Bovine coronavirus 5'-proximal genomic acceptor hotspot for discontinuous transcription is 65 nucleotides wide

Identification of subgenomic mRNAs derived from the coronavirus 1a/1b protein gene: Implications for coronavirus transcription

Synthesis of coronavirus subgenomic mRNA (sgmRNA) is guided by the transcription regulatory sequence (TRS). sgmRNA derived from the body TRS (TRS-B) located at the 1a/1b protein gene is designated 1ab/sgmRNA. In the current study, we comprehensively identified the 1ab/sgmRNAs synthesized from TRS-Bs located at the 1a/1b protein genes of different coronavirus genera both in vitro and in vivo by RT‒PCR and sequencing. The results suggested that the degree of sequence homology between the leader TRS (TRS-L) and TRS-B may not be a decisive factor for 1ab/sgmRNA synthesis. This observation led us to revisit the coronavirus transcription mechanism and to propose that the disassociation of coronavirus polymerase from the viral genome may be a prerequisite for sgmRNA synthesis. Once the polymerase can disassociate at TRS-B, the sequence homology between TRS-L and TRS-B is important for sgmRNA synthesis. The study therefore extends our understanding of transcription mechanisms.

Biological characterization of coronavirus noncanonical transcripts in vitro and in vivo

BACKGROUND

In addition to the well-known coronavirus genomes and subgenomic mRNAs, the existence of other coronavirus RNA species, which are collectively referred to as noncanonical transcripts, has been suggested; however, their biological characteristics have not yet been experimentally validated in vitro and in vivo.

METHODS

To comprehensively determine the amounts, species and structures of noncanonical transcripts for bovine coronavirus in HRT-18 cells and mouse hepatitis virus A59, a mouse coronavirus, in mouse L cells and mice, nanopore direct RNA sequencing was employed. To experimentally validate the synthesis of noncanonical transcripts under regular infection, Northern blotting was performed. Both Northern blotting and nanopore direct RNA sequencing were also applied to examine the reproducibility of noncanonical transcripts. In addition, Northern blotting was also employed to determine the regulatory features of noncanonical transcripts under different infection conditions, including different cells, multiplicities of infection (MOIs) and coronavirus strains.

RESULTS

In the current study, we (i) experimentally determined that coronavirus noncanonical transcripts were abundantly synthesized, (ii) classified the noncanonical transcripts into seven populations based on their structures and potential synthesis mechanisms, (iii) showed that the species and amounts of the noncanonical transcripts were reproducible during regular infection but regulated in altered infection environments, (iv) revealed that coronaviruses may employ various mechanisms to synthesize noncanonical transcripts, and (v) found that the biological characteristics of coronavirus noncanonical transcripts were similar between in vitro and in vivo conditions.

CONCLUSIONS

The biological characteristics of noncanonical coronavirus transcripts were experimentally validated for the first time. The identified features of noncanonical transcripts in terms of abundance, reproducibility and variety extend the current model for coronavirus gene expression. The capability of coronaviruses to regulate the species and amounts of noncanonical transcripts may contribute to the pathogenesis of coronaviruses during infection, posing potential challenges in disease control. Thus, the biology of noncanonical transcripts both in vitro and in vivo revealed here can provide a database for biological research, contributing to the development of antiviral strategies.

Unveiling the biology of defective viral genomes in vitro and in vivo: implications for gene expression and pathogenesis of coronavirus

BACKGROUND

Defective viral genome (DVG) is a truncated version of the full-length virus genome identified in most RNA viruses during infection. The synthesis of DVGs in coronavirus has been suggested; however, the fundamental characteristics of coronavirus DVGs in gene expression and pathogenesis have not been systematically analyzed.

METHODS

Nanopore direct RNA sequencing was used to investigate the characteristics of coronavirus DVGs in gene expression including reproducibility, abundance, species and genome structures for bovine coronavirus in cells, and for mouse hepatitis virus (MHV)-A59 (a mouse coronavirus) in cells and in mice. The MHV-A59 full-length genomic cDNAs (~ 31 kilobases) were in vitro constructed to experimentally validate the origin of coronavirus DVG. The synthesis of DVGs was also experimentally identified by RT-PCR followed by sequencing. In addition, the alterations of DVGs in amounts and species under different infection environments and selection pressures including the treatment of antiviral remdesivir and interferon were evaluated based on the banding patterns by RT-PCR.

RESULTS

The results are as follows: (i) the structures of DVGs are with diversity, (ii) DVGs are overall synthesized with moderate (MHV-A59 in cells) to high (BCoV in cells and MHV-A59 in mice) reproducibility under regular infection with the same virus inoculum, (iii) DVGs can be synthesized from the full-length coronavirus genome, (iv) the sequences flanking the recombination point of DVGs are AU-rich and thus may contribute to the recombination events during gene expression, (v) the species and amounts of DVG are altered under different infection environments, and (vi) the biological nature of DVGs between in vitro and in vivo is similar.

CONCLUSIONS

The identified biological characteristics of coronavirus DVGs in terms of abundance, reproducibility, and variety extend the current model for coronavirus gene expression. In addition, the biological features of alterations in amounts and species of coronavirus DVGs under different infection environments may assist the coronavirus to adapt to the altered environments for virus fitness and may contribute to the coronavirus pathogenesis. Consequently, the unveiled biological features may assist the community to study the gene expression mechanisms of DVGs and their roles in pathogenesis, contributing to the development of antiviral strategy and public health.

The coronavirus recombination pathway

Recombination is thought to be a mechanism that facilitates cross-species transmission in coronaviruses, thus acting as a driver of coronavirus spillover and emergence. Despite its significance, the mechanism of recombination is poorly understood, limiting our potential to estimate the risk of novel recombinant coronaviruses emerging in the future. As a tool for understanding recombination, here, we outline a framework of the recombination pathway for coronaviruses. We review existing literature on coronavirus recombination, including comparisons of naturally observed recombinant genomes as well as in vitro experiments, and place the findings into the recombination pathway framework. We highlight gaps in our understanding of coronavirus recombination illustrated by the framework and outline how further experimental research is critical for disentangling the molecular mechanism of recombination from external environmental pressures. Finally, we describe how an increased understanding of the mechanism of recombination can inform pandemic predictive intelligence, with a retrospective emphasis on SARS-CoV-2.

