Overexpression of the M2-2 Protein of Respiratory
http://www.100md.com
病菌学杂志 2005年第22期
MedImmune Vaccines Inc., 297 North Bernardo Ave., Mountain View, California 94043
ABSTRACT
The M2-2 protein of respiratory syncytial virus (RSV) is involved in regulation of viral RNA transcription and replication. Encoded by the next-to-last gene of RSV, the M2-2 open reading frame (ORF) overlaps with the upstream M2-1 ORF, suggesting that the production of the M2-2 protein might be tightly regulated during virus replication. To evaluate the effect of M2-2 overexpression on RSV replication, the M2-2 gene was separated from M2-1 by leaving it at the position prior to the M2-1 or moving it to the promoter proximal position as an independent transcriptional unit in the RSV A2 genome. Although recombinant viruses bearing the shuffled M2-2 gene were recovered and expressed higher levels of M2-2, most of these viruses grew poorly in HEp-2 cells. Sequence analysis revealed that various mutations (substitution, insertion, and deletion) occurred in the M2-2 gene, resulting in reduced M2-2 activity as measured by the RSV minigenome system. Further examination of the M2-2 sequence and its function showed that either one of the first two AUG codons located at the 5' end of M2-2 could be used to produce a functional M2-2 protein and that deletion of the first six amino acids from its N terminus or four amino acids from its C terminus greatly reduced its function. The effect of M2-2 protein on RSV replication was also studied by examining RSV replication in cells transiently expressing M2-2. The M2-2 protein expressed at a high level completely inhibited RSV replication. These results strongly suggested that the level of the M2-2 protein produced in the infected cells is critical to RSV replication.
INTRODUCTION
Respiratory syncytial virus (RSV) is a member of the Pneumovirus genus of the Paramyxoviridae family. It is the leading viral cause of serious lower-respiratory tract infection in infants and children worldwide (12, 15). The single-strand negative-sense RNA genome of the RSV A2 strain is 15,222 nucleotides (nt) in length and has 10 transcriptional units. Each transcriptional unit encodes a single protein, with the exception of the M2 gene that encodes two proteins, M2-1 and M2-2. As with all nonsegmented RNA viruses, RNA synthesis requires a genomic RNA encapsidated with nucleocapsid (N) protein and the virus-encoded components of the RNA polymerase, the large (L) polymerase protein and the phosphoprotein (P) (5, 22, 41). Each transcriptional unit is flanked by the gene-start (GS) sequence required for initiation of transcription and the gene-end (GE) sequence that directs polyadenylation and release of mRNA (24, 25, 31). The intergenic region (IGR), which varies in sequence and length, affects efficient transcription termination and downstream gene expression (32). Transcription begins at the 3' end of the genome and follows a linear, sequential, and polar gradient that requires termination of the upstream gene prior to initiating transcription of the downstream gene (16). The gene located at the 3' promoter proximal position is most abundantly transcribed, whereas the one most distal to the 3' promoter is expressed the least (16).
The M2 gene is unique to the genus Pneumovirus. It has two overlapping open reading frames (ORFs), M2-1 and M2-2 (14, 17), both involved in the viral RNA synthesis process. The processivity and antitermination functions of M2-1 are essential to RSV replication (13, 19, 20). In contrast, the M2-2 protein inhibits RSV minigenome transcription and replication and is dispensable for RSV replication (6, 27). However, recombinant virus lacking the M2-2 ORF grew poorly in HEp-2 cells and was attenuated in rodents and primates (11, 27, 40). The levels of the antigenomic and genomic RNA were significantly reduced compared to mRNA when the M2-2 gene was removed, suggesting that M2-2 was involved in regulating the switch between the viral RNA transcription and replication process (6, 27).
Three AUG codons are present at the 5' end of the M2-2 at amino acid positions 1, 3, and 7. Depending on the usage of the three AUG codons, the M2-2 gene overlaps with the C-terminal portion of M2-1 by 29, 23, or 11 nt, respectively. A previous study indicated that M2-2 could be initiated from any one of the three AUG codons by the ribosomes departing from the M2-1 ORF through a ribosomal termination-dependent reinitiation mechanism (3). A later study showed that disruption of the overlapping sequence between the ORFs of M2-1 and M2-2 did not affect initiation of M2-2 (37). However, the next-to-last gene location of M2 in the RSV genome and the initiation mechanism of M2-2 by ribosome reverse translocation (3) imply that the M2-2 protein is produced at a low level and is likely required during the late stage of virus infection (2).
To determine the effect of overproduction of M2-2 on virus replication, the M2-2 ORF was disentangled from M2-1 by moving it as an independent transcriptional unit to the position proximal to the M2-1 or proximal to the NS1 gene. It was expected that the M2-2 positioned at the first gene location would be expressed at a level much higher than that positioned at the ninth location and that the effect of the level of M2-2 protein on RSV replication could thus be evaluated. The recombinant viruses with the M2-2 ORF shuffled to these two different locations were produced, and their replication in cell culture was examined. In most circumstances, recombinant M2-2-shuffling viruses accumulated mutations in the M2-2 gene that lowered M2-2 function. In addition, it was found that the M2-2 protein supplied in trans inhibited virus replication, suggesting that the level of the M2-2 protein is critical to RSV replication.
MATERIALS AND METHODS
Cells and viruses. Monolayer cultures of HEp-2 and Vero cells obtained from the American Tissue Culture Collection were maintained in minimal essential medium (MEM) containing 5% fetal bovine serum (FBS). The modified vaccinia virus Ankara (MVA-T7) expressing bacteriophage T7 RNA polymerase was obtained from Bernard Moss and grown in CEK cells. The BSR T7/5 (9) cell line was maintained in Glasgow MEM (Invitrogen, Carlsbad, CA) supplemented with 10% FBS, 10% tryptose phosphate broth, and 0.05 mg/ml geneticin.
Construction of antigenomic cDNA and recovery of recombinant RSV. The M2-2 gene was inserted as an independent transcriptional unit into the RSV A2 antigenomic cDNA from which the M2-2 gene was deleted, yielding pA2M2-2 (27). To insert the M2-2 gene into the promoter proximal location, a KpnI restriction site was introduced at nt 97 upstream of the NS1 initiation site in the pET-X/A subclone that contained the RSV sequence of nt 1 to 2128 in the pET3 vector. The M2-2 gene was amplified by a 5' M2-2 gene-specific primer containing the KpnI site (underlined) and the M2-2 gene sequence (bold) starting from the second ATG (5'-GATCGGTACCATGCCAAAAATAATGATACTACC-3') and a 3' M2-2 gene-specific primer that contained the KpnI site (underlined), the GE sequence (italic and lowercase), the GS sequence (lowercase), and the 3' M2-2 sequence (bold) (5'-GACTGGTACCtatttgccccAGAttttaaataactgTTATGACACTAATATATATATTG-3'). The M2-2 gene cassette was inserted into the KpnI site in pET-X/A. The XmaI-AvrII fragment that contained the inserted M2-2 gene was then cloned into pA2M2-2 antigenomic cDNA (Fig. 1). The cDNA construct that had a short intergenic sequence of 3 nt between the M2-2 and NS1 gene junction was designated as pM2-2G1S. To ensure efficient transcription termination of the inserted M2-2 gene, the GE sequence with an additional A residue and the intergenic sequence from the SH/G junction were inserted between the M2-2 and NS1. The M2-2 GE sequence in pM2-2G1S was abolished by site-directed mutagenesis to create an EcoRI site in the pET-X/A subclone and inserted with the SH-G IGR amplified by PCR with the following primers: 5'-GTACGAATTCCATTCATCAATCCAACAGCCC-3' and 5'-GCATGAATTCAATGTTATTGTTAGTCTTG-3' (EcoRI site is underlined). Thus, pM2-2G1L contained 133 nt between the stop codon of M2-2 and its GE sequence and 51 nt from its GE sequence to the GS sequence of NS1.
