Expression of reporter genes from the defective ... - Semantic Scholar

Journal of General Virology (2000), 81, 1687–1698. Printed in Great Britain ...................................................................................................................................................................................................................................................................................

Expression of reporter genes from the defective RNA CD-61 of the coronavirus infectious bronchitis virus Kathleen Stirrups,† Kathleen Shaw, Sharon Evans, Kevin Dalton,‡ Rosa Casais, David Cavanagh and Paul Britton Division of Molecular Biology, Institute for Animal Health, Compton Laboratory, Compton, Newbury, Berkshire RG20 7NN, UK

The defective RNA (D-RNA) CD-61, derived from the Beaudette strain of the avian coronavirus infectious bronchitis virus (IBV), was used as an RNA vector for the expression of two reporter genes, luciferase and chloramphenicol acetyltransferase (CAT). D-RNAs expressing the CAT gene were demonstrated to be capable of producing CAT protein in a helper-dependent expression system to about 1n6 µg per 106 cells. The reporter genes were expressed from two different sites within the CD-61 sequence and expression was not affected by interruption of the CD-61-specific ORF. Expression of the reporter genes was under the control of a transcription-associated sequence (TAS) derived from the Beaudette gene 5, normally used for the transcription of IBV subgenomic mRNA 5. The Beaudette gene 5 TAS is composed of two tandem repeats of the IBV canonical consensus sequence involved in the acquisition of a leader sequence during the discontinuous transcription of IBV subgenomic mRNAs. It is demonstrated that only one canonical sequence is required for expression of mRNA 5 or for the expression of an mRNA from a D-RNA and that either sequence can function as an acceptor site for acquisition of the leader sequence.

Introduction Infectious bronchitis virus (IBV) is a highly infectious and contagious pathogen of chickens that replicates mainly in the respiratory tract but also in some epithelial cells of the gut, kidney and oviduct (Lambrechts et al., 1993 ; Cavanagh & Naqi, 1997). IBV, an avian coronavirus (order Nidovirales, family Coronaviridae, genus Coronavirus), is an enveloped virus that replicates in the cell cytoplasm and contains an unsegmented, 5h-capped and 3h-polyadenylated single-stranded, positivesense RNA genome of 27 608 nt (Boursnell et al., 1987). In addition to the genomic RNA, IBV produces five subgenomic mRNAs in IBV-infected cells, which possess a 64 nt leader Author for correspondence : Paul Britton. Fax j44 1635 577263. e-mail Paul.Britton!bbsrc.ac.uk † Present address : University of Cambridge, Department of Haematology, Division of Transfusion Medicine, Long Road, Cambridge CB2 2PT, UK. ‡ Present address : Departments of Pathology and Cell Biology (BML 342), Yale University School of Medicine, 310 Cedar St, New Haven, CT 06510, USA.

0001-6925 # 2000 SGM

sequence, which is also located at the 5h end of the genomic RNA, at their 5h ends. During replication, all coronaviruses produce a 3h-coterminal nested set of polycistronic subgenomic mRNAs by a discontinuous transcription mechanism (Baric et al., 1983 ; Sawicki & Sawicki, 1990, 1998). In general, only the 5h-most ORF of each coronavirus subgenomic mRNA is translated to produce the structural proteins, spike glycoprotein (S), small membrane protein (E), integral membrane protein (M) and nucleocapsid protein (N), and a number of other potential nonstructural proteins of as yet unknown function. As part of the discontinuous transcription mechanism for synthesis of the mRNAs, a common 5h-terminal leader sequence, derived from the 5h end of the genome, is attached to the body sequence of each subgenomic mRNA. Conserved sequences, which we have previously termed the transcription-associated sequence (TAS ; Hiscox et al., 1995), are present along the genomic RNA, corresponding to the sites where the subgenomic mRNAs are produced, and are involved in the acquisition of the leader sequence. The TASs are found proximal to the initiation codon of the first ORF for each particular subgenomic mRNA. However, the distance between the TAS and the AUG varies for each subgenomic mRNA. The canonical octameric IBV BGIH

K. Stirrups and others

TAS, CT(T\G)AACAA, is found 9–93 nt from the particular initiation codon of the 5h-most ORF for synthesis of the five IBV subgenomic mRNAs. The availability of complete full-length cDNA copies of RNA virus genomes that can be used for the production of infectious RNA copies has proved a powerful tool for understanding the molecular biology of the viruses and for studying the role of individual genes in pathogenesis. To date, there is no complete coronavirus cDNA available for generation of an infectious RNA. Therefore, we have been developing an alternative strategy, utilizing an IBV defective RNA (D-RNA), CD-61, that functions like a minigenome as a potential RNA vector both as an expression vector and for targetted recombination. IBV D-RNA CD-61 (Pe! nzes et al., 1994, 1996) lacks internal parts of the genome but contains the sequences required for replication and for packaging into virus particles and can therefore be replicated and packaged (rescued) in a helper virus-dependent manner.

Luciferase (Luc) and chloramphenicol acetyltransferase (CAT) gene cassettes were produced by PCR mutagenesis for insertion into D-RNA CD-61 by using oligonucleotides T7IBV5LUCSTART or T7IBV5CATSTART and SmaLUCEND or SmaCATEND at the 5h and 3h ends, respectively (Fig. 1 a). The 5h oligonucleotides included a SmaI restriction endonuclease site, a T7 promoter sequence, a PmaCI restriction endonuclease site, the Beaudette-derived gene 5 TAS and the first 20 nt of the gene. The 3h oligonucleotides consisted of the last 23 nt of the gene, including a termination codon, followed by a SmaI restriction endonuclease site. The genes were amplified from plasmid pGEMCAT\ EMC\LUC (Pause et al., 1994) (a gift from G. Belsham, Institute for Animal Health, UK) by using Pfu DNA polymerase (Stratagene). The cassettes were digested with SmaI and ligated into EcoRV-digested pBluescript SK for assessment of gene expression by using in vitro T7derived transcripts in the TNT\expression system (Promega). The TASgene regions were removed from the T7 promoter-containing cassettes by using PmaCI and SmaI and ligated into PmaCI- or SnaBI-digested pCD-61-PmaCI (Fig. 1 b, c).

Recombinant DNA techniques. Standard procedures were used for recombinant DNA techniques (Ausubel et al., 1987 ; Sambrook et al., 1989) or according to the manufacturers’ instructions.