Lessons Learned and Yet-to-Be Learned on the Importance of RNA Structure in SARS-CoV-2 Replication

SARS-CoV-2, the etiological agent responsible for the COVID-19 pandemic, is a member of the virus family Coronaviridae, known for relatively extensive (~30-kb) RNA genomes that not only encode for numerous proteins but are also capable of forming elaborate structures. As highlighted in this review, these structures perform critical functions in various steps of the viral life cycle, ultimately impacting pathogenesis and transmissibility. We examine these elements in the context of coronavirus evolutionary history and future directions for curbing the spread of SARS-CoV-2 and other potential human coronaviruses. While we focus on structures supported by a variety of biochemical, biophysical, and/or computational methods, we also touch here on recent evidence for novel structures in both protein-coding and noncoding regions of the genome, including an assessment of the potential role for RNA structure in the controversial finding of SARS-CoV-2 integration in “long COVID” patients. This review aims to serve as a consolidation of previous works on coronavirus and more recent investigation of SARS-CoV-2, emphasizing the need for improved understanding of the role of RNA structure in the evolution and adaptation of these human viruses.

Development of an Inactivated Vaccine against SARS CoV-2

The rapid spread of SARS-CoV-2 with its mutating strains has posed a global threat to safety during this COVID-19 pandemic. Thus far, there are 123 candidate vaccines in human clinical trials and more than 190 candidates in preclinical development worldwide as per the WHO on 1 October 2021. The various types of vaccines that are currently approved for emergency use include viral vectors (e.g., adenovirus, University of Oxford/AstraZeneca, Gamaleya Sputnik V, and Johnson & Johnson), mRNA (Moderna and Pfizer-BioNTech), and whole inactivated (Sinovac Biotech and Sinopharm) vaccines. Amidst the emerging cases and shortages of vaccines for global distribution, it is vital to develop a vaccine candidate that recapitulates the severe and fatal progression of COVID-19 and further helps to cope with the current outbreak. Hence, we present the preclinical immunogenicity, protective efficacy, and safety evaluation of a whole-virion inactivated SARS-CoV-2 vaccine candidate (ERUCoV-VAC) formulated in aluminium hydroxide, in three animal models, BALB/c mice, transgenic mice (K18-hACE2), and ferrets. The hCoV-19/Turkey/ERAGEM-001/2020 strain was used for the safety evaluation of ERUCoV-VAC. It was found that ERUCoV-VAC was highly immunogenic and elicited a strong immune response in BALB/c mice. The protective efficacy of the vaccine in K18-hACE2 showed that ERUCoV-VAC induced complete protection of the mice from a lethal SARS-CoV-2 challenge. Similar viral clearance rates with the safety evaluation of the vaccine in upper respiratory tracts were also positively appreciable in the ferret models. ERUCoV-VAC has been authorized by the Turkish Medicines and Medical Devices Agency and has now entered phase 3 clinical development (NCT04942405). The name of ERUCoV-VAC has been changed to TURKOVAC in the phase 3 clinical trial.

Structural phylogenetic analysis reveals lineage-specific RNA repetitive structural motifs in all coronaviruses and associated variations in SARS-CoV-2

In many single-stranded (ss) RNA viruses, the cis-acting packaging signal that confers selectivity genome packaging usually encompasses short structured RNA repeats. These structural units, termed repetitive structural motifs (RSMs), potentially mediate capsid assembly by specific RNA–protein interactions. However, general knowledge of the conservation and/or the diversity of RSMs in the positive-sense ssRNA coronaviruses (CoVs) is limited. By performing structural phylogenetic analysis, we identified a variety of RSMs in nearly all CoV genomic RNAs, which are exclusively located in the 5′-untranslated regions (UTRs) and/or in the inter-domain regions of poly-protein 1ab coding sequences in a lineage-specific manner. In all alpha- and beta-CoVs, except for Embecovirus spp, two to four copies of 5′-gUUYCGUc-3′ RSMs displaying conserved hexa-loop sequences were generally identified in Stem-loop 5 (SL5) located in the 5′-UTRs of genomic RNAs. In Embecovirus spp., however, two to eight copies of 5′-agc-3′/guAAu RSMs were found in the coding regions of non-structural protein (NSP) 3 and/or NSP15 in open reading frame (ORF) 1ab. In gamma- and delta-CoVs, other types of RSMs were found in several clustered structural elements in 5′-UTRs and/or ORF1ab. The identification of RSM-encompassing structural elements in all CoVs suggests that these RNA elements play fundamental roles in the life cycle of CoVs. In the recently emerged SARS-CoV-2, beta-CoV-specific RSMs are also found in its SL5, displaying two copies of 5′-gUUUCGUc-3′ motifs. However, multiple sequence alignment reveals that the majority of SARS-CoV-2 possesses a variant RSM harboring SL5b C241U, and intriguingly, several variations in the coding sequences of viral proteins, such as Nsp12 P323L, S protein D614G, and N protein R203K-G204R, are concurrently found with such variant RSM. In conclusion, the comprehensive exploration for RSMs reveals phylogenetic insights into the RNA structural elements in CoVs as a whole and provides a new perspective on variations currently found in SARS-CoV-2.