Another antigenomic cDNA clone was constructed to move the M2-2 gene to the ninth position prior to the M2-1 gene. The pET-S/B subclone containing the RSV sequence from SacI (RSV nt 4477) to BamHI (RSV nt 8498) was modified to introduce a BglII site in the F/M2-1 IGR at nt 7554. The pair of primers used to amplify the M2-2 ORF were 5'-GCTCAGATCTggggcaaatATGCCAAAAATAATGATGCTGCC and 5'-GACTAGATCTtttaaataactgTTATGACACTAATAT ATATATTG (the BglII sites are underlined, the GS and GE sequences are in lowercase, and the M2-2 sequences are in bold). The M2-2 gene cassette included the BglII sites, the GS signal, the M2-2 ORF, and the GE signal. TheSacI-BamHI fragment containing the M2-2 BglII fragment inserted into the BglII site at nt 7554 was cloned into the antigenomic cDNA of pA2M2-2, and the plasmid was designated pM2-2G9. The IGRs between F and M2-2 and between M2-2 and M2-1 were 7 nt and 44 nt, respectively.
Recovering of recombinant viruses. Recovery of infectious viruses from RSV antigenomic cDNA was performed as described previously (28). Briefly, antigenomic cDNA and supporting plasmids encoding the N, P, L, and M2-1 genes, each under the control of the T7 promoter, were cotransfected into MVA-T7-infected HEp-2 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). The culture medium was replaced after overnight incubation at 35°C, and 3 days later the supernatant from the transfected cells was passaged onto fresh Vero cells to amplify any rescued viruses. Infectious viruses recovered from the transfected cells were plaque purified three times and amplified in Vero cells. Virus recovered from pM2-2G1S, pM2-2G1L, and pM2-2G9 was designated as G1S, G1L, and G9, respectively.
Replication of M2-2 recombinant viruses in cell cultures. Replication of M2-2 recombinant viruses was analyzed in HEp-2 and Vero cells. Cells grown in six-well plates were infected with virus at a multiplicity of infection (MOI) of 0.2. After a 1-h adsorption at room temperature, the infected cells were washed three times with phosphate-buffered saline, the medium was replaced with 2 ml of OptiMEM (Invitrogen, Carlsbad, CA), and the culture was incubated at 35°C in an incubator containing 5% CO2. At various times postinfection, 250 μl of culture supernatant was collected and stored at –80°C until virus titration. Each aliquot taken was replaced with an equal volume of fresh medium. The virus titer was determined by plaque assay on Vero cells, using an overlay of 1% methylcellulose and Eagle's MEM/L15 medium (JRB Biosciences, Lenexas, Kansas) containing 2% FBS. The plaques were enumerated by immunostaining using goat anti-RSV polyclonal antibody (Biogenesis, Sandown, NH).
Northern blot analysis. For Northern blot analysis, total cellular RNA was extracted from virus-infected HEp-2 or Vero cells using Trizol reagent (Invitrogen, Carlsbad, CA), followed by an additional round of phenol-chloroform extraction and ethanol precipitation. RNA was electrophoresed on 1.5% agarose gel containing 0.44 M formaldehyde and transferred to a nylon membrane (Amersham Pharmacia Biotech, Piscataway, NJ). The membrane was hybridized with a gene-specific riboprobe labeled with digoxigenin. The hybridized RNA bands were visualized with a Dig-Luminescent Detection Kit for Nucleic Acids (Roche Diagnostics Corportions, Indianapolis, IN).
Western blot analysis and immunoprecipitation. For Western blotting analysis, Vero or HEp-2 cells were infected with virus at an MOI of 1.0 and incubated at 35°C for 48 h. The infected cells were lysed in protein lysis buffer, and the cells lysates were electrophoresed on 10 to 20% Tris glycine gels (Invitrogen, Carlsbad, CA). The proteins were transferred to a nylon membrane using a semidry transfer cell (Bio-Rad, Hercules, CA). The membrane was blotted with anti-N monoclonal antibody (gift of Jose Melero), rabbit antiserum against NS1 protein, or guinea pig antiserum against M2-1 protein (gift of Jeyesh Meanger). The membrane was then incubated with a secondary antibody conjugated with horseradish peroxidase, and the protein bands were visualized after incubation with the ECL substrate (Amersham Pharmacia Biotech, Piscataway, NJ). For immunoprecipitation, the cells were infected with virus at an MOI of 1.0 and labeled with 100 μCi/ml of [35S]cysteine and [35S]methionine (35S-promix; Amersham Pharmacia Biotech, Piscataway, NJ) at 16 h postinfection for 4 h. The labeled cell monolayers were lysed in radioimmunoprecipitation assay buffer (0.15 M NaCl, 1% Na-deoxycholate, 1% Triton X-100, 0.1% sodium dodecyl sulfate [SDS], and 0.01 M Tris, pH 7.4), and the polypeptides were immunoprecipitated with an appropriate antibody, electrophoresed on a 17.5% polyacrylamide gel containing 0.1% SDS and 4 M urea or a 4 to 20% polyacrylamide gel containing 0.1% SDS (Invitrogen, Carlsbad, CA), and detected by autoradiography.
Sequence analysis of the M2-2 gene from the recovered M2-2 recombinant viruses. The M2-2 gene was amplified by reverse transcription-PCR (RT-PCR) using the primers flanking the M2-2 gene and viral RNA extracted with a viral RNA extraction mini kit (QIAGEN, Valencia, CA). For G1 variants, the M2-2 gene was amplified with V1964 primer (nt 1 to 18 of the antigenomic sense) and V1950 (nt 649 to 622 of the genomic sense). For G9, the M2-2 gene was amplified with V1959 (nt 7200 to 7229 of the antigenomic sense) and V1963 (nt7710 to 7691 of the genomic sense). The amplified M2-2 cDNAs were examined by DNA sequencing. To determine the mixed sequence populations, the RT-PCR-amplified M2-2 cDNAs were also cloned into the pPCR2.1 vector (Invitrogen, Carlsbad, CA) for sequence analysis.
Functional analysis of M2-2 cDNA by a minigenome assay. To examine the M2-2 function by a minigenome assay, the M2-2 cDNA amplified from viral RNA was cloned into the pCITE2a vector (Novagen, Madison, WI) under the control of the T7 promoter. Mutagenesis of pCITE2a-M2-2 was performed using a QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) for evaluation of M2-2 sequence variations and its in vitro function. The inhibition effect of the M2-2 protein on RSV minigenome transcription and protein expression was analyzed essentially as previously described (37). Briefly, duplicate wells of HEp-2 cells in six-well plates were infected with MVA-T7 at an MOI of 1.0 and transfected with plasmid encoding the N, P, and L genes and the pRSVCAT minigenome (0.2, 0.2, 0.1, and 0.4 μg per well, respectively), together with 0.2 μg of wild-type M2-2 or an M2-2 variant, using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Two days after transfection, the cell lysates were prepared and assayed for chloramphenicol acetyltransferase activity (Roche Applied Science, Indianapolis, IN). The relative activity of each M2-2 mutant was expressed as the percentage of activity relative to that of wild-type M2-2, and the data obtained were averaged from a minimum of three experiments.
Replication of RSV in cells transiently expressing M2-2. The effect of the M2-2 protein expressed transiently on RSV infection was examined by immunofluorescence. The M2-2 gene in pCITE2a vector was fused with the ICP4 sequence (5'-GATGAATATGATGATGCAGCTGATGCAGCAGGAGATCGA GCACCAGGA-3') (26) at its C terminus. The ICP4 tag allowed differentiation of the transiently expressed M2-2 from virus-encoded M2-2 produced in the infected cells. The BSR T7/5 cells in 24-well plates were transfected with 2 μg of wild-type M2-2 or the influenza A/AA/6/60 NP plasmid (29) using Lipofectamine 2000. After incubation at 37°C for 3 h, the transfected cells were infected with rA2 at an MOI of 10.0, followed by incubation at 35°C for 24 h. The cell monolayers were fixed with 1% paraformaldehyde and permeabilized with 0.2% Triton X-100. The cells were immunostained with rabbit polyclonal antibody against the NS1 protein of RSV (30) and mouse monoclonal antibody against ICP4 (Rumbaugh-Goodwin, Plantation, FL) or mouse monoclonal antibody against the influenza virus NP protein (Biodesign, Saco, Maine). The cells were then incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit antibodies to detect RSV infected cells and Texas Red-conjugated goat anti-mouse to detect ICP4- or NP-expressing cells. The cell images were captured by an ORCA-100 digital camera attached to the fluorescence microscope.