RNA electroporation of primary CK cells. T7-derived RNAs corresponding to the various D-RNAs were synthesized in vitro from 2 µg of the corresponding NotI-linearized D-RNA-containing plasmids (Pe! nzes et al., 1996). CK cells (P ) were grown to 80–90 % confluence in ! 25 cm# tissue culture flasks (Falcon) and infected with 0n5 ml Beaudette helper virus (7i10( p.f.u.\ml) in allantoic fluid. At 8 h p.i., the cells were electroporated with the transcription reactions (Stirrups et al., 2000). Following incubation of the electroporated cells for 16 h, virus (V ) in " 1 ml of the supernatants was used to infect CK cells (P ) and, after " 20–24 h p.i., virus (V ) from the supernatants was passaged on CK cells # (P ) for up to P . # "#

Analysis of IBV-derived RNAs. Total cellular RNA was extracted from the Beaudette-infected CK cells (Pe! nzes et al., 1994) and electrophoresed in denaturing 1 % agarose–2n2 M formaldehyde gels (Sambrook et al., 1989). The RNA was Northern blotted onto Hybond-C extra 0n45 µm nitrocellulose membranes (Amersham) and IBV-specific RNAs were detected by hybridization with $#P-labelled DNA. Several different probes were used : (i) a 590 bp IBV 3h probe, to detect all IBV-derived RNAs, corresponding to nt 27017–27607 at the 3h end of IBV genome, was generated by PCR with oligonucleotides N1145 and 93\100 ; (ii) an IBV 5h probe, minus the leader sequence, to detect IBV genomic RNA and D-RNAs consisted of a 1120 bp AgeI–SphI fragment (nt 338–1458) ; (iii) a 1664 bp Luc-specific probe, to detect D-RNAs containing the Luc gene, was produced by PCR with oligonucleotides corresponding to the 5h and 3h ends of the Luc gene ; and (iv) a CAT-specific probe, to detect D-RNAs containing the CAT gene, consisted of a 305 bp MroI–NcoI fragment derived from the CAT gene. All the probes were labelled with [$#P]dCTP by using the random oligonucleotide-primed synthesis method (Feinberg & Vogelstein, 1983).

Production of gene cassettes. A unique PmaCI site within the IBV D-RNA CD-61 was initially chosen for insertion of genes. The PmaCI site is within domain III of CD-61 but not within the D-RNAspecific ORF (Pe! nzes et al., 1994, 1996). However, the pZSL1190 sequence of pCD-61 (Pe! nzes et al., 1996), into which the CD-61 cDNA was inserted, contained a PmaCI site within the multiple cloning site. A 72 nt PstI–SfiI fragment containing the PmaCI site was removed from pCD-61 and the two ends were converted to blunt ends and ligated together. A resultant plasmid, pCD-61-PmaCI, was sequenced and found to contain an 81 nt deletion and was subsequently used for the insertion of genes. A second unique site, SnaBI, which was contained in pCD-61PmaCI but within domain II of CD-61 (Fig. 1 b) and which interrupts the D-RNA-specific ORF, was also used.

Analysis of reporter gene activities. CK cells (approximately 2i10') were disrupted in cell medium and centrifuged at 2500 r.p.m. The pelleted cells were resuspended in 1 ml PBSa (Stirrups et al., 2000), of which 0n5 ml was used for reporter gene assay and 0n5 ml for RNA extraction. For the luciferase assay, the resuspended cells were centrifuged at 2500 r.p.m. and lysed with 0n5 ml lysis buffer (Promega). Fifty µl of the cell extract was added to 50 µl luciferase assay reagent (Promega) and analysed in a luminometer (Labtech, model Jade 1253). For the CAT assay, the resuspended cells were washed three times in PBSa, lysed in 1 ml lysis buffer (Boehringer Mannheim) and incubated at room temperature for 30 min. CAT protein was detected by ELISA (Boehringer Mannheim, product no. 1363727). Serial dilutions of the cell extracts were made and the amount of CAT protein present in the cell

Methods

Virus and cells. IBV strains Beaudette, M41, H120 (Darbyshire et al., 1979), D207 (Davelaar et al., 1984) and B1648 (Meulemans et al., 1987) were grown in 11-day-old embryonated domestic fowl eggs at 37 mC and harvested from allantoic fluid at 24 h p.i. Beaudette was used as helper virus for the rescue of IBV CD-61-based D-RNAs (Pe! nzes et al., 1996). Virions from M41, H120, D207 and B1648 were used for preparation of genomic RNA. Beaudette, M41 and H120 belong to the same serotype, Massachusetts, according to either serology of the S protein (Darbyshire et al., 1979) or S gene sequence identity determined from Beaudette (Binns et al., 1985 ; Boursnell et al., 1987), M41 (Binns et al., 1986 ; Niesters et al., 1986) and H120 (Kusters et al., 1989). D207 is of a different antigenic serotype from the Massachusetts strains, according to serology of the S protein (Davelaar et al., 1984) and S gene sequence (Kusters et al., 1989). B1648 is a nephropathogenic IBV strain that is of a different antigenic serotype from both the Massachusetts strains and D207-type viruses, according to serology of the S protein and S gene sequence (Shaw et al., 1996). The passage and growth of IBV in chick kidney (CK) cells was done as described previously (Pe! nzes et al., 1994).

Oligonucleotides. The oligonucleotides used in this study were obtained from MWG-Biotech or Pharmacia and are listed in Table 1.

BGII

A D-RNA of IBV that expresses reporter genes

Table 1. Oligonucleotides used for RT–PCR and sequencing Underlined residues represent non-IBV sequence. Residues in italics represent T7 promoter sequences and reporter gene initiation codons and residues in bold represent IBV canonical consensus TASs. Nucleotides in lower case represent substitutions used to mutate IBV-derived sequences. The positions of the nucleotides in the Beaudette sequence (Boursnell et al., 1987) are given. , Not applicable. Oligonucleotide 94\155 43 93\106 93\136 MEND3j TAS5aj TAS5bj TAS5cj TAS5ck TAS5ak NSTART2 NSTARTk 41 N1145 93\100 CATINTk T7IBV5LUCSTART* SmaLUCEND* T7IBV5CATSTART* SmaCATEND* T7IBV5CATSTARTGGG* IBV5CATScrambled* IBV5CAT-TAS-1-Scr* IBV5CAT-TAS-2-Scr*

Sequence GAAGGATCCATTAATACGACTCACTATAGGGACTTAAGATAGAT ATTAATATAT GGGCCCACTTAAGATAGATATTAATATA GGCAGAAGTTTGACCGTAG GTCCCATTTTAGCCAACATG GTAGATACTGGCGAGCTAG TACTACGAAGGAACACCAG TTCCAAAAAGGTTGTTGTAG GATGTGGTCCAATTATAAG TAACTGCTCTTCCAAAAC GCGTAGTAGTCCGTGATC GTTAAGAAAGTAAACACAATC GCTTGCCATGACAAAAGATT GGAACAGGACCTGCCGC AGAGGAACAATGCACAGCTGG CAGGATATCGCTCTAACTCTATACTAGCCT GTAACAAGGGTGAACACTAT TCCCCCGGGGGATTATTAATACGACTCACTATAGGGCACGTGTT TTACTTAACAAAAACTTAACAAATACGGACGATG GAAGACGCCAAAAAC ACCCCCGGGGGATTACAATTTGGACTTTCCGCCC TCCCCCGGGGGATTATTAATACGACTCACTATAGGGCACGTGTT TTACTTAACAAAAACTTAACAAATACGGACGATG GAGAAAAAAATCACTGG ACCCCCGGGGGTTACGCCCCGCCCTGCCACTCAT TCCCCCGGGGGATTATTAATACGACTCACTATAGGGCACGTGTT TTACTTAACAAAAACTTAACAAATACGGACG gggGAGAAAAAAATCACTGG TCCCCCGGGGGATTATTAATACGACTCACTATAGGGCACGTGTT TTtacgtgatcgtgaactcgtgtTACGGACGATG GAGAAAAAAATCACTGG TCCCCCGGGGGATTATTAATACGACTCACTATAGGGCACGTGTT TTtacgtgatcgtACTTAACAAATACGGACGATG GAGAAAAAAATCACTGG TCCCCCGGGGGATTATTAATACGACTCACTATAGGGCACGTGTT TTACTTAACAAAAgaactcgtgtTACGGACGATG GAGAAAAAAATCACTGG