Effects of Coronavirus Persistence on the Genome Structure and Subsequent Gene Expression, Pathogenicity and Adaptation Capability

Coronaviruses are able to establish persistence. However, how coronaviruses react to persistence and whether the selected viruses have altered their characteristics remain unclear. In this study, we found that the persistent infection of bovine coronavirus (BCoV), which is in the same genus as SARS-COV-2, led to alterations of genome structure, attenuation of gene expression, and the synthesis of subgenomic mRNA (sgmRNA) with a previously unidentified pattern. Subsequent analyses revealed that the altered genome structures were associated with the attenuation of gene expression. In addition, the genome structure at the 5' terminus and the cellular environment during the persistence were responsible for the sgmRNA synthesis, solving the previously unanswered question regarding the selection of transcription regulatory sequence for synthesis of BCoV sgmRNA 12.7. Although the BCoV variants (BCoV-p95) selected under the persistence replicated efficiently in cells without persistent infection, its pathogenicity was still lower than that of wild-type (wt) BCoV. Furthermore, in comparison with wt BCoV, the variant BCoV-p95 was not able to efficiently adapt to the challenges of alternative environments, suggesting wt BCoV is genetically robust. We anticipate that the findings derived from this fundamental research can contribute to the disease control and treatments against coronavirus infection including SARS-CoV-2.

The Impacts of Antivirals on the Coronavirus Genome Structure and Subsequent Pathogenicity, Virus Fitness and Antiviral Design

With the global threat of SARS-CoV-2, much effort has been focused on treatment and disease control. However, how coronaviruses react to the treatments and whether the surviving viruses have altered their characteristics are also unanswered questions with medical importance. To this end, bovine coronavirus (BCoV), which is in the same genus as SARS-CoV-2, was used as a test model and the findings were as follows. With the treatment of antiviral remdesivir, the selected BCoV variant with an altered genome structure developed resistance, but its pathogenicity was not increased in comparison to that of wild type (wt) BCoV. Under the selection pressure of innate immunity, the genome structure was also altered; however, neither resistance developed nor pathogenicity increased for the selected BCoV variant. Furthermore, both selected BCoV variants showed a better efficiency in adapting to alternative host cells than wt BCoV. In addition, the previously unidentified feature that the spike protein was a common target for mutations under different antiviral treatments might pose a problem for vaccine development because spike protein is a common target for antibody and vaccine designs. The findings derived from this fundamental research may contribute to the disease control and treatments against coronaviruses, including SARS-CoV-2.

Isolation and characterization of severe acute respiratory syndrome coronavirus 2 in Turkey

Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and associated with severe respiratory illness emerged in Wuhan, China, in late 2019. The virus has been able to spread promptly across all continents in the world. The current pandemic has posed a great threat to public health concern and safety. Currently, there are no specific treatments or licensed vaccines available for COVID-19. We isolated SARS-CoV-2 from the nasopharyngeal sample of a patient in Turkey with confirmed COVID-19. We determined that the Vero E6 and MA-104 cell lines are suitable for supporting SARS-CoV-2 that supports viral replication, development of cytopathic effect (CPE) and subsequent cell death. Phylogenetic analyses of the whole genome sequences showed that the hCoV-19/Turkey/ERAGEM-001/2020 strain clustered with the strains primarily from Australia, Canada, England, Iran and Kuwait and that the cases in the nearby clusters were reported to have travel history to Iran and to share the common unique nucleotide substitutions.

Characterization of the Role of Hexamer AGUAAA and Poly(A) Tail in Coronavirus Polyadenylation

Similar to eukaryotic mRNA, the positive-strand coronavirus genome of ~30 kilobases is 5’-capped and 3’-polyadenylated. It has been demonstrated that the length of the coronaviral poly(A) tail is not static but regulated during infection; however, little is known regarding the factors involved in coronaviral polyadenylation and its regulation. Here, we show that during infection, the level of coronavirus poly(A) tail lengthening depends on the initial length upon infection and that the minimum length to initiate lengthening may lie between 5 and 9 nucleotides. By mutagenesis analysis, it was found that (i) the hexamer AGUAAA and poly(A) tail are two important elements responsible for synthesis of the coronavirus poly(A) tail and may function in concert to accomplish polyadenylation and (ii) the function of the hexamer AGUAAA in coronaviral polyadenylation is position dependent. Based on these findings, we propose a process for how the coronaviral poly(A) tail is synthesized and undergoes variation. Our results provide the first genetic evidence to gain insight into coronaviral polyadenylation.

Coronavirus cis-Acting RNA Elements

Coronaviruses have exceptionally large RNA genomes of approximately 30 kilobases. Genome replication and transcription is mediated by a multisubunit protein complex comprised of more than a dozen virus-encoded proteins. The protein complex is thought to bind specific cis-acting RNA elements primarily located in the 5′- and 3′-terminal genome regions and upstream of the open reading frames located in the 3′-proximal one-third of the genome. Here, we review our current understanding of coronavirus cis-acting RNA elements, focusing on elements required for genome replication and packaging. Recent bioinformatic, biochemical, and genetic studies suggest a previously unknown level of conservation of cis-acting RNA structures among different coronavirus genera and, in some cases, even beyond genus boundaries. Also, there is increasing evidence to suggest that individual cis-acting elements may be part of higher-order RNA structures involving long-range and dynamic RNA–RNA interactions between RNA structural elements separated by thousands of nucleotides in the viral genome. We discuss the structural and functional features of these cis-acting RNA elements and their specific functions in coronavirus RNA synthesis.