RESULTS
Construction and recovery of recombinant M2-2-shuffled viruses. To examine the effect of the increased expression of M2-2 on RSV replication, the M2-2 gene was separated from its overlapping M2-1 gene and moved as an independent transcriptional unit into two different locations in order to increase the quantity of RNA transcription and possibly protein expression. The F-M2 IGR was selected to insert the M2-2 gene (pM2-2G9) in order to have its transcriptional level similar to M2-1. The second location for the M2-2 gene insertion was the 3' promoter proximal region that would allow maximal expression of the M2-2 protein by the recombinant virus. To minimize the effect of the inserted M2-2 gene on the expression of the NS1 gene and subsequent protein expression, two antigenomic cDNA clones were constructed to contain a short (pM2-2G1S) or long (pM2-2G1L) intergenic sequence between the inserted M2-2 gene and the downstream NS1 gene (Fig. 1). The M2-2 gene transcriptional unit in G1S has an inefficient GE sequence compared to the GE sequence that has five A residues (24, 36). These three M2-2 recombinant viruses were recovered from cDNA using a previously described reverse genetics system (28). At least two different virus stocks obtained from several independent rescue experiments were produced. The recombinant viruses obtained from pM2-2G1S were designated G1S-A, G1S-B, G1S-C and G1S-D; the one from pM2-2G1L was designated G1L-A; and those from pM2-2G9 were designated G9-A, G9-B, and G9-C.
Replication of the M2-2 recombinant viruses in tissue culture. Replication of the M2-2 recombinant viruses was compared to rA2 and a recombinant lacking M2-2, rA2M2-2, in HEp-2 and Vero cells (Fig. 2). As previously observed (28), the growth kinetics of rA2M2-2 in Vero cells was slower than rA2 but eventually reached a peak titer only slightly lower than rA2. In contrast, replication of rA2M2-2 in HEp-2 cells was approximately 100-fold lower than rA2. Except for G9-A that replicated similar to rA2, the other three M2-2 recombinants with shuffled M2-2 replicated less well than rA2, especially in the HEp-2 cells. G1S-C replicated similarly to rA2M2-2 in both cell lines; G9-C and G1L-A had intermediate growth kinetics, and their peak titers in HEp-2 cells were much lower than rA2 in HEp-2 cells. Therefore, shuffling of the M2-2 gene to either the first or the ninth position affected RSV replication in tissue culture cells differently.
Effect of M2-2 shuffling on downstream viral gene expression. The effect of the shuffled M2-2 gene on viral RNA and protein synthesis was examined in both HEp-2 and Vero cells and was shown to have similar patterns in both cell lines. The antigenomic RNA produced in G1S-C-infected cells was greatly reduced, and its level was similar to rA2M2-2; antigenomic RNA levels produced in G9-C- and G1L-A-infected cells were moderately reduced, with a decrease of approximately two- to threefold (Fig. 3A). G9-A had antigenomic RNA slightly higher than rA2. This result suggested that G1S-C might have completely lost the M2-2 function, whereas G9-C and G1L-A might have partially lost M2-2 function. To examine the impact of the inserted M2-2 on the downstream gene expression, levels of NS1 and M2-1 RNAs were examined by Northern blotting. As predicted, there was a significant amount of M2-2/NS1 readthrough RNA produced in G1S-infected cells, resulting in reduction of the NS1 mRNA. The amount of NS1 mRNA produced in G1L-A infected cells was similar to rA2. The M2-1 mRNA in G9-A- and G9-C-infected cells was slightly reduced, and a readthrough M2-2/M2-1 RNA was detected. Insertion of M2-2 at the first position (G1S and G1L) had no effect on the synthesis of M2-1. The level of the N mRNA produced by the recombinant viruses was comparable to rA2. Viral proteins produced in the recombinant virus-infected cells were examined by Western blotting analysis (Fig. 3B). The amount of the N protein was equivalent among different viruses. Consistent with the Northern blotting result, the NS1 protein was reduced in G1S-C-infected cells. Both G9-A and G9-C had reduced M2-1 protein expression.
Aberrant expression of the M2-2 proteins in M2-2 recombinant virus-infected cells. There was an overall higher level of M2-2 mRNA produced in cells infected with the viruses that had the M2-2 inserted at the first gene position compared to those at the ninth position (Fig. 4A). A high level of M2-2/NS1 and M2-2/NS1/NS2 readthrough RNA was produced in G1S-infected cells. G1L that had the GE sequence and the intergenic sequence derived from SH/G inserted between the M2-2 and NS1 junction terminated M2-2 mRNA transcription efficiently, and very little M2-2/NS1 readthrough RNA was detected. Thus, the amount of M2-2 mRNA produced by G1L was higher than that of the G1S viruses. The M2-2 mRNA produced by the G9 viruses was higher than rA2 but slightly lower than most of the G1S viruses. These results confirmed that the virus containing an independent M2-2 transcriptional unit synthesized higher levels of mRNA; in addition, comparison of G1L and G1S demonstrated that the GE sequence and the length of the intergenic sequence affected the gene expression level.
The M2-2 proteins produced in the M2-2 recombinant virus-infected cells were examined by immunoprecipitation using anti-M2-2 antibody (Fig. 4B). Only G9-A (lane 6) produced a protein that migrated identically to that of rA2 and at a level slightly higher than that of rA2; the M2-2 polypeptides detected in other M2-2 recombinant virus-infected cells were heterogeneous in size and amount. G1L-A produced a high level of a single M2-2 band that was smaller than wild-type M2-2 (Fig. 4B). No M2-2 protein was detected in G1S-C-infected cells. Multiple M2-2 bands were detected in cells infected with G1S-A, G1S-B, G1S-D, G9-B, and G9-C. These data indicated that the M2-2 protein produced by most of the M2-2 recombinant viruses did not have the correct protein size, which most likely resulted from different sequence changes in the M2-2 ORF.
Sequence variations of the M2-2 gene in M2-2 recombinant viruses. The M2-2 gene sequences of all the obtained M2-2 recombinant viruses were examined by sequencing of the cDNAs amplified from the viral RNA. Consistent with the previous protein analysis result, only G9-A had the correct M2-2 sequence; all the other viruses had sequence changes clustered in three different regions (Table 1). The first region was A/U rich and located between the second and the third AUG as found in M2-2 of G1L-A and G1S-A. G1L-A had a single A nucleotide inserted at position 16, and thus M2-2 could only be translated from the third initiation codon, resulting in a protein smaller than M2-2 (Fig. 4B, lane 1). G1S-A had nucleotide U-17 replaced by A, resulting in a substitution of Ile-6 by Lys. The second mutation region was at nt 100 of the M2-2 gene as shown for G1S-C, which had an A nucleotide inserted at nt 100 that resulted in the frameshift of the M2-2 ORF. Therefore, the M2-2 protein was not detected in G1S-C-infected cells (Fig. 4B, lane 4). The third highly mutated region occurred in an A/U-rich region at nt 213 to 216. The sequence downstream of nt 213 contained mixed sequences for G1S-B, G1S-D, G9-B, and G9-C, indicating that these viruses contained mixed virus populations despite three rounds of virus plaque purification. G1S-B and G1S-D had 1 nt inserted at position 213, and G9-B contained 3 or 6 nt inserted at position 213; G9-C had three to six U or A nucleotides inserted at the same position. Insertion of 1 or 2 nt at position 213 would cause the frameshift of M2-2 ORF and result in the smallest-sized M2-2 protein produced in cells infected with G1S-B, G1S-D, G9-B, and G9-C (Fig. 4B, lanes 3, 5, 7, and 8).
To eliminate the possibility that the inserted M2-2 might cause mutations in the other RSV genomes, the NS2, N, P, and M2-1 genes were sequenced from all the obtained M2-2 recombinant viruses. No mutations were identified in these genes, indicating that the mutation that occurred in the M2-2 was the result of the M2-2 overexpression instead of its ability to cause an increased mutation rate in the RSV genome.