Position

Polarity

1–23

j

1–22 674–692 12857–12876 25126–25144 25350–25368 25371–25390 25391–25409 25504–25521 25581–25598 25757–25777 25862–25881 26161–26177 27017–27037 27587–27607  

j j k j j j j k k k k j j k k j

 

k j

 

k j



j



j



j

* Oligonucleotides for generation of reporter gene cassettes had a SmaI site included at the 5h end for cloning purposes.

supernatants was determined by comparison with standard amounts of CAT protein.

Sequence analysis of D-RNA-derived mRNAs and IBV gene 5 TASs. Total cellular RNA was extracted from P CK cells infected with ' a CAT-containing D-RNA. RT–PCR was used to amplify the 5h ends of the CAT mRNAs by using oligonucleotides 43 and CATINTk and Pfu polymerase. RT–PCR products were sequenced directly or cloned into SmaI-digested pTarget (Promega), from which the D-RNA-derived cDNA was sequenced. D-RNA-derived cDNAs from the cloned PCR products were analysed by A-track sequencing. RT–PCR products amplified from virion RNA, derived from IBV strains H120, M41, B1648 and D207, were used to determine the genomic sequences corresponding

to the gene 5 TAS region by using oligonucleotides MEND3j and NSTARTk.

Results Insertion of genes into IBV D-RNA CD-61

The expression of coronavirus genes under the control of a coronavirus RNA-dependent RNA polymerase requires the synthesis of subgenomic mRNAs. Synthesis of coronavirus mRNAs is dependent on TASs found along the genomic RNA. Expression of a heterologous gene from a coronavirus D-RNA BGIJ

K. Stirrups and others

(a)

(b)

(c)

compared (Table 2). The TAS responsible for the synthesis of mRNA 5 was chosen for the expression of heterologous genes from IBV D-RNA CD-61 because it had the shortest sequence between the 3h end of the TAS and the AUG of ORF 5a. IBV Beaudette mRNA 5 is one of the abundantly expressed mRNAs, indicating that the associated TAS is efficient for the synthesis of an mRNA. However, the Beaudette gene 5 TAS has two canonical consensus sequences, CTTAACAA, in a tandem repeat (CTTAACAAAAACTTAACAA), and it was not known whether one or other or both are utilized during transcription of mRNA 5. Therefore, for the expression of genes from CD-61, both canonical sequences were retained. The genomic sequence between the 3h end of the second canonical sequence (TAS-2), the gene and the 8 nt proximal to the first canonical sequence (TAS-1) were kept identical to those on the IBV Beaudette genome (Fig. 1 a). Initially the Luc gene and, subsequently, the CAT gene, under the control of the Beaudette-derived gene 5 TAS, were inserted into the D-RNA sequence in pCD-61-PmaCI either in the PmaCI site in domain III or the SnaBI site in domain II. The resulting constructs are outlined in Fig. 1 (c). D-RNAs containing the CAT gene inserted into the PmaCI site of CD-61, either with the TASs scrambled or containing the gene 5 TAS but with the AUG initiation codon mutated to GGG, were generated by using oligonucleotides IBV5CATScrambled or T7IBV5CATSTARTGGG, respectively. Expression of genes by helper virus-dependent rescue of IBV D-RNAs

Fig. 1. Schematic diagram outlining the production of reporter gene cassettes for insertion into IBV D-RNA CD-61. (a) Generation of reporter gene cassettes by PCR mutagenesis. Oligonucleotides T7IBV5LUCSTART and SmaLUCEND or T7IBV5CATSTART and SmaCATEND were used for the Luc and CAT gene cassettes, respectively, to generate the gene cassettes by PCR mutagenesis. The resulting PCR products contained a T7 promoter, a PmaCI site for future manipulation, the IBV Beaudette gene 5 TAS and the appropriate reporter gene. The sequence from GTG of the PmaCI site to the ATG of the reporter gene was identical to the sequence in the Beaudette genome for ORF 5a. The cassettes were cloned into pBluescript and in vitro T7-transcripts were assessed for the expression of the reporter gene. (b) Structure of IBV CD-61 showing the positions of the two cloning sites. (c) D-RNAs after insertion of the reporter genes. The three domains I, II and III of CD-61 derived from different regions of the IBV genome (Pe! nzes et al., 1994, 1996) are indicated as patterned boxes. The position of the nucleotide deletion, A-749, characteristic of CD-61 (Pe! nzes et al., 1994) is indicated in (b). The D-RNA-specific ORFs are shown as thick lines. The reporter gene cassettes were removed from pBluescript by digestion with PmaCI and SmaI and inserted into the D-RNA CD-61 sequence in PmaCI- or SnaBI-digested pCD-61-PmaCI. The expected sizes of the D-RNA-specific mRNAs are shown. The Luccontaining D-RNAs were 7n8 kb and the CAT containing D-RNAs were 6n8 kb.

also requires a TAS proximal to the gene for the synthesis of an mRNA from the D-RNA. The nucleotides adjacent to the TAS and the distance between the 3h end of the TAS and the AUG of the adjacent ORF have been suggested to play a role in the transcription of mRNAs. The IBV TASs, responsible for generation of the five Beaudette-derived mRNAs, were BGJA

In vitro T7-transcribed RNAs corresponding to D-RNAs in which the Luc gene had been inserted into either the PmaCI or SnaBI sites of CD-61 were electroporated into Beaudetteinfected CK cells. Luciferase activity was detected in cell lysates from the electroporated cells (P ) when compared with ! controls in which the Luc-containing D-RNAs were absent (Fig. 2), indicating that the D-RNAs were replicated and that a Luc mRNA was transcribed from the D-RNAs. However, serial passage of the D-RNAs rescued by helper virus resulted in lower luciferase activities, which declined after P (Fig. 2). % Northern blot analyses with IBV 5h and 3h probes were carried out on total RNA isolated from the P –P cells. No RNA of " ( 7n9 kb, corresponding to a CD-61–Luc D-RNA, was detected (data not shown). However, other RNAs of unknown origin, not detectable with the Luc-specific probe, were detected in addition to the IBV helper virus-derived RNAs. The amounts and sizes of the extraneous D-RNAs varied with different rescue experiments. Although luciferase activity was dependent on the presence of IBV helper virus and the spatial position of the Luc gene within the D-RNA did not affect expression, we concluded that the Luc-containing D-RNAs were inherently unstable. Whether the instability was due to the Luc gene sequence specifically, the presence of any heterologous sequence or expression of the luciferase protein was not known.