Isolation and Characterization of Dromedary Camel Coronavirus UAE-HKU23 from Dromedaries of the Middle East: Minimal Serological Cross-Reactivity between MERS Coronavirus and Dromedary Camel Coronavirus UAE-HKU23

Recently, we reported the discovery of a dromedary camel coronavirus UAE-HKU23 (DcCoV UAE-HKU23) from dromedaries in the Middle East. In this study, DcCoV UAE-HKU23 was successfully isolated in two of the 14 dromedary fecal samples using HRT-18G cells, with cytopathic effects observed five days after inoculation. Northern blot analysis revealed at least seven distinct RNA species, corresponding to predicted subgenomic mRNAs and confirming the core sequence of transcription regulatory sequence motifs as 5′-UCUAAAC-3′ as we predicted previously. Antibodies against DcCoV UAE-HKU23 were detected in 58 (98.3%) and 59 (100%) of the 59 dromedary sera by immunofluorescence and neutralization antibody tests, respectively. There was significant correlation between the antibody titers determined by immunofluorescence and neutralization assays (Pearson coefficient = 0.525, p < 0.0001). Immunization of mice using recombinant N proteins of DcCoV UAE-HKU23 and Middle East respiratory syndrome coronavirus (MERS-CoV), respectively, and heat-inactivated DcCoV UAE-HKU23 showed minimal cross-antigenicity between DcCoV UAE-HKU23 and MERS-CoV by Western blot and neutralization antibody assays. Codon usage and genetic distance analysis of RdRp, S and N genes showed that the 14 strains of DcCoV UAE-HKU23 formed a distinct cluster, separated from those of other closely related members of Betacoronavirus 1, including alpaca CoV, confirming that DcCoV UAE-HKU23 is a novel member of Betacoronavirus 1.

Discovery of a novel coronavirus, China Rattus coronavirus HKU24, from Norway rats supports the murine origin of Betacoronavirus 1 and has implications for the ancestor of Betacoronavirus lineage A

We discovered a novel Betacoronavirus lineage A coronavirus, China Rattus coronavirus (ChRCoV) HKU24, from Norway rats in China. ChRCoV HKU24 occupied a deep branch at the root of members of Betacoronavirus 1, being distinct from murine coronavirus and human coronavirus HKU1. Its unique putative cleavage sites between nonstructural proteins 1 and 2 and in the spike (S) protein and low sequence identities to other lineage A betacoronaviruses (βCoVs) in conserved replicase domains support ChRCoV HKU24 as a separate species. ChRCoV HKU24 possessed genome features that resemble those of both Betacoronavirus 1 and murine coronavirus, being closer to Betacoronavirus 1 in most predicted proteins but closer to murine coronavirus by G+C content, the presence of a single nonstructural protein (NS4), and an absent transcription regulatory sequence for the envelope (E) protein. Its N-terminal domain (NTD) demonstrated higher sequence identity to the bovine coronavirus (BCoV) NTD than to the mouse hepatitis virus (MHV) NTD, with 3 of 4 critical sugar-binding residues in BCoV and 2 of 14 contact residues at the MHV NTD/murine CEACAM1a interface being conserved. Molecular clock analysis dated the time of the most recent common ancestor of ChRCoV HKU24, Betacoronavirus 1, and rabbit coronavirus HKU14 to about the year 1400. Cross-reactivities between other lineage A and B βCoVs and ChRCoV HKU24 nucleocapsid but not spike polypeptide were demonstrated. Using the spike polypeptide-based Western blot assay, we showed that only Norway rats and two oriental house rats from Guangzhou, China, were infected by ChRCoV HKU24. Other rats, including Norway rats from Hong Kong, possessed antibodies only against N protein and not against the spike polypeptide, suggesting infection by βCoVs different from ChRCoV HKU24. ChRCoV HKU24 may represent the murine origin of Betacoronavirus 1, and rodents are likely an important reservoir for ancestors of lineage A βCoVs.

IMPORTANCE

While bats and birds are hosts for ancestors of most coronaviruses (CoVs), lineage A βCoVs have never been found in these animals and the origin of Betacoronavirus lineage A remains obscure. We discovered a novel lineage A βCoV, China Rattus coronavirus HKU24 (ChRCoV HKU24), from Norway rats in China with a high seroprevalence. The unique genome features and phylogenetic analysis supported the suggestion that ChRCoV HKU24 represents a novel CoV species, occupying a deep branch at the root of members of Betacoronavirus 1 and being distinct from murine coronavirus. Nevertheless, ChRCoV HKU24 possessed genome characteristics that resemble those of both Betacoronavirus 1 and murine coronavirus. Our data suggest that ChRCoV HKU24 represents the murine origin of Betacoronavirus 1, with interspecies transmission from rodents to other mammals having occurred centuries ago, before the emergence of human coronavirus (HCoV) OC43 in the late 1800s. Rodents are likely an important reservoir for ancestors of lineage A βCoVs.

RNA structure analysis of alphacoronavirus terminal genome regions

•

Review of current knowledge of cis-acting RNA elements essential to coronavirus replication.

•

Identification of RNA structural elements in alphacoronavirus terminal genome regions.

•

Discussion of intra- and intergeneric conservation of genomic cis-acting RNA elements in alpha- and betacoronaviruses.

Coronavirus genome replication is mediated by a multi-subunit protein complex that is comprised of more than a dozen virally encoded and several cellular proteins. Interactions of the viral replicase complex with cis-acting RNA elements located in the 5′ and 3′-terminal genome regions ensure the specific replication of viral RNA. Over the past years, boundaries and structures of cis-acting RNA elements required for coronavirus genome replication have been extensively characterized in betacoronaviruses and, to a lesser extent, other coronavirus genera. Here, we review our current understanding of coronavirus cis-acting elements located in the terminal genome regions and use a combination of bioinformatic and RNA structure probing studies to identify and characterize putative cis-acting RNA elements in alphacoronaviruses. The study suggests significant RNA structure conservation among members of the genus Alphacoronavirus but also across genus boundaries. Overall, the conservation pattern identified for 5′ and 3′-terminal RNA structural elements in the genomes of alpha- and betacoronaviruses is in agreement with the widely used replicase polyprotein-based classification of the Coronavirinae, suggesting co-evolution of the coronavirus replication machinery with cognate cis-acting RNA elements.