The effect of M2-2 gene sequence variations on M2-2 function. To further determine the nature of the mixed mutations in the M2-2 gene of the recombinant viruses, the M2-2 cDNAs amplified from the viral RNAs of M2-2 recombinant viruses were cloned into the pPCR2.1 vector, and approximately 10 cDNA clones from each RT-PCR DNA product were sequenced. A few of the representative M2-2 cDNAs were further cloned into the pCITE2a vector such that the M2-2 gene was under the control of the T7 promoter. The functions of these M2-2 genes were analyzed using the minigenome assay as described previously (38). Consistent with the sequencing result using RT-PCR DNA as a template, G1L-A had a nucleotide inserted at nt 16, and the protein could only be made from the third AUG codon, resulting in a shortened M2-2 protein (Table 2). This protein almost completely lost its M2-2 transcriptional inhibition function; only a residual activity (1%) could be detected. The G1S-A contained three different sequence changes. One species (5 out of 13 clones) had a point mutation at nt 17 from U to A, which resulted in the substitution of Lys for Ile-6 (I6K). This protein maintained 81% of M2-2 activity. The other two types of mutations in the M2-2 gene of G1S-A had either 1 or 2 nt deleted between the second and the third AUG codons such that the protein could only be produced from the third AUG, and thus some of the M2-2 in G1S-A-infected cells migrated faster on the polyacrylamide gel (Fig. 4B, lane 2). Similar to G1L-A, these two deletion mutations had very little M2-2 inhibition function as assayed in the minigenome system.
To further determine the function of M2-2 produced from each of the three AUG codons, the M2-2 cDNA was mutagenized by removing two of the three alternative AUG codons. As shown in Table 2, the M2-2 protein initiated from either the first (AUG1) or the second (AUG2) AUG only had slightly reduced activity at a level of 94% and 93% of wild-type M2-2, respectively. The protein produced from the third AUG (AUG3) was not functional as determined by the minigenome assay. To determine the protein sequence change in the N terminus of M2-2 on protein function, the individual amino acids between the first and the third AUG codon were replaced by the Ala codon, and the protein function was analyzed in vitro. Substituting Ala for Thr-2, Pro-4, and Lys-5 had little impact on the in vitro activity of M2-2 (Table 2); only I6A reduced the protein function by 20%, which was similar to the result obtained with G1S-A#1 that had the I6K mutation. Deletion of Pro-4 or Lys-5 between the second and third AUG slightly reduced M2-2 activity; only deletion of the Ile-6 codon eliminated M2-2 function. This result indicated that the Ile-6 is a residue critical to the M2-2 function.
The second most frequently occurring insertion mutations at nt 213 to 216 were also evaluated for their effect on M2-2 function (Table 3). The M2-2 cDNA amplified from the viral RNA of G9-C was cloned into the expression vector and shown to have 3 to 6 nt inserted at nt 213. G9-C#1 (two out of nine clones) had an insertion of three U nucleotides, resulting in an insertion of a Phe residue that had little effect on M2-2 function. However, a second Phe residue insertion from the inserted six U nucleotides in G9-C#2 reduced M2-2 function by 38% (Table 3). G9-C#3 also had six nucleotides (AAUUUU) inserted at nt 213 that resulted in substitution of N71K in addition to the insertion of Ile and Phe. G9-C#3 had only 32% of M2-2 activity. The fourth version of the G9-C mutation accounting for 50% of the clones was the insertion of the four U nucleotides that resulted in the shift of the M2-2 ORF (G9-C#4). However, the G9-C#4 M2-2 protein still had 76% of M2-2 function, which was higher than G9-C#3, which maintained the M2-2 ORF. Amino acid sequence comparisons of these four G9-C-derived insertion mutations indicated that F72 might be critical to M2-2 function. To test this hypothesis, a mutant that lacked the C-terminal 18 residues (18#1) was made by inserting a U nucleotide at position 213 that would result in the shift of the M2-2 ORF. Interestingly, this M2-2 had almost no activity (8%) compared to G9-C#4 (73%). In contrast, 18#2, which had the four-U nucleotide insertion and one Phe residue added at residue 73, had activity (58%) significantly higher than 18#1 (8%). 18#3 with Ile-73 had no M2-2 activity, indicating that F73 was important in maintaining some of the M2-2 activity. These results demonstrated that introduction of specific mutations at several selected regions without completely abolishing M2-2 activity was important for replication of these M2-2 recombinant viruses that had the M2-2 gene overexpressed.
The requirement of the C-terminal amino acid residues on M2-2 function was analyzed by introduction of premature stop codons (Table 3). A single amino acid deletion from the C-terminal M2-2 resulted in a 32% reduction in its activity (C1). A much greater loss of M2-2 activity was observed when two (C2; 48%), four (C4; 17%), or eight (C8; 4%) amino acids were deleted.
Effect of M2-2 overexpression on RSV infection. To examine whether the M2-2 protein supplied in trans would have any inhibitory effect on RSV infection, replication of RSV in cells transiently expressing M2-2 was studied. To distinguish the transiently expressed M2-2 from the virus-encoded M2-2 produced during RSV infection, the M2-2 gene under the control of the T7 promoter in pCITE2a vector was tagged with the ICP4 sequence at its C terminus. The ICP4-tagged M2-2 inhibited RSV minigenome synthesis similarly to M2-2. Thus, the ICP4 tag had no adverse effect on M2-2 function. Three hours after transfection of the BSR T7/5 cells with the ICP4-tagged M2-2 plasmid, the cells were superinfected with RSV A2 at an MOI of 10 and incubated overnight at 35°C. The infected cells were examined by immunofluorescence microscopy (Fig. 5). The influenza virus A/AA/6/60 NP plasmid (29) was used as a control. The transiently expressed M2-2 or NP proteins are shown in red and RSV infected cells are shown in green. RSV was able to replicate in the cells expressing influenza NP but not in the cells producing M2-2. These data indicated that the M2-2 protein expressed at high levels inhibited RSV replication.
DISCUSSION
The amount of M2-2 produced in RSV-infected cells was relatively low and had a slower synthesis kinetics compared to other viral proteins (2). It is thus speculated that the overlapping configuration of M2-2 with the M2-1 ORF might impose a strict regulation on M2-2 gene expression. Additionally, it was found that the M2-1 protein had a negative feedback on the transcription termination of the preceding F gene (23), which may exert an additional control on the level of the M2-1 and M2-2 produced in the infected cells. The exogenously expressed M2-2 strongly inhibited RSV replication in the transfection and superinfection assay. When the M2-2 protein was overexpressed through the change of its position, virus was unable to tolerate a high level of M2-2 protein, and most of the recombinant viruses had mutations in the M2-2 gene, resulting in reduced M2-2 gene function that enabled their replication better than rA2M2-2. The fact that most of the M2-2 mutations were not lethal to M2-2 gene function further demonstrated an important role of M2-2 in RSV replication.
Although three tandem AUG codons are present at the 5' end of the M2-2 gene, only the first and the second AUG codons can be used to make a functional M2-2 protein. When two of the alternative AUG initiation sites in the M2-2 gene were removed, any one of the three AUG sites could be used by ribosomes, but the protein initiated from the third AUG was not functional as assayed in the minigenome system. G1L-A that had the M2-2 protein initiated from the third AUG replicated better than rA2M2-2, and the level of the antigenomic RNA detected in G1L-A-infected cells was also higher than that of rA2M2-2, indicating that M2-2 produced from the third AUG might have some residual function in vivo, at least when overexpressed. Because most of the inserted M2-2 gene was mutated in the recombinant viruses, it was difficult to correlate the level of active M2-2 protein with virus replication in this study.
There is a good correlation between the function of M2-2 and the amount of genome and antigenome RNA synthesized in the infected cells. As shown previously, complete removal of the M2-2 ORF significantly reduced the level of the genomic and antigenomic RNA produced in the infected cells (6, 27). G1S-C could only make the N-terminal 33 amino acids of M2-2 and likely lacked any M2-2 function. Its replication in vitro was very similar to rA2M2-2, and a very low level of antigenomic and genomic RNA was produced in the infected cells. The other M2-2 recombinant viruses had antigenomic and genomic RNA produced at a level intermediate between rA2 and rA2M2-2 and replicated better than rA2M2-2 but worse than rA2 in HEp-2 cells. Thus, the reduced growth of these M2-2 recombinant viruses in HEp-2 cells reflected their reduced M2-2 function. The level of RNA and protein synthesis of the M2-2-shuffled recombinant viruses was slightly lower in HEp-2 cells than Vero cells, but their patterns were very similar in these two cell lines. Further studies are needed to elucidate the mechanism by which the M2-2 protein affects virus replication in HEp-2 cells.
