning in translation initiation, is of critical importance for EAV replication.

Fig. 1

Schematic representation of the EAV mRNA7-specific plasmid and its derivatives. The parental cDNA clone was generated by RT-PCR and inserted into pUC18 as described in Materials and methods. Indicated are, going from 5( to 3(: the T7 RNA polymerase promoter sequence (T7, black box), the leader sequence (LS, large black line) containing the ILO start codon with the desired point mutations as indicated (capitals in open box), the intergenic sequence (IS) between the leader sequence, the nucleocapsid start codon (ATG[N]), and the poly(A) tail

Fig. 2

In vitro synthesis of the EAV IL peptide and N protein. (A) Direct visualization of [³⁵S] methionine-labeled in vitro translation products in tricine-buffered SDS–10% PAA gels. The transcripts used for in vitro translation were derived from F7(wt)ATG (lane 1), F7(koz)ATG (lane 2), F7(opt)ATG (lane 3), F7(−)ATG (lane 4) and F7(nf)ATG (lane 5). Lanes 6 and 7 show the result of in vitro translation reactions without RNA (negative control) and with a transcript encoding the luciferase (Luc) protein (positive control), respectively. The positions of the IL peptide (ILP) and the nucleocapsid protein (N) are indicated. (B) RIPA of the EAV IL peptide synthesized in vitro using a rabbit antiserum specific for the IL peptide (+) or the corresponding preimmune serum (−) The lane number assignments are the same as in panel A. [¹⁴C]-labeled molecular weight (MW) standards (kDa) (GE Healthcare) are indicated at the left

Fig. 3

Effect of the ILO start codon and its direct surroundings on the in vitro translation of the EAV N protein. In vitro translation products of the mRNA7 expression plasmids [F7(wt)ATG, F7(−)ATG, F7(nf)ATG, F7(opt)ATG, and F7(koz)ATG; lanes 1–5, respectively] were analyzed directly by SDS-PAGE and after immunoprecipitation (RIPA) using a rabbit antiserum raised against the EAV N protein (+) or the corresponding preimmune serum (−). The mean relative amounts of N protein were calculated from both experiments and for each construct and normalized to the parental construct (F7(wt)ATG) as indicated at the bottom

Fig. 4

Growth curves of the EAV recombinant viruses collected after cell transfection (P0) or at the fifteenth viral passage (P15). (a) Comparison of growth kinetics of the virus in which the ILO start codon was knocked out (recEAV[ko]AUG-P0) with those of the recombinant wild-type virus (recEAV[wt]AUG-P0) and recombinant viruses in which the wild-type leader sequence was restored by genetic manipulation (recEAV[ko-to-wt]AUG-P0) or by repeated passages (recEAV[ko]AUG-P15). (b) Comparison of growth kinetics of the virus in which the context of the ILO start codon was improved (recEAV[opt]AUG-P0) with those of the recombinant wild-type virus (recEAV[wt]AUG-P0) and recombinant viruses in which the wild-type leader sequence was restored by genetic manipulation (recEAV[opt-to-wt]AUG-P0) or by repeated passages (recEAV[opt]AUG-P15). The results are expressed as the mean titers (three independent experiments) ( SD

Fig. 5

Plaque morphology of EAV recombinant viruses. Monolayers of Vero cells were infected with recombinant viruses recEAV[ko]AUG-P0 (a), recEAV[ko]AUG-P15 (b), recEAV[ko-to-wt]AUG-P0 (c), recEAV[opt]AUG-P0 (d), recEAV[opt]AUG-P15 (e), recEAV[opt-to-wt]AUG-P0 (f), or recEAV[wt]AUG-P0 (g), collected after cell transfection (P0) or at the fifteenth viral passage (P15), or mock-infected (negative control) (h) or infected with the Bucyrus strain of EAV (positive control) (i), and covered with an agar overlay. At 4 days pi, the overlay was removed and the monolayers were fixed and stained with 0.9% violet crystal in 25% formaldehyde and 5% ethanol