Dependence of coronavirus RNA replication on an NH2-terminal partial nonstructural protein 1 in cis

Genomes of positive (+)-strand RNA viruses use cis-acting signals to direct both translation and replication. Here we examine two 5'-proximal cis-replication signals of different character in a defective interfering (DI) RNA of the bovine coronavirus (BCoV) that map within a 322-nucleotide (nt) sequence (136 nt from the genomic 5' untranslated region and 186 nt from the nonstructural protein 1 [nsp1]-coding region) not found in the otherwise-identical nonreplicating subgenomic mRNA7 (sgmRNA7). The natural DI RNA is structurally a fusion of the two ends of the BCoV genome that results in a single open reading frame between a partial nsp1-coding region and the entire N gene. (i) In the first examination, mutation analyses of a recently discovered long-range RNA-RNA base-paired structure between the 5' untranslated region and the partial nsp1-coding region showed that it, possibly in concert with adjacent stem-loops, is a cis-acting replication signal in the (+) strand. We postulate that the higher-order structure promotes (+)-strand synthesis. (ii) In the second examination, analyses of multiple frame shifts, truncations, and point mutations within the partial nsp1-coding region showed that synthesis of a PEFP core amino acid sequence within a group A lineage betacoronavirus-conserved NH2-proximal WAPEFPWM domain is required in cis for DI RNA replication. We postulate that the nascent protein, as part of an RNA-associated translating complex, acts to direct the DI RNA to a critical site, enabling RNA replication. We suggest that these results have implications for viral genome replication and explain, in part, why coronavirus sgmRNAs fail to replicate.

IMPORTANCE

cis-Acting RNA and protein structures that regulate (+)-strand RNA virus genome synthesis are potential sites for blocking virus replication. Here we describe two: a previously suspected 5'-proximal long-range higher-order RNA structure and a novel nascent NH2-terminal protein component of nsp1 that are common among betacoronaviruses of group A lineage.

Identification of cis-acting elements on positive-strand subgenomic mRNA required for the synthesis of negative-strand counterpart in bovine coronavirus

It has been demonstrated that, in addition to genomic RNA, sgmRNA is able to serve as a template for the synthesis of the negative-strand [(−)-strand] complement. However, the cis-acting elements on the positive-strand [(+)-strand] sgmRNA required for (−)-strand sgmRNA synthesis have not yet been systematically identified. In this study, we employed real-time quantitative reverse transcription polymerase chain reaction to analyze the cis-acting elements on bovine coronavirus (BCoV) sgmRNA 7 required for the synthesis of its (−)-strand counterpart by deletion mutagenesis. The major findings are as follows. (1) Deletion of the 5'-terminal leader sequence on sgmRNA 7 decreased the synthesis of the (−)-strand sgmRNA complement. (2) Deletions of the 3' untranslated region (UTR) bulged stem-loop showed no effect on (−)-strand sgmRNA synthesis; however, deletion of the 3' UTR pseudoknot decreased the yield of (−)-strand sgmRNA. (3) Nucleotides positioned from −15 to −34 of the sgmRNA 7 3'-terminal region are required for efficient (−)-strand sgmRNA synthesis. (4) Nucleotide species at the 3'-most position (−1) of sgmRNA 7 is correlated to the efficiency of (−)-strand sgmRNA synthesis. These results together suggest, in principle, that the 5'- and 3'-terminal sequences on sgmRNA 7 harbor cis-acting elements are critical for efficient (−)-strand sgmRNA synthesis in BCoV.

The 3'-terminal 55 nucleotides of bovine coronavirus defective interfering RNA harbor cis-acting elements required for both negative- and positive-strand RNA synthesis

The synthesis of the negative-strand [(−)-strand] complement of the ∼30 kilobase, positive-strand [(+)-strand] coronaviral genome is a necessary early step for genome replication. The identification of cis-acting elements required for (−)-strand RNA synthesis in coronaviruses, however, has been hampered due to insufficiencies in the techniques used to detect the (−)-strand RNA species. Here, we employed a method of head-to-tail ligation and real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR) to detect and quantitate the synthesis of bovine coronavirus (BCoV) defective interfering (DI) RNA (−) strands. Furthermore, using the aforementioned techniques along with Northern blot assay, we specifically defined the cis-acting RNA elements within the 3′-terminal 55 nucleotides (nts) which function in the synthesis of (−)- or (+)-strand BCoV DI RNA. The major findings are as follows: (i) nts from -5 to -39 within the 3′-terminal 55 nts are the cis-acting elements responsible for (−)-strand BCoV DI RNA synthesis, (ii) nts from −3 to −34 within the 3′-terminal 55 nts are cis-acting elements required for (+)-strand BCoV DI RNA synthesis, and (iii) the nucleotide species at the 3′-most position (−1) is important, but not critical, for both (−)- and (+)-strand BCoV DI RNA synthesis. These results demonstrate that the 3′-terminal 55 nts in BCoV DI RNA harbor cis-acting RNA elements required for both (−)- and (+)-strand DI RNA synthesis and extend our knowledge on the mechanisms of coronavirus replication. The method of head-to-tail ligation and qRT-PCR employed in the study may also be applied to identify other cis-acting elements required for (−)-strand RNA synthesis in coronaviruses.

A leaderless genome identified during persistent bovine coronavirus infection is associated with attenuation of gene expression

The establishment of persistent viral infection is often associated with the selection of one or more mutant viruses. For example, it has been found that an intraleader open reading frame (ORF) in genomic and subgenomic mRNA (sgmRNA) molecules is selected during bovine coronavirus (BCoV) persistence which leads to translation attenuation of the downstream ORF. Here, we report the unexpected identification of leaderless genomes, in addition to leader-containing genomes, in a cell culture persistently infected with BCoV. The discovery was made by using a head-to-tail ligation method that examines genomic 5′-terminal sequences at different times postinfection. Functional analyses of the leaderless genomic RNA in a BCoV defective interfering (DI) RNA revealed that (1) the leaderless genome was able to serve as a template for the synthesis of negative-strand genome, although it cannot perform replicative positive-strand genomic RNA synthesis, and (2) the leaderless genome retained its function in translation and transcription, although the efficiency of these processes was impaired. Therefore, this previously unidentified leaderless genome is associated with the attenuation of genome expression. Whether the leaderless genome contributes to the establishment of persistent infection remains to be determined.