It has been shown that transcription of nonsegmented negative-strand RNA viruses is obligatorily sequential (1, 4). The genes located at the 3' promoter proximal position were produced more abundantly than those positioned distal to the promoter (16). Thus, the NS1 protein was the most abundantly produced protein in RSV-infected cells. The NS1 and NS2 proteins have been shown to be interferon antagonists (8, 34, 35), and the removal of either of these two proteins resulted in poor virus replication in HEp-2 cells (30, 39). Insertion of the M2-2 gene positioned before the NS1 gene could thus affect the production of the NS1 and NS2 proteins. As expected, much less NS1 was produced in cells infected with G1S virus that had a less efficient GE sequence and a short intergenic sequence between M2-2 and NS1, but the level of NS1 produced in G1L-A-infected cells was normal because the GE sequence and the intergenic sequence derived from the SG/G junction in G1L-A terminates M2-2 efficiently. However, virus replication appeared to be impaired mainly by the reduction of M2-2 function rather than the reduction of NS1. Some of the M2-2 molecules produced by the G1S-A recombinant had an I6K mutation (78% activity), and G1S-A grew better than G1S-C that had no M2-2 function. Insertion of M2-2 into the F-M2 junction also reduced the level of M2-1 produced in the infected cells because of insufficient termination of the M2-2 transcription. Nevertheless, the reduced M2-1 protein did not have a significant effect on virus replication. Only G9-B and G9-C that had the impaired M2-2 function replicated poorly in HEp-2 cells; G9-A containing the correct M2-2 gene replicated as efficiently as wild-type rA2 in both Vero and HEp-2 cells, despite its reduced M2-1 protein synthesis and increased levels of the M2-2 protein and antigenomic RNA produced in the infected cells.
Except for G1S-C that had a nucleotide insertion at nt 100, resulting in complete loss of M2-2 function, most of the viruses had mixed mutations in the M2-2 gene. The attempts to isolate viruses with homogenous M2-2 gene populations by several additional rounds of plaque purification were not successful, despite the fact that RSV lacking the M2-2 ORF could be obtained. This indicated that M2-2 produced from the M2-2 gene of different mutations might have resulted in a more balanced M2-2 function in the infected cells. Interestingly, most of the mutations were clustered in two regions that were rich in the A and U sequences, as also seen in the RSV G gene (21). The nucleotide insertions or deletions were possibly generated by polymerase stuttering or slippage errors in these regions (7, 18). The NS2, N, P, and M2-1 genes in each of the M2-2 recombinant viruses did not contain any mutations. In addition, the recombinant RSV with the lacZ gene inserted at the 3' promoter proximal position reported previously by our group did not contain any mutation in the lacZ gene (10). These results indicated that the mutations occurred in the inserted M2-2 gene in these recombinant RSV mutants were due to the overexpression of the M2-2 protein instead of a global phenomenon caused by the inserted gene.
Here we showed that the RSV minigenome assay was very useful in screening amino acid residues in M2-2 critical to RSV replication. Deletion of the first five amino acids individually had a minimal effect on its function. The only residue in the N-terminal M2-2 critical to its in vitro function was Ile-6. Deletion of Ile-6 abolished M2-2 function, but replacement of Ile-6 by Ala or Lys was tolerated and resulted in a slight reduction in its M2-2 activity. Unexpectedly, G9-C#4, which lacked the C-terminal 18 amino acids, still maintained most of the M2-2 function as measured by the minigenome assay, which was much better than the clone with four or eight amino acids deleted from its C terminus. The addition of a Phe at position 73 was the accidental result of the insertion of a UUU codon. The M2-2 does not have a Phe at its C-terminal 18 amino acids; this Phe-73 may have performed some function that enabled the protein to have a better conformation compared to the M2-2 molecules with shorter deletions.
The mechanism by which M2-2 influences the viral RNA synthesis process remains to be determined. Most likely, M2-2 needs to interact with one of the viral proteins that are involved in RNA transcription and replication. We examined the interaction of the ICP4-tagged M2-2 and hemagglutinin-tagged L protein by coimmunoprecipitation, but the results were not conclusive (data not shown). Recently, it was found that RSV M2-1 was detected in the nucleus of the infected cells, and it interacted with Rel A protein to induce the activation of nuclear factor kappa B (33). More studies need to be done to investigate whether M2-2 functions through its interaction with a viral protein(s) or a host factor(s).
ACKNOWLEDGMENTS
We thank the tissue culture facility of MedImmune Vaccine for providing tissue culture cells, Chengjun Mo for helping with immunofluorescence microscopy, Winnie Chan for technical assistance, members of Hong Jin's group for assistance and discussions, and Bin Lu, Richard Spaete, and George Kemble for review of the manuscript.
REFERENCES
Abraham, G., and A. K. Banerjee. 1976. Sequential transcription of the genes of vesicular stomatitis virus. Proc. Natl. Acad. Sci. USA 73:1504-1508.
Ahmadian, G., P. Chambers, and A. J. Easton. 1999. Detection and characterization of proteins encoded by the second ORF of the M2 gene of pneumoviruses. J. Gen. Virol. 80:2011-2016.
Ahmadian, G., J. S. Randhawa, and A. J. Easton. 2000. Expression of the ORF-2 protein of the human respiratory syncytial M2 gene is initiated by aribosomal termination-dependent reinitiation mechanism. EMBO J. 19: 2681-2689.
Ball, L. A., and C. N. White. 1976. Order of transcription of genes of vesicular stomatitis virus. Proc. Natl. Acad. Sci. USA 73:442-446.
Barik, S. 1992. Transcription of human respiratory syncytial virus genome RNA in vitro: requirement of cellular factor(s). J. Virol. 66:6813-6818.
Bermingham, A., and P. L. Collins. 1999. The M2-2 protein of human respiratory syncytial virus is a regulatory factor involved in the balance between RNA replication and transcription. Proc. Natl. Acad. Sci. USA 96:11259-11264.
Bilsel, P. A., and S. T. Nichol. 1990. Polymerase errors accumulating during natural evolution of the glycoprotein gene of vesicular stomatitis virus Indiana serotype isolates. J. Virol. 64:4873-4883.
Bossert, B., S. Marozin, and K. K. Conzelmann. 2003. Nonstructural proteins NS1 and NS2 of bovine respiratory syncytial virus block activation of interferon regulatory factor 3. J. Virol. 77:8661-8668.
Buchholz, U. J., S. Finke, and K. K. Conzelmann. 1999. Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter. J. Virol. 73:251-259.
Cheng, X., M. Munoz, H. Zhou, and H. Jin. 2002. Expression of beta-galactosidase by recombinant respiratory syncytial viruses for microneutralization assay. J. Virol. Methods 105:287-295.
Cheng, X., H. Zhou, R. S. Tang, M. G. Munoz, and H. Jin. 2001. Chimeric subgroup A respiratory syncytial virus with the glycoproteins substituted by those of subgroup B and RSV without the M2-2 gene are attenuated in African green monkeys. Virology 283:59-68.
Collins P. L., R. M. Chanock, and B. R. Murphy. 2001. Respiratory syncytial virus, p. 1443-1485. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Collins, P. L., M. G. Hill, J. Cristina, and H. Grosfeld. 1996. Transcription elongation factor of respiratory syncytial virus, a nonsegmented negative-strand RNA virus. Proc. Natl. Acad. Sci. USA 93:81-85.
Collins, P. L., M. G. Hill, and P. R. Johnson. 1990. The two open reading frames of the 22K mRNA of human respiratory syncytial virus: sequence comparison of antigenic subgroups A and B and expression in vitro. J. Gen. Virol. 71:3015-3020.
Collins, P. L., and B. R. Murphy. 2002. Respiratory syncytial virus: reverse genetics and vaccine strategies. Virology 296:204-211.
Collins, P. L., and G. W. Wertz. 1983. cDNA cloning and transcriptional mapping of nine polyadenylylated RNAs encoded by the genome of human respiratory syncytial virus. Proc. Natl. Acad. Sci. USA 80:3208-3212.