A D-RNA of IBV that expresses reporter genes

Table 2. Comparison of canonical IBV TASs

Gene Polymerase* Spike Gene 3 Membrane Gene 5† Nucleoprotein

Canonical TAS

No. of nucleotides from TAS 3h end to ATG of first ORF

CTTAACAA CTGAACAA CTGAACAA CTTAACAA CTTAACAAAAACTTAACAA CTTAACAA

465 52 23 77 9 93

* No specific mRNA is produced for expression of the IBV polymerase gene. The TAS on the genomic RNA is involved in leader sequence acquisition during the discontinuous transcription of the subgenomic RNAs. † The tandem TASs are shown in bold, TAS-1 and TAS-2 (from the left).

Fig. 2. Detection of luciferase activity in CK cell supernatants after serial passage (P1–P7) of IBV and Luc-containing D-RNAs after electroporation of in vitro T7-transcribed D-RNA into Beaudette-infected CK cells (P0). Luciferase activities were detected from cells that contained D-RNA CD-61 with the Luc gene, under the control of the Beaudette gene 5 TAS, inserted into either the PmaCI site (filled bars) or SnaBI site (hatched bars). Each value represents the mean of duplicate electroporations ; similar profiles were obtained in two separate experiments. No luciferase activity was detected in cells infected with IBV helper virus containing D-RNA CD61. RLU, Relative light units.

In order to investigate further the potential for expressing heterologous genes from CD-61, we decided to use the CAT gene in place of the Luc gene. D-RNAs with the TAS–CAT sequence inserted in the anti-sense orientation, a scrambled gene 5 TAS sequence upstream of the CAT gene and a CAT gene in which the AUG was mutated to GGG were also used. Expression of CAT protein, following rescue of D-RNAs with the CAT gene inserted in either the PmaCI or SnaBI sites of CD-61, was again dependent on the presence of IBV helper virus, a functional IBV TAS and a complete 3h UTR. Analysis of the amounts of CAT protein detected in cells following rescue of the D-RNAs in passages after P routinely showed a ! drop in the amount of CAT protein at P , followed by a 6-fold " increase in the amount of CAT protein detected at P \P & ' compared with that observed at P (Fig. 3 a). However, the ! amount of CAT detected routinely decreased on further

passage. Similar patterns of CAT expression were observed whether the CAT gene was inserted in the PmaCI site or the SnaBI site of CD-61 (Fig. 3 a). The expression of CAT protein was routinely observed over a higher passage number, up to P in one experiment, compared with expression of luciferase "# activity and the amount of CAT protein produced was observed to be as high as 1n6 µg from 10' cells (Fig. 3 b). It should be noted that the highest levels of CAT expression were observed following electroporation of the T7 D-RNA transcripts into Beaudette-infected Vero cells (P ) followed by ! serial passage of the D-RNAs in CK cells. No expression of CAT protein was observed from D-RNAs either with the TAS–CAT sequence inserted in the anti-sense orientation or under the control of a scrambled TAS. There was a minimal amount of CAT protein detected following rescue of the DRNA containing the CAT gene with a mutated initiation codon, possibly resulting from detection of a truncated CAT protein expressed from a downstream in-frame AUG. Northern blot analyses, with the IBV 3h or 5h probe or a CAT-specific probe, detected an RNA species of 6n9 kb, corresponding to the size of the in vitro T7-transcribed D-RNA transcripts (Fig. 3 c). Detection of the 6n9 kb RNA by both IBV probes showed that the RNA was an IBV D-RNA and detection with the CAT-specific probe indicated that the DRNA contained the CAT gene. The probes also detected the CAT-containing D-RNA with the TAS–CAT sequence in the anti-sense orientation (data not shown). At least one extra RNA, of unknown origin, was identified in addition to the 6n9 kb D-RNA (Fig. 3 c), and was detected with the CAT probe (data not shown). These observations indicated that the smaller RNAs were D-RNAs generated after the loss of part of the CAT gene. The observation that the CAT-containing D-RNAs were detected whereas the Luc-containing D-RNAs were not detected and the increased expression of CAT upon serial passage compared with the amount of CAT produced at P ! indicated that the CAT-containing D-RNAs were more stable BGJB

K. Stirrups and others (a)

(c)

(b)

Fig. 3. Detection of CAT protein by ELISA in cell supernatants after passage of IBV and CAT-containing D-RNAs after electroporation of in vitro T7-transcribed D-RNA into Beaudette-infected cells (P0). (a) Detection of CAT protein from CK cells infected with Beaudette containing the D-RNA with the CAT gene, under the control of the Beaudette gene 5 TAS, inserted into either the PmaCI site (filled bars) or SnaBI site (hatched bars) of CD-61. No CAT protein was detected in cells infected with IBV helper virus containing D-RNA CD-61. (b) Comparison of CAT protein production in CK cells (P1–P6) infected with IBV containing the D-RNA CD-61–CAT with CAT inserted in the PmaCI site after electroporation of in vitro T7-transcribed D-RNA initially into either CK cells (P0) (open bars) or Vero cells (P0) (shaded bars). The virus plus D-RNA from these P0 cells was then passaged serially on CK cells. The Beaudette strain of IBV is able to infect and grow on Vero cells. (c) Northern blot analysis of total RNA isolated from cells after serial passage (P0–P6) of CD-61–CAT in the presence of helper IBV. The CAT gene was inserted in the PmaCI site of the D-RNA. The RNA was isolated after electroporation of either IBV-infected Vero cells with subsequent passage (P1–P6) on CK cells (VERO) or IBV-infected CK cells with subsequent passage on CK cells (CKC). The IBV-derived RNAs were detected by using the IBV 3h probe, used to detect all IBV-derived RNAs. The positions of the IBV subgenomic mRNAs 2 (7n3 kb), 3 (3n9 kb), 4 (3n3 kb), 5 (2n5 kb) and 6 (2n1 kb), are indicated. The RNA corresponding to CD-61–CAT (6n9 kb) is also indicated. A smaller RNA (6n1 kb) of unknown origin was observed routinely from P4. No RNA of 1n9 kb, corresponding to the D-RNA-derived CAT mRNA, was detected. The RNAs detected between mRNAs 4 and 5 are observed routinely for all strains of IBV, as identified originally by Stern & Kennedy (1980), and are of unknown origin.