Reselection of a genomic upstream open reading frame in mouse hepatitis coronavirus 5'-untranslated-region mutants

An AUG-initiated upstream open reading frame (uORF) encoding a potential polypeptide of 3 to 13 amino acids (aa) is found within the 5' untranslated region (UTR) of >75% of coronavirus genomes based on 38 reference strains. Potential CUG-initiated uORFs are also found in many strains. The AUG-initiated uORF is presumably translated following genomic 5'-end cap-dependent ribosomal scanning, but its function is unknown. Here, in a reverse-genetics study with mouse hepatitis coronavirus, the following were observed. (i) When the uORF AUG-initiating codon was replaced with a UAG stop codon along with a U112A mutation to maintain a uORF-harboring stem-loop 4 structure, an unimpaired virus with wild-type (WT) growth kinetics was recovered. However, reversion was found at all mutated sites within five virus passages. (ii) When the uORF was fused with genomic (main) ORF1 by converting three in-frame stop codons to nonstop codons, a uORF-ORF1 fusion protein was made, and virus replicated at WT levels. However, a frameshifting G insertion at virus passage 7 established a slightly 5'-extended original uORF. (iii) When uAUG-eliminating deletions of 20, 30, or 51 nucleotides (nt) were made within stem-loop 4, viable but debilitated virus was recovered. However, a C80U mutation in the first mutant and an A77G mutation in the second appeared by passage 10, which generated alternate uORFs that correlated with restored WT growth kinetics. In vitro, the uORF-disrupting nondeletion mutants showed enhanced translation of the downstream ORF1 compared with the WT. These results together suggest that the uORF represses ORF1 translation yet plays a beneficial but nonessential role in coronavirus replication in cell culture.

Regulation of coronaviral poly(A) tail length during infection

The positive-strand coronavirus genome of ~30 kilobase in length and subgenomic (sg) mRNAs of shorter lengths, are 5’ and 3’-co-terminal by virtue of a common 5’-capped leader and a common 3’-polyadenylated untranslated region. Here, by ligating head-to-tail viral RNAs from bovine coronavirus-infected cells and sequencing across the ligated junctions, it was learned that at the time of peak viral RNA synthesis [6 hours postinfection (hpi)] the 3’ poly(A) tail on genomic and sgmRNAs is ~65 nucleotides (nt) in length. Surprisingly, this length was found to vary throughout infection from ~45 nt immediately after virus entry (at 0 to 4 hpi) to ~65 nt later on (at 6 h to 9 hpi) and from ~65 nt (at 6 h to 9 hpi) to ~30 nt (at 120-144 hpi). With the same method, poly(U) sequences of the same lengths were simultaneously found on the ligated viral negative-strand RNAs. Functional analyses of poly(A) tail length on specific viral RNA species, furthermore, revealed that translation, in vivo, of RNAs with the longer poly(A) tail was enhanced over those with the shorter poly(A). Although the mechanisms by which the tail lengths vary is unknown, experimental results together suggest that the length of the poly(A) and poly(U) tails is regulated. One potential function of regulated poly(A) tail length might be that for the coronavirus genome a longer poly(A) favors translation. The regulation of coronavirus translation by poly(A) tail length resembles that during embryonal development suggesting there may be mechanistic parallels.

Isolation and characterization of a novel Betacoronavirus subgroup A coronavirus, rabbit coronavirus HKU14, from domestic rabbits

We describe the isolation and characterization of a novel Betacoronavirus subgroup A coronavirus, rabbit coronavirus HKU14 (RbCoV HKU14), from domestic rabbits. The virus was detected in 11 (8.1%) of 136 rabbit fecal samples by reverse transcriptase PCR (RT-PCR), with a viral load of up to 10(8) copies/ml. RbCoV HKU14 was able to replicate in HRT-18G and RK13 cells with cytopathic effects. Northern blotting confirmed the production of subgenomic mRNAs coding for the HE, S, NS5a, E, M, and N proteins. Subgenomic mRNA analysis revealed a transcription regulatory sequence, 5'-UCUAAAC-3'. Phylogenetic analysis showed that RbCoV HKU14 formed a distinct branch among Betacoronavirus subgroup A coronaviruses, being most closely related to but separate from the species Betacoronavirus 1. A comparison of the conserved replicase domains showed that RbCoV HKU14 possessed <90% amino acid identities to most members of Betacoronavirus 1 in ADP-ribose 1″-phosphatase (ADRP) and nidoviral uridylate-specific endoribonuclease (NendoU), indicating that RbCoV HKU14 should represent a separate species. RbCoV HKU14 also possessed genomic features distinct from those of other Betacoronavirus subgroup A coronaviruses, including a unique NS2a region with a variable number of small open reading frames (ORFs). Recombination analysis revealed possible recombination events during the evolution of RbCoV HKU14 and members of Betacoronavirus 1, which may have occurred during cross-species transmission. Molecular clock analysis using RNA-dependent RNA polymerase (RdRp) genes dated the most recent common ancestor of RbCoV HKU14 to around 2002, suggesting that this virus has emerged relatively recently. Antibody against RbCoV was detected in 20 (67%) of 30 rabbit sera tested by an N-protein-based Western blot assay, whereas neutralizing antibody was detected in 1 of these 20 rabbits.