Collins, P. L., and G. W. Wertz. 1985. The envelope-associated 22K protein of human respiratory syncytial virus: nucleotide sequence of the mRNA and a related polytranscript. J. Virol. 54:65-71.
Cunningham, P. R., C. J. Weitzmann, and J. Ofengand. 1991. SP6 RNA polymerase stutters when initiating from an AAA... sequence. Nucleic Acids Res. 19:4669-4673.
Fearns, R., and P. L Collins. 1999. Model for polymerase access to the overlapped L gene of respiratory syncytial virus. J. Virol. 73:388-397.
Fearns, R., and P. L. Collins. 1999. Role of the M2-1 transcription antitermination protein of respiratory syncytial virus in sequential transcription. J. Virol. 73:5852-5864.
Garcia-Barreno, B., A. Portela, T. Delgado, J. A. Lopez, and J. A. Melero. 1990. Frame shift mutations as a novel mechanism for the generation of neutralization resistant mutants of human respiratory syncytial virus. EMBO J. 9:4181-4187.
Grosfeld, H., M. G. Hill, and P. L. Collins. 1995. RNA replication by respiratory syncytial virus (RSV) is directed by the N, P, and L protein; transcription also occurs under these conditions but requires RSV superinfection for efficient synthesis of full-length mRNA. J. Virol. 69:5677-5686.
Hardy, R. W., S. B. Harmon, and G. W. Wertz. 1999. Diverse gene junctions of respiratory syncytial virus modulate the efficiency of transcription termination and respond differently to M2-mediated antitermination. J. Virol. 73:170-176.
Harmon, S. B., A. G. Megaw, and G. W. Wertz. 2001. RNA sequences involved in transcriptional termination of respiratory syncytial virus. J. Virol. 75:36-44.
Huang, Y. T., P. L. Collins, and G. W. Wertz. 1985. Characterization of the 10 proteins of human respiratory syncytial virus: identification of a fourth envelope-associated protein. Virus Res. 2:157-173.
Hubenthal-Voss, J., R. A. Houghten, L. Pereira, and B. Roizman. 1988. Mapping of functional and antigenic domains of the alpha 4 protein of herpes simplex virus 1. J. Virol. 62:454-462.
Jin, H., X. Cheng, H. Z. Zhou, S. Li, and A. Seddiqui. 2000. Respiratory syncytial virus that lacks open reading frame 2 of the M2 gene (M2-2) has altered growth characteristics and is attenuated in rodents. J. Virol. 74: 74-82.
Jin, H., D. Clarke, Z.-Y.H. Zhou, et al. 1998. Recombinant human respiratory syncytial virus (RSV) from cDNA and construction of subgroup A and B chimeric RSV. Virology 251:206-214.
Jin, H., B. Lu, H. Zhou, et al. 2003. Multiple amino acid residues confer temperature sensitivity to human influenza virus vaccine strains (FluMist) derived from cold-adapted A/Ann. Arbor/6/60. Virology 306:18-24.
Jin, H., H. Zhou, X. Cheng, R. Tang, M. Munoz, and N. Nguyen. 2000. Recombinant respiratory syncytial viruses with deletions in the NS1, NS2, SH, and M2-2 genes are attenuated in vitro and in vivo. Virology 273: 210-218.
Kuo, L., H. Grosfeld, J. Cristina, M. G. Hill, and P. L. Collins. 1996. Effects of mutations in the gene-start and gene-end sequence motifs on transcription of monocistronic and dicistronic minigenomes of respiratory syncytial virus. J. Virol. 70:6892-6901.
Moudy, R. M., W. M. Sullender, and G. W. Wertz. 2004. Variations in intergenic region sequences of human respiratory syncytial virus clinical isolates: analysis of effects on transcriptional regulation. Virology 327: 121-133.
Reimers, K., K. Buchholz, and H. Werchau. 2005. Respiratory syncytial virus M2-1 protein induces the activation of nuclear factor kappa B. Virology 331:260-268.
Schlender, J., B. Bossert, U. Buchholz, and K. K. Conzelmann. 2000. Bovine respiratory syncytial virus nonstructural proteins NS1 and NS2 cooperatively antagonize alpha/beta interferon-induced antiviral response. J. Virol. 74:8234-8242.
Spann, K. M., K. C. Tran, B. Chi, R. L. Rabin, and P. L. Collins. 2004. Suppression of the induction of alpha, beta, and lambda interferons by the NS1 and NS2 proteins of human respiratory syncytial virus in human epithelial cells and macrophages. J. Virol. 78:4363-4369.
Sutherland, K. A., P. L. Collins, and M. E. Peeples. 2001. Synergistic effects of gene-end signal mutations and the M2-1 protein on transcription termination by respiratory syncytial virus. Virology 288:295-307.
Tang, R. S., N. Nguyen, X. Cheng, and H. Jin. 2001. Requirement of cysteines and length of the human respiratory syncytial virus M2-1 protein for protein function and virus viability. J. Virol. 75:11328-11335.
Teng, M. N., and P. L. Collins. 1998. Identification of the respiratory syncytial virus proteins required for formation and passage of helper-dependent infectious particles. J. Virol. 72:5707-5716.
Teng, M. N., and P. L. Collins. 1999. Altered growth characteristics of recombinant respiratory syncytial viruses which do not produce NS2 protein. J. Virol. 73:466-473.
Teng, M. N., S. S. Whitehead, A. Bermingham, M. St. Claire, W. R. Elkins, B. R. Murphy, and P. L. Collins. 2000. Recombinant respiratory syncytial virus that does not express the NS1 or M2-2 protein is highly attenuated and immunogenic in chimpanzees. J. Virol. 74:9317-9321.
Yu, Q., R. W. Hardy, and G. W. Wertz. 1995. Functional cDNA clones of the human respiratory syncytial (RS) virus N, P, and L proteins support replication of RS virus genomic RNA analogs and define minimal trans-acting requirements for RNA replication. J. Virol. 69:2412-2419.(Xing Cheng, HyunJung Park)
ABSTRACT
The M2-2 protein of respiratory syncytial virus (RSV) is involved in regulation of viral RNA transcription and replication. Encoded by the next-to-last gene of RSV, the M2-2 open reading frame (ORF) overlaps with the upstream M2-1 ORF, suggesting that the production of the M2-2 protein might be tightly regulated during virus replication. To evaluate the effect of M2-2 overexpression on RSV replication, the M2-2 gene was separated from M2-1 by leaving it at the position prior to the M2-1 or moving it to the promoter proximal position as an independent transcriptional unit in the RSV A2 genome. Although recombinant viruses bearing the shuffled M2-2 gene were recovered and expressed higher levels of M2-2, most of these viruses grew poorly in HEp-2 cells. Sequence analysis revealed that various mutations (substitution, insertion, and deletion) occurred in the M2-2 gene, resulting in reduced M2-2 activity as measured by the RSV minigenome system. Further examination of the M2-2 sequence and its function showed that either one of the first two AUG codons located at the 5' end of M2-2 could be used to produce a functional M2-2 protein and that deletion of the first six amino acids from its N terminus or four amino acids from its C terminus greatly reduced its function. The effect of M2-2 protein on RSV replication was also studied by examining RSV replication in cells transiently expressing M2-2. The M2-2 protein expressed at a high level completely inhibited RSV replication. These results strongly suggested that the level of the M2-2 protein produced in the infected cells is critical to RSV replication.
INTRODUCTION
Respiratory syncytial virus (RSV) is a member of the Pneumovirus genus of the Paramyxoviridae family. It is the leading viral cause of serious lower-respiratory tract infection in infants and children worldwide (12, 15). The single-strand negative-sense RNA genome of the RSV A2 strain is 15,222 nucleotides (nt) in length and has 10 transcriptional units. Each transcriptional unit encodes a single protein, with the exception of the M2 gene that encodes two proteins, M2-1 and M2-2. As with all nonsegmented RNA viruses, RNA synthesis requires a genomic RNA encapsidated with nucleocapsid (N) protein and the virus-encoded components of the RNA polymerase, the large (L) polymerase protein and the phosphoprotein (P) (5, 22, 41). Each transcriptional unit is flanked by the gene-start (GS) sequence required for initiation of transcription and the gene-end (GE) sequence that directs polyadenylation and release of mRNA (24, 25, 31). The intergenic region (IGR), which varies in sequence and length, affects efficient transcription termination and downstream gene expression (32). Transcription begins at the 3' end of the genome and follows a linear, sequential, and polar gradient that requires termination of the upstream gene prior to initiating transcription of the downstream gene (16). The gene located at the 3' promoter proximal position is most abundantly transcribed, whereas the one most distal to the 3' promoter is expressed the least (16).