than the Luc-containing D-RNAs. No RNAs, of 5n1 kb or 1n9 kb depending on the site of the CAT gene in CD-61 (Fig. 1 c), corresponding to the potential mRNAs transcribed from the D-RNAs were detected by Northern blot analysis (Fig. 3 c). RT–PCR analysis of RNA extracted from P cells con' taining CD-61 with the CAT gene inserted into the PmaCI site, with oligonucleotides 93\136 and 93\106, resulted in a 910 bp product. This confirmed the presence of a D-RNA with the CD-61-specific domain I\II junction (Pe! nzes et al., 1996). RT–PCR analysis of the P RNA extracts with a CAT-specific ' oligonucleotide, CATINTk, and an IBV leader-specific oligonucleotide, 94\155, generated a 353 bp product, indicative of a CAT-specific mRNA containing the IBV leader sequence. Detection of CAT-specific mRNAs was only successful from rescue experiments that yielded the highest levels of CAT protein, supporting the results of Northern blot analysis, which BGJC

suggested that the amounts of the CAT-specific mRNAs were very low. Analysis of CAT expression from D-RNAs containing modified TASs

Both of the canonical TASs within the Beaudette gene 5 TAS were retained for expression of genes from CD-61 because it was not known whether one or other or both sequences are utilized for mRNA synthesis. Therefore, a series of gene cassettes with the CAT gene under the control of one TAS with the other sequence scrambled was produced for insertion into CD-61 to determine whether one sequence was sufficient or used preferentially or whether both can be utilized. The D-RNAs were constructed as before except that oligonucleotides IBV5CAT-TAS-1-Scr, IBV5CAT-TAS-2-Scr and

A D-RNA of IBV that expresses reporter genes (a)

(b)

other sequence and that there was no observable advantage of two canonical sequences. In order to confirm that RNA isolated from the P –P cells $ ) contained D-RNAs, two separate RT–PCRs were carried out following rescue of D-RNAs CD-61–T\T–CAT, CD-61– ST\T–CAT, CD-61–T\ST–CAT and CD-61–ST\ST–CAT. The first RT–PCR used oligonucleotides 93\106 and 93\136, which produced a product of 910 bp, confirming the presence of an RNA containing the CD-61 domain I\II-specific junction. The second RT–PCR used oligonucleotides 41 and CATINTk. The former anneals proximal to the PmaCI site in CD-61 and the latter corresponds to a sequence within the CAT gene, to confirm the presence of a CAT-containing DRNA. A product of 400 bp was produced, confirming the presence of an RNA containing the CAT gene within the PmaCI site of a CD-61-derived D-RNA. Sequence analysis of mRNAs transcribed from CD61–CAT D-RNAs

Fig. 4. Analysis of CAT protein production by IBV D-RNAs containing modified TASs. (a) Modified regions of the Beaudette gene 5 TASs in the various D-RNAs. The boxed regions represent the IBV canonical sequences involved in leader sequence acquisition by the mRNAs. The nucleotides in lower-case represent the modified bases. The top sequence, T/T, represents the gene 5 TAS present on the Beaudette genome. The ATG of the CAT gene is underlined. ST designates that the canonical sequence was scrambled. The modified TAS–CAT cassettes were inserted into the PmaCI site of CD-61. (b) Detection of CAT protein by ELISA in CK cell supernatants after passage of IBV and the D-RNAs containing the CAT gene under the control of the unmodified and modified Beaudette gene 5 TASs, T/T (open bars), ST/T (hatched bars), T/ST (filled bars) and ST/ST (shaded bars). Each value represents the mean of duplicate electroporations ; similar profiles were obtained in three separate experiments. No CAT protein was detectable in cell supernatants analysed from cells infected with D-RNA CD-61-ST/ST-CAT.

IBV5CATScrambled replaced oligonucleotide T7IBV5CATStart for generation of the gene cassettes. The resulting gene cassettes (Fig. 4 a) were initially cloned into pBluescript to assess expression of the CAT gene. The TAS–CAT cassettes containing the modified gene 5 TASs ST\T, T\ST and ST\ST, in which the first canonical consensus sequence (TAS-1), the second canonical sequence (TAS-2) or both sequences scrambled, respectively, were inserted into the PmaCI site in pCD-61-PmaCI. The TAS–CAT cassette containing both TASs was designated T\T. Sequence analysis confirmed that the sequences incorporated into the D-RNAs were as expected. In vitro T7-transcribed D-RNAs containing the modified TASs were electroporated into Beaudette-infected CK cells and the D-RNAs were serially passaged. Total cellular RNA was isolated and cell supernatants were taken from P –P cells for ! "! RT–PCR analysis and CAT protein ELISA. The amounts of CAT protein detected in the cell supernatants are shown in Fig. 4 (b). Similar amounts of CAT protein were detected following rescue of D-RNAs with one TAS or both TASs (Fig. 4 b). This indicated that both canonical sequences, TAS-1 and TAS-2, were utilized equally for mRNA synthesis in the absence of the

RT–PCR with oligonucleotides 43 and CATINTk was used to analyse the 5h ends of the CAT mRNAs transcribed from CD-61–ST\T–CAT, CD-61–T\ST–CAT and CD-61– T\T–CAT. Oligonucleotide 43 corresponded to the 5h end of the IBV leader sequence. The RT–PCRs were expected to generate either a 342 or 353 bp product, depending on the canonical TAS used for leader sequence acquisition. The RT–PCR products were sequenced directly and after cloning to determine whether there was heterogeneity in the use of the two TASs from CD-61–T\T–CAT. Sequence analysis of mRNAs transcribed from the D-RNAs showed that leader sequence acquisition occurred on the unmodified TASs in CD61–ST\T–CAT and CD-61–T\ST–CAT (Fig. 5 a, b). The sequence between the TAS and the CAT AUG from the mRNA transcribed from CD-61–T\ST–CAT was 11 nt, representing scrambled TAS-2, longer than the corresponding sequence on the mRNA transcribed from CD-61–ST\T–CAT (Fig. 5 a). No nucleotide substitutions were found between the 11 nt sequence determined from the mRNA and the sequence present on the input D-RNA, indicating that the heterologous sequence did not appear to be detrimental to transcription of the CAT mRNA from CD-61–T\ST–CAT. Leader acquisition by the mRNA transcribed from CD61–T\T–CAT could potentially have occurred at either of the two TASs. Sequence analysis of the RT–PCR products derived from the mRNA transcribed from CD-61–T\T–CAT showed that the 342 bp RT–PCR product was generated preferentially (Fig. 5 a), indicating that the mRNA was predominantly transcribed from the TAS-2 canonical sequence, the same TAS that was utilized for transcription of the mRNA from CD61–ST\T–CAT. The RT–PCR products derived from the mRNA transcribed from CD-61–T\T–CAT were cloned and 49 resultant plasmids were analysed by A-track sequencing. None of the 49 cloned RT–PCR products contained the two TASs, indicating either that TAS-1 was not utilized or that the BGJD

K. Stirrups and others (a)

(b)