Genetic evidence of a long-range RNA-RNA interaction between the genomic 5' untranslated region and the nonstructural protein 1 coding region in murine and bovine coronaviruses

Higher-order RNA structures in the 5' untranslated regions (UTRs) of the mouse hepatitis coronavirus (MHV) and bovine coronavirus (BCoV), separate species in the betacoronavirus genus, appear to be largely conserved despite an ∼36% nucleotide sequence divergence. In a previous study, each of three 5'-end-proximal cis-acting stem-loop domains in the BCoV genome, I/II, III, and IV, yielded near-wild-type (wt) MHV phenotypes when used by reverse genetics to replace its counterpart in the MHV genome. Replacement with the BCoV 32-nucleotide (nt) inter-stem-loop fourth domain between stem-loops III and IV, however, required blind cell passaging for virus recovery. Here, we describe suppressor mutations within the transplanted BCoV 32-nt domain that along with appearance of potential base pairings identify an RNA-RNA interaction between this domain and a 32-nt region ∼200 nt downstream within the nonstructural protein 1 (Nsp1)-coding region. Mfold and phylogenetic covariation patterns among similarly grouped betacoronaviruses support this interaction, as does cotransplantation of the BCoV 5' UTR and its downstream base-pairing domain. Interestingly, cotransplantation of the BCoV 5' UTR and BCoV Nsp1 coding region directly yielded an MHV wt-like phenotype, which demonstrates a cognate interaction between these two BCoV regions, which in the MHV genome act in a fully interspecies-compliant manner. Surprisingly, the 30-nt inter-stem-loop domain in the MHV genome can be deleted and viral progeny, although debilitated, are still produced. These results together identify a previously undescribed long-range RNA-RNA interaction between the 5' UTR and Nsp1 coding region in MHV-like and BCoV-like betacoronaviruses that is cis acting for viral fitness but is not absolutely required for viral replication in cell culture.

Mouse hepatitis virus stem-loop 4 functions as a spacer element required to drive subgenomic RNA synthesis

The 5' 140 nucleotides of the mouse hepatitis virus (MHV) 5' untranslated region (5'UTR) are predicted to contain three secondary structures, stem-loop 1 (SL1), SL2, and SL4. SL1 and SL2 are required for subgenomic RNA synthesis. The current study focuses on SL4, which contains two base-paired regions, SL4a and SL4b. A series of reverse genetic experiments show that SL4a is not required to be base paired. Neither the structure, the sequence, nor the putative 8-amino-acid open reading frame (ORF) in SL4b is required for viral replication. Viruses containing separate deletions of SL4a and SL4b are viable. However, deletion of SL4 is lethal, and genomes carrying this deletion are defective in directing subgenomic RNA synthesis. Deletion of (131)ACA(133) just 3' to SL4 has a profound impact on viral replication. Viruses carrying the (131)ACA(133) deletion were heterogeneous in plaque size. We isolated three viruses with second-site mutations in the 5'UTR which compensated for decreased plaque sizes, delayed growth kinetics, and lower titers associated with the (131)ACA(133) deletion. The second-site mutations are predicted to change either the spacing between SL1 and SL2 or that between SL2 and SL4 or to destabilize the proximal portion of SL4a in our model. A mutant constructed by replacing SL4 with a shorter sequence-unrelated stem-loop was viable. These results suggest that the proposed SL4 in the MHV 5'UTR functions in part as a spacer element that orients SL1, SL2, and the transcriptional regulatory sequence (TRS), and this spacer function may play an important role in directing subgenomic RNA synthesis.

An optimal cis-replication stem-loop IV in the 5' untranslated region of the mouse coronavirus genome extends 16 nucleotides into open reading frame 1

The 288-nucleotide (nt) 3' untranslated region (UTR) in the genome of the bovine coronavirus (BCoV) and 339-nt 3' UTR in the severe acute respiratory syndrome (SARS) coronavirus (SCoV) can each replace the 301-nt 3' UTR in the mouse hepatitis coronavirus (MHV) for virus replication, thus demonstrating common 3' cis-replication signals. Here, we show that replacing the 209-nt MHV 5' UTR with the ∼63%-sequence-identical 210-nt BCoV 5' UTR by reverse genetics does not yield viable virus, suggesting 5' end signals are more stringent or possibly are not strictly 5' UTR confined. To identify potential smaller, 5'-common signals, each of three stem-loop (SL) signaling domains and one inter-stem-loop domain from the BCoV 5' UTR was tested by replacing its counterpart in the MHV genome. The SLI/II domain (nucleotides 1 to 84) and SLIII domain (nucleotides 85 to 141) each immediately enabled near-wild-type (wt) MHV-like progeny, thus behaving similarly to comparable 5'-proximal regions of the SCoV 5' UTR as shown by others. The inter-stem-loop domain (nt 142 to 173 between SLs III and IV) enabled small plaques only after genetic adaptation. The SLIV domain (nt 174 to 210) required a 16-nt extension into BCoV open reading frame 1 (ORF1) for apparent stabilization of a longer BCoV SLIV (nt 174 to 226) and optimal virus replication. Surprisingly, pleiomorphic SLIV structures, including a terminal loop deletion, were found among debilitated progeny from intra-SLIV chimeras. The results show the inter-stem-loop domain to be a potential novel species-specific cis-replication element and that cis-acting SLIV in the viral genome extends into ORF1 in a manner that stabilizes its lower stem and is thus not 5' UTR confined.

Subgenomic messenger RNA amplification in coronaviruses

Coronaviruses possess the largest known RNA genome, a 27- to 32-kb (+)-strand molecule that replicates in the cytoplasm. During virus replication, a 3' coterminal nested set of five to eight subgenomic (sg) mRNAs are made that are also 5' coterminal with the genome, because they carry the genomic leader as the result of discontinuous transcription at intergenic donor signals during (-)-strand synthesis when templates for sgmRNA synthesis are made. An unanswered question is whether the sgmRNAs, which appear rapidly and abundantly, undergo posttranscriptional amplification. Here, using RT-PCR and sequence analyses of head-to-tail-ligated (-) strands, we show that after transfection of an in vitro-generated marked sgmRNA into virus-infected cells, the sgmRNA, like the genome, can function as a template for (-)-strand synthesis. Furthermore, when the transfected sgmRNA contains an internally placed RNA-dependent RNA polymerase template-switching donor signal, discontinuous transcription occurs at this site, and a shorter, 3' terminally nested leader-containing sgmRNA is made, as evidenced by its leader-body junction and by the expression of a GFP gene. Thus, in principle, the longer-nested sgmRNAs in a natural infection, all of which contain potential internal template-switching donor signals, can function to increase the number of the shorter 3'-nested sgmRNAs. One predicted advantage of this behavior for coronavirus survivability is an increased chance of maintaining genome fitness in the 3' one-third of the genome via a homologous recombination between the (now independently abundant) WT sgmRNA and a defective genome.

