Abstract/Summary

Human coronaviruses are large enveloped viruses with a single stranded, positive-sense RNA genome, the largest described among all RNA viruses. Infection gives rise to a number of pathologies ranging from mild upper respiratory infections (the common cold) to acute respiratory syndromes, with documented evidence of additional involvement of the gut and renal systems. Human coronavirus HKU1, classified as a group 2 coronavirus, was first isolated in Hong Kong in 2003 and generally causes mild infections. Although the structure of the virus and its genome are typical of coronaviruses there has been limited experimental confirmation of some of the key features of the virus life cycle and its pathogenicity. Coronavirus genomes contain extensive secondary structure elements, for example at the 5' and 3' extremities as well as internally at the junctions between open reading frames but their role in HKU1 has not been documented. This project aims to investigate the RNA secondary structures of human HKU1 coronavirus to determine how these structures affect virus transcription and translation. Bioinformatics analysis identified potential RNA secondary structures in the 5’-UTR, 3’-UTR and ORF1a/1b ribosomal frameshift regions of HKU1. Subsequently, the identified 5’-UTR, 3’-UTR and frameshift sequences were constructed de novo and introduced into appropriate vectors in order to transcribe RNA in vitro and in vivo. Cloned HKU1 sequences were then flanked by two reporter genes, mCherry and eGFP, in the same vectors to provide an easily assayed readout of the function of the cloned segments and mutants thereof. In vitro and in vivo analysis using a cell free coupled transcription-and translation system and expression in insect cells respectively revealed protein expression that was dependent on the sequence appended to the reporter. In particular, the results showed that conserved stem loops in the 5’-UTR of HKU1 stimulated translation while those present in the 3’-UTR had no effect. Time course experiments suggested a more efficient initiation of translation consistent with the presence of an internal ribosome binding site (IRES) and mutational analysis using 12 structure designed nucleotide changes highlighted critical residues, notably those indicated by mutant 10, which essentially abolished the stimulatory activity shown by the parental sequence. Analysis of the 1ab junction sequence of HKU1, predicted to be a ribosomal frameshift sequence, using similar genetic constructs, confirmed for the first time a bona fide frameshift function in HKU1. Further, the rate of shift was formally measured as approximatively 25%, that is about 1 in 4 messenger RNAs encoding the 1ab junction led to translation of the downstream sequence while the majority of transcripts ceased translation within the 1ab region. The deduced rate is similar to that reported for other coronaviruses confirming that the experimental set-up provides an accurate model for measuring HKU1 folded RNA function. A limited mutational analysis of the folded structure abolished the frameshift, mapping some of the critical nucleotides concerned. Together the experimental evidence obtained provides the first formal demonstration of activity for folded RNA structures found in the genome of HKU1 and indicated some of the critical residues involved. Further study focussing on these areas might include antisense or other small molecule antiviral therapy targeting the 5’-UTR and 1abFS and apply these methods on other human coronaviruses.