Fig. 5. Analysis of the D-RNA-derived mRNAs to determine the site of leader sequence acquisition. (a) Sequence analysis of RT–PCR products derived from mRNAs transcribed from the D-RNAs, isolated from passage P6 CK cells, expressing CAT protein under control of the modified Beaudette gene 5 TASs. Two sequencing reactions were carried out per RT–PCR. The radioactive band at the top of the gel represents the complete RT–PCR product and the sequence immediately below is the 5h end of the leader sequence. The CAT-derived sequence is at the bottom of the gel. The IBV canonical consensus sequence, CTTAACAA, is indicated, as is the scrambled sequence in D-RNA CD-61–T/TS–CAT. (b) Sequence comparisons of the RT–PCR products

BGJE

A D-RNA of IBV that expresses reporter genes

Fig. 6. Sequence comparison of the gene 5 TASs derived from seven strains of IBV. The two Beaudette gene 5 canonical consensus sequences, TAS-1 and TAS-2, are boxed and the canonical sequences of the other strains are double-underlined, as are the initiation codons corresponding to IBV ORF 5a, the first ORF of gene 5. The underlined nucleotides represent substitutions within the different IBV strains when compared with the Beaudette sequence.

frequency of use of TAS-1 for leader acquisition was very low. Interestingly, a number of mutations were observed in the RT–PCR products, resulting in the addition of two extra adenosine nucleotides in the A-rich region at the 5h end of the CAT gene, interrupting the CAT ORF. Sequence analysis of IBV gene 5 TASs

Expression of the CAT gene from IBV D-RNAs using the Beaudette gene 5 TAS or modified versions of the sequence showed that both canonical consensus sequences present in the Beaudette gene 5 TAS were functional. However, when both sequences were present, the 3h canonical sequence, TAS-2, was used for leader sequence acquisition. In addition to the Beaudette sequence, only two other IBV gene 5 sequences have been determined : CU-T2 (Jia & Naqi, 1997) and KB8523 (Sutou et al., 1988). Comparison of the gene 5 sequences from these two strains with the Beaudette sequence showed that CU-T2 and KB8523 only had the canonical sequence equivalent to Beaudette TAS-2. The sequence equivalent to the Beaudette gene 5 TAS-1 site in CU-T2 and KB8523 contained two nucleotide substitutions. In order to investigate whether other strains of IBV contained one or two canonical TASs for gene 5, we determined the gene 5 TASs from four other IBV strains, H120, M41, B1648 and D207. RT–PCRs were carried out on genomic RNA with oligonucleotides MEND3j and NSTARTk, which corresponded to sequences in the 3h end of the M gene and the 5h end of the N gene, respectively. RT–PCR products of the expected size, 755 bp, were obtained from the four genomic RNAs and sequenced directly by using oligonucleotides TAS5aj, TAS5ak, TAS5bj, TAS5cj and TAS5ck. Comparison of the gene 5 TASs determined from the seven IBV strains (Fig. 6) showed that all of the strains

except Beaudette contained a single TAS, equivalent to Beaudette TAS-2. The region of the genomic RNAs corresponding to the Beaudette TAS-1 site contained two or three nucleotide substitutions. The observation that TAS-2 of Beaudette gene 5 was used preferentially for mRNA transcription and that Beaudette is the only strain of IBV identified to date to have two TASs for gene 5 led us to investigate which sequence is utilized for transcription of Beaudette subgenomic mRNA 5. The 5h end of Beaudette mRNA 5 was amplified by RT–PCR from total cellular RNA isolated from Beaudette-infected CK cells by using oligonucleotides 43 and NSTART2. The potential products of the RT–PCRs were 374 or 363 bp, depending on whether leader acquisition occurred on TAS-1 or TAS-2. The RT–PCR products were sequenced directly and after cloning. Sequence analysis of the RT–PCR products showed that the TAS-2 site was utilized for acquisition of leader sequence during transcription of mRNA 5. RT–PCR products from 22 clones were sequenced, 10 completely and 12 by A-track sequencing. Analysis of the sequences showed that 21 were derived from an mRNA 5 transcribed from TAS-2 and one from an mRNA 5 transcribed from TAS-1. Our results showed that the Beaudette gene 5 TAS-2 canonical sequence is used preferentially for leader sequence acquisition. However, the TAS-1 site can function as a leader sequence acquisition site in the absence of the TAS-2 site.

Discussion This study reports for the first time expression of heterologous genes from an IBV D-RNA under the control of an IBV TAS. Expression of the heterologous genes was demonstrated to occur from two insertion sites within D-RNA CD-61. One site (SnaBI) was within domain II of CD-61 and interrupted the D-RNA-specific ORF. The second site (PmaCI) was within domain III of CD-61 and not within the D-RNAspecific ORF or within the 3h UTR (Fig. 1). No observable differences were observed in expression of the two genes at either site, supporting our previous results that interruption of the IBV D-RNA-specific ORF does not affect replication (Pe! nzes et al., 1996). The overall pattern of heterologous gene expression from the four IBV D-RNAs was the same in CK cells (Figs 2 and 3 a, b). Expression of the CAT gene appeared to be more stable than that of the Luc gene. Other coronavirus-derived D-RNAs have been used for the expression of heterologous genes. D-RNAs based on the D-

derived from mRNAs transcribed from the D-RNAs. The top line shows the complete IBV Beaudette leader sequence with the leader junction site, equivalent to a TAS, underlined. The middle panel shows the sequences proximal to the CAT gene in the four D-RNAs cDNA constructs. The Beaudette gene 5 TASs are boxed and the canonical consensus sequence is underlined. The modified nucleotides are in lower case. The ATG of the CAT gene is double-underlined. The lower panel shows part of the 5h-end sequence of mRNAs transcribed from the three D-RNAs expressing CAT protein. The canonical sequence of the TAS used for leader sequence acquisition is boxed and the ATG of the CAT gene is double-underlined. The sequence underlined with dots proximal to the TAS corresponds to the 3h end of the IBV leader sequence. Nucleotides in lower case represent the modified bases and account for the 11 nt insertion in the mRNA transcribed from D-RNA CD-61–T/ST–CAT.