Group-specific structural features of the 5'-proximal sequences of coronavirus genomic RNAs

Global predictions of the secondary structure of coronavirus (CoV) 5′ untranslated regions and adjacent coding sequences revealed the presence of conserved structural elements. Stem loops (SL) 1, 2, 4, and 5 were predicted in all CoVs, while the core leader transcription-regulating sequence (L-TRS) forms SL3 in only some CoVs. SL5 in group I and II CoVs, with the exception of group IIa CoVs, is characterized by the presence of a large sequence insertion capable of forming hairpins with the conserved 5′-UUYCGU-3′ loop sequence. Structure probing confirmed the existence of these hairpins in the group I Human coronavirus-229E and the group II Severe acute respiratory syndrome coronavirus (SARS-CoV). In general, the pattern of the 5′ cis-acting elements is highly related to the lineage of CoVs, including features of the conserved hairpins in SL5. The function of these conserved hairpins as a putative packaging signal is discussed.

Bovine coronavirus nonstructural protein 1 (p28) is an RNA binding protein that binds terminal genomic cis-replication elements

Nonstructural protein 1 (nsp1), a 28-kDa protein in the bovine coronavirus (BCoV) and closely related mouse hepatitis coronavirus, is the first protein cleaved from the open reading frame 1 (ORF 1) polyprotein product of genome translation. Recently, a 30-nucleotide (nt) cis-replication stem-loop VI (SLVI) has been mapped at nt 101 to 130 within a 288-nt 5'-terminal segment of the 738-nt nsp1 cistron in a BCoV defective interfering (DI) RNA. Since a similar nsp1 coding region appears in all characterized groups 1 and 2 coronavirus DI RNAs and must be translated in cis for BCoV DI RNA replication, we hypothesized that nsp1 might regulate ORF 1 expression by binding this intra-nsp1 cistronic element. Here, we (i) establish by mutation analysis that the 72-nt intracistronic SLV immediately upstream of SLVI is also a DI RNA cis-replication signal, (ii) show by gel shift and UV-cross-linking analyses that cellular proteins of approximately 60 and 100 kDa, but not viral proteins, bind SLV and SLVI, (SLV-VI) and (iii) demonstrate by gel shift analysis that nsp1 purified from Escherichia coli does not bind SLV-VI but does bind three 5' untranslated region (UTR)- and one 3' UTR-located cis-replication SLs. Notably, nsp1 specifically binds SLIII and its flanking sequences in the 5' UTR with approximately 2.5 muM affinity. Additionally, under conditions enabling expression of nsp1 from DI RNA-encoded subgenomic mRNA, DI RNA levels were greatly reduced, but there was only a slight transient reduction in viral RNA levels. These results together indicate that nsp1 is an RNA-binding protein that may function to regulate viral genome translation or replication but not by binding SLV-VI within its own coding region.

Coronavirus Genome Replication

Viruses belonging to the family Coronaviridae are unique among RNA viruses because of the unusually large size of their genome, which is of messenger- or positive- or plus-sense. It is ∼30,000 bases or 2–3 times larger than the genomes of most other RNA viruses. Coronaviruses belong to the order Nidovirales, the other three families being the Arteriviridae, Toroviridae and Roniviridae. (For a review of classification and evolutionary relatedness of Nidovirales see Gorbalenya et al. 2006.) This grouping is based on the arrangement and relatedness of open reading frames within their genomes and on the presence in infected cells of multiple subgenomic mRNAs that form a 3'-co-terminal, nested set with the genome. Among the Nidovirales, coronaviruses (and toroviruses) are unique in their possession of a helical nucleocapsid, which is unusual for plus-stranded but not minus-stranded RNA viruses; plus-stranded RNA-containing plant viruses in the Closteroviridae and in the Tobamovirus genus also possess helical capsids. Coronaviruses are very successful and have infected many species of animals, including bats, birds (poultry) and mammals, such as humans and livestock. Coronavirus species are classified into three groups, which were based originally on cross-reacting antibodies and more recently on nucleotide sequence relatedness (Gonzalez et al. 2003). There have been several reviews of coronaviruses published recently and the reader is referred to them for more extensive references (Enjuanes et al. 2006; Masters 2006; Pasternak et al. 2006; Sawicki and Sawicki 2005; Sawicki et al. 2007; Ziebuhr 2005).

5'-proximal hot spot for an inducible positive-to-negative-strand template switch by coronavirus RNA-dependent RNA polymerase

Coronaviruses have a positive-strand RNA genome and replicate through the use of a 3' nested set of subgenomic mRNAs each possessing a leader (65 to 90 nucleotides [nt] in length, depending on the viral species) identical to and derived from the genomic leader. One widely supported model for leader acquisition states that a template switch takes place during the generation of negative-strand antileader-containing templates used subsequently for subgenomic mRNA synthesis. In this process, the switch is largely driven by canonical heptameric donor sequences at intergenic sites on the genome that match an acceptor sequence at the 3' end of the genomic leader. With experimentally placed 22-nt-long donor sequences within a bovine coronavirus defective interfering (DI) RNA we have shown that matching sites occurring anywhere within a 65-nt-wide 5'-proximal genomic acceptor hot spot (nt 33 through 97) can be used for production of templates for subgenomic mRNA synthesis from the DI RNA. Here we report that with the same experimental approach, template switches can be induced in trans from an internal site in the DI RNA to the negative-strand antigenome of the helper virus. For these, a 3'-proximal 89-nt acceptor hot spot on the viral antigenome (nt 35 through 123), largely complementary to that described above, was found. Molecules resulting from these switches were not templates for subgenomic mRNA synthesis but, rather, ambisense chimeras potentially exceeding the viral genome in length. The results suggest the existence of a coronavirus 5'-proximal partially double-stranded template switch-facilitating structure of discrete width that contains both the viral genome and antigenome.