The M2 gene is unique to the genus Pneumovirus. It has two overlapping open reading frames (ORFs), M2-1 and M2-2 (14, 17), both involved in the viral RNA synthesis process. The processivity and antitermination functions of M2-1 are essential to RSV replication (13, 19, 20). In contrast, the M2-2 protein inhibits RSV minigenome transcription and replication and is dispensable for RSV replication (6, 27). However, recombinant virus lacking the M2-2 ORF grew poorly in HEp-2 cells and was attenuated in rodents and primates (11, 27, 40). The levels of the antigenomic and genomic RNA were significantly reduced compared to mRNA when the M2-2 gene was removed, suggesting that M2-2 was involved in regulating the switch between the viral RNA transcription and replication process (6, 27).
Three AUG codons are present at the 5' end of the M2-2 at amino acid positions 1, 3, and 7. Depending on the usage of the three AUG codons, the M2-2 gene overlaps with the C-terminal portion of M2-1 by 29, 23, or 11 nt, respectively. A previous study indicated that M2-2 could be initiated from any one of the three AUG codons by the ribosomes departing from the M2-1 ORF through a ribosomal termination-dependent reinitiation mechanism (3). A later study showed that disruption of the overlapping sequence between the ORFs of M2-1 and M2-2 did not affect initiation of M2-2 (37). However, the next-to-last gene location of M2 in the RSV genome and the initiation mechanism of M2-2 by ribosome reverse translocation (3) imply that the M2-2 protein is produced at a low level and is likely required during the late stage of virus infection (2).
To determine the effect of overproduction of M2-2 on virus replication, the M2-2 ORF was disentangled from M2-1 by moving it as an independent transcriptional unit to the position proximal to the M2-1 or proximal to the NS1 gene. It was expected that the M2-2 positioned at the first gene location would be expressed at a level much higher than that positioned at the ninth location and that the effect of the level of M2-2 protein on RSV replication could thus be evaluated. The recombinant viruses with the M2-2 ORF shuffled to these two different locations were produced, and their replication in cell culture was examined. In most circumstances, recombinant M2-2-shuffling viruses accumulated mutations in the M2-2 gene that lowered M2-2 function. In addition, it was found that the M2-2 protein supplied in trans inhibited virus replication, suggesting that the level of the M2-2 protein is critical to RSV replication.
MATERIALS AND METHODS
Cells and viruses. Monolayer cultures of HEp-2 and Vero cells obtained from the American Tissue Culture Collection were maintained in minimal essential medium (MEM) containing 5% fetal bovine serum (FBS). The modified vaccinia virus Ankara (MVA-T7) expressing bacteriophage T7 RNA polymerase was obtained from Bernard Moss and grown in CEK cells. The BSR T7/5 (9) cell line was maintained in Glasgow MEM (Invitrogen, Carlsbad, CA) supplemented with 10% FBS, 10% tryptose phosphate broth, and 0.05 mg/ml geneticin.
Construction of antigenomic cDNA and recovery of recombinant RSV. The M2-2 gene was inserted as an independent transcriptional unit into the RSV A2 antigenomic cDNA from which the M2-2 gene was deleted, yielding pA2M2-2 (27). To insert the M2-2 gene into the promoter proximal location, a KpnI restriction site was introduced at nt 97 upstream of the NS1 initiation site in the pET-X/A subclone that contained the RSV sequence of nt 1 to 2128 in the pET3 vector. The M2-2 gene was amplified by a 5' M2-2 gene-specific primer containing the KpnI site (underlined) and the M2-2 gene sequence (bold) starting from the second ATG (5'-GATCGGTACCATGCCAAAAATAATGATACTACC-3') and a 3' M2-2 gene-specific primer that contained the KpnI site (underlined), the GE sequence (italic and lowercase), the GS sequence (lowercase), and the 3' M2-2 sequence (bold) (5'-GACTGGTACCtatttgccccAGAttttaaataactgTTATGACACTAATATATATATTG-3'). The M2-2 gene cassette was inserted into the KpnI site in pET-X/A. The XmaI-AvrII fragment that contained the inserted M2-2 gene was then cloned into pA2M2-2 antigenomic cDNA (Fig. 1). The cDNA construct that had a short intergenic sequence of 3 nt between the M2-2 and NS1 gene junction was designated as pM2-2G1S. To ensure efficient transcription termination of the inserted M2-2 gene, the GE sequence with an additional A residue and the intergenic sequence from the SH/G junction were inserted between the M2-2 and NS1. The M2-2 GE sequence in pM2-2G1S was abolished by site-directed mutagenesis to create an EcoRI site in the pET-X/A subclone and inserted with the SH-G IGR amplified by PCR with the following primers: 5'-GTACGAATTCCATTCATCAATCCAACAGCCC-3' and 5'-GCATGAATTCAATGTTATTGTTAGTCTTG-3' (EcoRI site is underlined). Thus, pM2-2G1L contained 133 nt between the stop codon of M2-2 and its GE sequence and 51 nt from its GE sequence to the GS sequence of NS1.
Another antigenomic cDNA clone was constructed to move the M2-2 gene to the ninth position prior to the M2-1 gene. The pET-S/B subclone containing the RSV sequence from SacI (RSV nt 4477) to BamHI (RSV nt 8498) was modified to introduce a BglII site in the F/M2-1 IGR at nt 7554. The pair of primers used to amplify the M2-2 ORF were 5'-GCTCAGATCTggggcaaatATGCCAAAAATAATGATGCTGCC and 5'-GACTAGATCTtttaaataactgTTATGACACTAATAT ATATATTG (the BglII sites are underlined, the GS and GE sequences are in lowercase, and the M2-2 sequences are in bold). The M2-2 gene cassette included the BglII sites, the GS signal, the M2-2 ORF, and the GE signal. TheSacI-BamHI fragment containing the M2-2 BglII fragment inserted into the BglII site at nt 7554 was cloned into the antigenomic cDNA of pA2M2-2, and the plasmid was designated pM2-2G9. The IGRs between F and M2-2 and between M2-2 and M2-1 were 7 nt and 44 nt, respectively.
Recovering of recombinant viruses. Recovery of infectious viruses from RSV antigenomic cDNA was performed as described previously (28). Briefly, antigenomic cDNA and supporting plasmids encoding the N, P, L, and M2-1 genes, each under the control of the T7 promoter, were cotransfected into MVA-T7-infected HEp-2 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). The culture medium was replaced after overnight incubation at 35°C, and 3 days later the supernatant from the transfected cells was passaged onto fresh Vero cells to amplify any rescued viruses. Infectious viruses recovered from the transfected cells were plaque purified three times and amplified in Vero cells. Virus recovered from pM2-2G1S, pM2-2G1L, and pM2-2G9 was designated as G1S, G1L, and G9, respectively.
Replication of M2-2 recombinant viruses in cell cultures. Replication of M2-2 recombinant viruses was analyzed in HEp-2 and Vero cells. Cells grown in six-well plates were infected with virus at a multiplicity of infection (MOI) of 0.2. After a 1-h adsorption at room temperature, the infected cells were washed three times with phosphate-buffered saline, the medium was replaced with 2 ml of OptiMEM (Invitrogen, Carlsbad, CA), and the culture was incubated at 35°C in an incubator containing 5% CO2. At various times postinfection, 250 μl of culture supernatant was collected and stored at –80°C until virus titration. Each aliquot taken was replaced with an equal volume of fresh medium. The virus titer was determined by plaque assay on Vero cells, using an overlay of 1% methylcellulose and Eagle's MEM/L15 medium (JRB Biosciences, Lenexas, Kansas) containing 2% FBS. The plaques were enumerated by immunostaining using goat anti-RSV polyclonal antibody (Biogenesis, Sandown, NH).