BGJF

K. Stirrups and others

RNA DIssE (Makino & Lai, 1989) from murine hepatitis virus (MHV) have been used to express CAT (Liao & Lai, 1994), MHV haemagglutinin–esterase (Liao et al., 1995 ; Lai et al., 1997 ; Zhang et al., 1998) and murine IFN-γ (Lai et al., 1997 ; Zhang et al., 1997). The D-RNA DI-C from the porcine coronavirus transmissible gastroenteritis virus (TGEV) has been used to express β-glucuronidase (GUS ; Izeta et al., 1999). Expression of genes from MHV D-RNAs was not stable ; CAT activity was not observed beyond P (Liao & Lai, 1994), # expression of MHV haemagglutinin–esterase beyond P (Liao $ et al., 1995) and expression of murine IFN-γ beyond P (Zhang % et al., 1997). Expression of GUS in the TGEV D-RNA system was also unstable, although expression of GUS was detected up to P in some cases. Expression of CAT appeared to be "! more stable than expression of Luc in the IBV system, but even expression of CAT decreased after P . The observation that ' coronavirus D-RNAs expressing heterologous genes are relatively unstable could result from several factors. The most likely explanation may be the presence of non-coronavirusderived sequences in the D-RNAs. This may have several effects, ranging from some fundamental interference on replication or packaging of the D-RNAs to the natural tendency of D-RNAs to evolve by removal of unnecessary sequences. An effect on packaging is harder to explain, as gene expression was observed after serial passage of the IBV and TGEV D-RNAs. The D-RNAs were sufficiently packaged during passages P –P to infect cells on subsequent passage. " ' The IBV system was demonstrated to be capable of producing CAT protein to levels greater than 1n6 µg per 10' cells and to be capable of producing CAT protein up to P . "# The TGEV D-RNA system produced GUS protein to about 1n0 µg per 10' cells. The authors reported that this could be increased 5–10-fold by using specific transcription regulatory sequences and that the system was capable of expressing GUS up to P (Izeta et al., 1999). The TGEV D-RNA expression "! system used a two-step amplification system, in which the DRNA was expressed initially in P cells by Pol II transcription ! of a transfected cDNA under the control of a cytomegalovirus promoter in the cell nucleus, followed by TGEV helperdependent replication of the D-RNA in the cytoplasm. Studies to investigate coronavirus transcription have used either MHV (Makino et al., 1991 ; Joo & Makino, 1992, 1995 ; Makino & Joo, 1993 ; van der Most et al., 1994 ; van Marle et al., 1995) or bovine coronavirus (BCoV ; Krishnan et al., 1996) D-RNAs containing one or more TASs for the expression of mRNAs. Insertion of two TASs within an MHV D-RNA resulted in decreased transcription of the larger mRNA if the two TASs were 23 nt apart. The inhibition decreased as the distance between the two TASs was increased, resulting in equal amounts of the mRNAs if the TASs were separated by 124 nt (Joo & Makino, 1995). The authors concluded that the TAS for the smaller mRNA had some inhibitory effect on the upstream TAS. Further studies on MHV D-RNAs containing combinations of up to three TASs separated by 361–761 nt BGJG

showed that the position of the TASs affected the amounts of the mRNAs produced (van Marle et al., 1995). The largest mRNA (the 5h-most TAS) was produced in the smallest amount whether it was the only TAS in the D-RNA or it was in combination with one or more TASs. The middle TAS [site B of van Marle et al. (1995)] produced the largest amount of mRNA whether alone or in conjunction with one or more TASs. The observation that the downstream TASs attenuated upstream TASs but not vice versa led the authors to conclude that their results were consistent with the transcription model proposed by Sawicki & Sawicki (1990). Studies on coronavirus transcription using a BCoV D-RNA showed that transcription occurred preferentially at the 3h-most TAS after insertion of either a duplicate or triplicate 27 nt tandem repeat sequence containing the BCoV canonical TAS within a D-RNA (Krishnan et al., 1996). The BCoV heptameric canonical TASs, UCUAAAC, were separated by 20 nt in the tandem repeats. The IBV TAS derived from gene 5 of the Beaudette strain of IBV, used in this study, naturally contains a tandem repeat of the octameric IBV canonical TAS, CUUAACAA, separated by 3 nt. We observed that both TASs were functional for the production of a translationally active mRNA in the absence of the other sequence. Analysis of the mRNA produced from a DRNA containing the CAT gene under the control of the gene 5 TAS showed that the downstream or 3h-most TAS (TAS-2) was used preferentially for transcription of the CAT mRNA. Analysis of subgenomic mRNA 5 derived from gene 5 in Beaudette-infected cells confirmed that TAS-2 was the preferential site for mRNA 5 transcription. Analysis of the gene 5 sequences from other IBV strains showed that only one canonical TAS was present, corresponding to the Beaudette TAS-2. Our results are consistent with the observation from both the MHV and BCoV experiments that, if two or more canonical TASs are present in close proximity, the 3h-most sequence is used preferentially. The observation that the TAS-2 site of Beaudette gene 5 is used preferentially is in agreement with the conclusions of van Marle et al. (1995) and consistent with the coronavirus transcription model of Sawicki & Sawicki (1990). This can be explained if the polymerase terminates preferentially at TAS-2 for synthesis of mRNA 5 and if read-through for synthesis from TAS-1 is rare. Presumably, if both TASs acted as efficient terminators for mRNA 5 synthesis, this would have a detrimental effect on the synthesis of the longer mRNAs. This work was supported by the Ministry of Agriculture, Fisheries and Food, UK (project code OD1905) and by grant number CT950064 of the Fourth RTD Framework Programme of the European Commission. K. Stirrups and K. Dalton were the holders of Research Studentships from the Biotechnology and Biological Sciences Research Council (BBSRC). S. Evans was supported by a BBSRC Realising Our Potential Award. R. Casais was the recipient of an EU TMR Marie Curie Research Training Grant. We would like to thank Dr G. Belsham, IAH Pirbright Laboratory, for providing plasmid pGEMCAT\EMC\LUC.

A D-RNA of IBV that expresses reporter genes

References Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A. & Struhl, K. (editors) (1987). Current Protocols in Molecular

Biology. New York : John Wiley. Baric, R. S., Stohlman, S. A. & Lai, M. M. C. (1983). Characterization of

replicative intermediate RNA of mouse hepatitis virus : presence of leader RNA sequences on nascent chains. Journal of Virology 48, 633–640. Binns, M. M., Boursnell, M. E. G., Cavanagh, D., Pappin, D. J. C. & Brown, T. D. K. (1985). Cloning and sequencing of the gene encoding

the spike protein of the coronavirus IBV. Journal of General Virology 66, 719–726. Binns, M. M., Boursnell, M. E. G., Tomley, F. M. & Brown, T. D. K. (1986). Comparison of the spike precursor sequences of coronavirus IBV

strains M41 and 6\82 with that of IBV Beaudette. Journal of General Virology 67, 2825–2831. Boursnell, M. E. G., Brown, T. D. K., Foulds, I. J., Green, P. F., Tomley, F. M. & Binns, M. M. (1987). Completion of the sequence of the genome

of the coronavirus avian infectious bronchitis virus. Journal of General Virology 68, 57–77. Cavanagh, D. & Naqi, S. (1997). Infectious bronchitis. In Diseases of Poultry, 10th edn, pp. 511–526. Edited by B. W. Calnek, H. J. Barnes, C. W. Beard, W. M. Reid & H. W. Yoda. Ames, IA : Iowa State University Press. Darbyshire, J. H., Rowell, J. G., Cook, J. K. A. & Peters, R. W. (1979).

Taxonomic studies on strains of avian infectious bronchitis virus using neutralisation tests in tracheal organ cultures. Archives of Virology 61, 227–238. Davelaar, F. G., Kouwenhoven, B. & Burger, A. G. (1984). Occurrence and significance of infectious bronchitis virus variant strains in egg and broiler production in the Netherlands. Veterinary Quarterly 6, 114–120. Feinberg, A. P. & Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Analytical Biochemistry 132, 6–13. Hiscox, J. A., Mawditt, K. L., Cavanagh, D. & Britton, P. (1995).

Investigation of the control of coronavirus subgenomic mRNA transcription by using T7-generated negative-sense RNA transcripts. Journal of Virology 69, 6219–6227. Izeta, A., Smerdou, C., Alonso, S., Pe! nzes, Z., Mendez, A., PlanaDura! n, J. & Enjuanes, L. (1999). Replication and packaging of transmissible gastroenteritis coronavirus-derived synthetic minigenomes. Journal of Virology 73, 1535–1545. Jia, W. & Naqi, S. A. (1997). Sequence analysis of gene 3, gene 4 and gene 5 of avian infectious bronchitis virus strain CU-T2. Gene 189, 189–193. Joo, M. & Makino, S. (1992). Mutagenic analysis of the coronavirus intergenic consensus sequence. Journal of Virology 66, 6330–6337. Joo, M. & Makino, S. (1995). The effect of two closely inserted transcription consensus sequences on coronavirus transcription. Journal of Virology 69, 272–280. Krishnan, R., Chang, R. Y. & Brian, D. A. (1996). Tandem placement of a coronavirus promoter results in enhanced mRNA synthesis from the downstream-most initiation site. Virology 218, 400–405. Kusters, J. G., Niesters, H. G. M., Lenstra, J. A., Horzinek, M. C. & van der Zeijst, B. A. M. (1989). Phylogeny of antigenic variants of avian

coronavirus IBV. Virology 169, 217–221. Lai, M. M., Zhang, X., Hinton, D. & Stohlman, S. (1997). Modulation

of mouse hepatitis virus infection by defective-interfering RNA-mediated expression of viral proteins and cytokines. Journal of Neurovirology 3 (Suppl. 1), S33–S34.

Lambrechts, C., Pensaert, M. & Ducatelle, R. (1993). Challenge

experiments to evaluate cross-protection induced at the trachea and kidney level by vaccine strains and Belgian nephropathogenic isolates of avian infectious bronchitis virus. Avian Pathology 22, 577–590. Liao, C.-L. & Lai, M. M. C. (1994). Requirement of the 5h-end genomic sequence as an upstream cis-acting element for coronavirus subgenomic mRNA transcription. Journal of Virology 68, 4727–4737. Liao, C.-L., Zhang, X. & Lai, M. M. C. (1995). Coronavirus defectiveinterfering RNA as an expression vector : the generation of a pseudorecombinant mouse hepatitis virus expressing hemagglutinin–esterase. Virology 208, 319–327. Makino, S. & Joo, M. (1993). Effect of intergenic consensus sequence flanking sequences on coronavirus transcription. Journal of Virology 67, 3304–3311. Makino, S. & Lai, M. M. C. (1989). High-frequency leader sequence switching during coronavirus defective interfering RNA replication. Journal of Virology 63, 5285–5292. Makino, S., Joo, M. & Makino, J. K. (1991). A system for study of coronavirus mRNA synthesis : a regulated, expressed subgenomic defective interfering RNA results from intergenic site insertion. Journal of Virology 65, 6031–6041. Meulemans, G., Carlier, M. C., Gonze, M., Petit, P. & Vandenbroeck, M. (1987). Incidence, characterisation and prophylaxis of nephropathogenic

avian infectious bronchitis viruses. Veterinary Record 120, 205–206. Niesters, H. G. M., Lenstra, J. A., Spaan, W. J. M., Zijderveld, A. J., Bleumink-Pluym, N. M. C., Hong, F., van Scharrenburg, G. J. M., Horzinek, M. C. & van der Zeijst, B. A. M. (1986). The peplomer

protein sequence of the M41 strain of coronavirus IBV and its comparison with Beaudette strains. Virus Research 5, 253–263. Pause, A., Belsham, G. J., Gingras, A. C., Donze, O., Lin, T. A., Lawrence, J. C., Jr & Sonenberg, N. (1994). Insulin-dependent

stimulation of protein synthesis by phosphorylation of a regulator of 5hcap function. Nature 371, 762–767. Pe! nzes, Z., Tibbles, K., Shaw, K., Britton, P., Brown, T. D. K. & Cavanagh, D. (1994). Characterization of a replicating and packaged defective RNA of avian coronavirus infectious bronchitis virus. Virology 203, 286–293. Pe! nzes, Z., Wroe, C., Brown, T. D., Britton, P. & Cavanagh, D. (1996). Replication and packaging of coronavirus infectious bronchitis virus defective RNAs lacking a long open reading frame. Journal of Virology 70, 8660–8668. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning. A Laboratory Manual, 2nd edn. Cold Spring Harbor, NY : Cold Spring Harbor Laboratory. Sawicki, S. G. & Sawicki, D. L. (1990). Coronavirus transcription : subgenomic mouse hepatitis virus replicative intermediates function in RNA synthesis. Journal of Virology 64, 1050–1056. Sawicki, S. G. & Sawicki, D. L. (1998). A new model for coronavirus transcription. Advances in Experimental Medicine and Biology 440, 215–219. Shaw, K., Britton, P. & Cavanagh, D. (1996). Sequence of the spike protein of the Belgian B1648 isolate of nephropathogenic infectious bronchitis virus. Avian Pathology 25, 607–611. Stern, D. F. & Kennedy, S. I. T. (1980). Coronavirus multiplication strategy. I. Identification and characterization of virus-specific RNA. Journal of Virology 34, 665–674. Stirrups, K., Shaw, K., Evans, S., Dalton, K., Cavanagh, D. & Britton, P. (2000). Leader switching occurs during the rescue of defective RNAs by

heterologous strains of the coronavirus infectious bronchitis virus. Journal of General Virology 81, 791–801. BGJH

K. Stirrups and others Sutou, S., Sato, S., Okabe, T., Nakai, M. & Sasaki, N. (1988). Cloning

and sequencing of genes encoding structural proteins of avian infectious bronchitis virus. Virology 165, 589–595. van der Most, R. G., de Groot, R. J. & Spaan, W. J. M. (1994).

Subgenomic RNA synthesis directed by a synthetic defective interfering RNA of mouse hepatitis virus : a study of coronavirus transcription initiation. Journal of Virology 68, 3656–3666. van Marle, G., Luytjes, W., van der Most, R. G., van der Straaten, T. & Spaan, W. J. (1995). Regulation of coronavirus mRNA transcription.

Journal of Virology 69, 7851–7856.

BGJI

Zhang, X., Hinton, D. R., Cua, D. J., Stohlman, S. A. & Lai, M. M. (1997). Expression of interferon-gamma by a coronavirus defective-

interfering RNA vector and its effect on viral replication, spread, and pathogenicity. Virology 233, 327–338. Zhang, X., Hinton, D. R., Park, S., Parra, B., Liao, C.-L., Lai, M. M. & Stohlman, S. A. (1998). Expression of hemagglutinin\esterase by a

mouse hepatitis virus coronavirus defective-interfering RNA alters viral pathogenesis. Virology 242, 170–183. Received 13 January 2000 ; Accepted 16 March 2000