Northern blot analysis. For Northern blot analysis, total cellular RNA was extracted from virus-infected HEp-2 or Vero cells using Trizol reagent (Invitrogen, Carlsbad, CA), followed by an additional round of phenol-chloroform extraction and ethanol precipitation. RNA was electrophoresed on 1.5% agarose gel containing 0.44 M formaldehyde and transferred to a nylon membrane (Amersham Pharmacia Biotech, Piscataway, NJ). The membrane was hybridized with a gene-specific riboprobe labeled with digoxigenin. The hybridized RNA bands were visualized with a Dig-Luminescent Detection Kit for Nucleic Acids (Roche Diagnostics Corportions, Indianapolis, IN).
Western blot analysis and immunoprecipitation. For Western blotting analysis, Vero or HEp-2 cells were infected with virus at an MOI of 1.0 and incubated at 35°C for 48 h. The infected cells were lysed in protein lysis buffer, and the cells lysates were electrophoresed on 10 to 20% Tris glycine gels (Invitrogen, Carlsbad, CA). The proteins were transferred to a nylon membrane using a semidry transfer cell (Bio-Rad, Hercules, CA). The membrane was blotted with anti-N monoclonal antibody (gift of Jose Melero), rabbit antiserum against NS1 protein, or guinea pig antiserum against M2-1 protein (gift of Jeyesh Meanger). The membrane was then incubated with a secondary antibody conjugated with horseradish peroxidase, and the protein bands were visualized after incubation with the ECL substrate (Amersham Pharmacia Biotech, Piscataway, NJ). For immunoprecipitation, the cells were infected with virus at an MOI of 1.0 and labeled with 100 μCi/ml of [35S]cysteine and [35S]methionine (35S-promix; Amersham Pharmacia Biotech, Piscataway, NJ) at 16 h postinfection for 4 h. The labeled cell monolayers were lysed in radioimmunoprecipitation assay buffer (0.15 M NaCl, 1% Na-deoxycholate, 1% Triton X-100, 0.1% sodium dodecyl sulfate [SDS], and 0.01 M Tris, pH 7.4), and the polypeptides were immunoprecipitated with an appropriate antibody, electrophoresed on a 17.5% polyacrylamide gel containing 0.1% SDS and 4 M urea or a 4 to 20% polyacrylamide gel containing 0.1% SDS (Invitrogen, Carlsbad, CA), and detected by autoradiography.
Sequence analysis of the M2-2 gene from the recovered M2-2 recombinant viruses. The M2-2 gene was amplified by reverse transcription-PCR (RT-PCR) using the primers flanking the M2-2 gene and viral RNA extracted with a viral RNA extraction mini kit (QIAGEN, Valencia, CA). For G1 variants, the M2-2 gene was amplified with V1964 primer (nt 1 to 18 of the antigenomic sense) and V1950 (nt 649 to 622 of the genomic sense). For G9, the M2-2 gene was amplified with V1959 (nt 7200 to 7229 of the antigenomic sense) and V1963 (nt7710 to 7691 of the genomic sense). The amplified M2-2 cDNAs were examined by DNA sequencing. To determine the mixed sequence populations, the RT-PCR-amplified M2-2 cDNAs were also cloned into the pPCR2.1 vector (Invitrogen, Carlsbad, CA) for sequence analysis.
Functional analysis of M2-2 cDNA by a minigenome assay. To examine the M2-2 function by a minigenome assay, the M2-2 cDNA amplified from viral RNA was cloned into the pCITE2a vector (Novagen, Madison, WI) under the control of the T7 promoter. Mutagenesis of pCITE2a-M2-2 was performed using a QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) for evaluation of M2-2 sequence variations and its in vitro function. The inhibition effect of the M2-2 protein on RSV minigenome transcription and protein expression was analyzed essentially as previously described (37). Briefly, duplicate wells of HEp-2 cells in six-well plates were infected with MVA-T7 at an MOI of 1.0 and transfected with plasmid encoding the N, P, and L genes and the pRSVCAT minigenome (0.2, 0.2, 0.1, and 0.4 μg per well, respectively), together with 0.2 μg of wild-type M2-2 or an M2-2 variant, using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Two days after transfection, the cell lysates were prepared and assayed for chloramphenicol acetyltransferase activity (Roche Applied Science, Indianapolis, IN). The relative activity of each M2-2 mutant was expressed as the percentage of activity relative to that of wild-type M2-2, and the data obtained were averaged from a minimum of three experiments.
Replication of RSV in cells transiently expressing M2-2. The effect of the M2-2 protein expressed transiently on RSV infection was examined by immunofluorescence. The M2-2 gene in pCITE2a vector was fused with the ICP4 sequence (5'-GATGAATATGATGATGCAGCTGATGCAGCAGGAGATCGA GCACCAGGA-3') (26) at its C terminus. The ICP4 tag allowed differentiation of the transiently expressed M2-2 from virus-encoded M2-2 produced in the infected cells. The BSR T7/5 cells in 24-well plates were transfected with 2 μg of wild-type M2-2 or the influenza A/AA/6/60 NP plasmid (29) using Lipofectamine 2000. After incubation at 37°C for 3 h, the transfected cells were infected with rA2 at an MOI of 10.0, followed by incubation at 35°C for 24 h. The cell monolayers were fixed with 1% paraformaldehyde and permeabilized with 0.2% Triton X-100. The cells were immunostained with rabbit polyclonal antibody against the NS1 protein of RSV (30) and mouse monoclonal antibody against ICP4 (Rumbaugh-Goodwin, Plantation, FL) or mouse monoclonal antibody against the influenza virus NP protein (Biodesign, Saco, Maine). The cells were then incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit antibodies to detect RSV infected cells and Texas Red-conjugated goat anti-mouse to detect ICP4- or NP-expressing cells. The cell images were captured by an ORCA-100 digital camera attached to the fluorescence microscope.
RESULTS
Construction and recovery of recombinant M2-2-shuffled viruses. To examine the effect of the increased expression of M2-2 on RSV replication, the M2-2 gene was separated from its overlapping M2-1 gene and moved as an independent transcriptional unit into two different locations in order to increase the quantity of RNA transcription and possibly protein expression. The F-M2 IGR was selected to insert the M2-2 gene (pM2-2G9) in order to have its transcriptional level similar to M2-1. The second location for the M2-2 gene insertion was the 3' promoter proximal region that would allow maximal expression of the M2-2 protein by the recombinant virus. To minimize the effect of the inserted M2-2 gene on the expression of the NS1 gene and subsequent protein expression, two antigenomic cDNA clones were constructed to contain a short (pM2-2G1S) or long (pM2-2G1L) intergenic sequence between the inserted M2-2 gene and the downstream NS1 gene (Fig. 1). The M2-2 gene transcriptional unit in G1S has an inefficient GE sequence compared to the GE sequence that has five A residues (24, 36). These three M2-2 recombinant viruses were recovered from cDNA using a previously described reverse genetics system (28). At least two different virus stocks obtained from several independent rescue experiments were produced. The recombinant viruses obtained from pM2-2G1S were designated G1S-A, G1S-B, G1S-C and G1S-D; the one from pM2-2G1L was designated G1L-A; and those from pM2-2G9 were designated G9-A, G9-B, and G9-C.
Replication of the M2-2 recombinant viruses in tissue culture. Replication of the M2-2 recombinant viruses was compared to rA2 and a recombinant lacking M2-2, rA2M2-2, in HEp-2 and Vero cells (Fig. 2). As previously observed (28), the growth kinetics of rA2M2-2 in Vero cells was slower than rA2 but eventually reached a peak titer only slightly lower than rA2. In contrast, replication of rA2M2-2 in HEp-2 cells was approximately 100-fold lower than rA2. Except for G9-A that replicated similar to rA2, the other three M2-2 recombinants with shuffled M2-2 replicated less well than rA2, especially in the HEp-2 cells. G1S-C replicated similarly to rA2M2-2 in both cell lines; G9-C and G1L-A had intermediate growth kinetics, and their peak titers in HEp-2 cells were much lower than rA2 in HEp-2 cells. Therefore, shuffling of the M2-2 gene to either the first or the ninth position affected RSV replication in tissue culture cells differently.
Effect of M2-2 shuffling on downstream viral gene expression. The effect of the shuffled M2-2 gene on viral RNA and protein synthesis was examined in both HEp-2 and Vero cells and was shown to have similar patterns in both cell lines. The antigenomic RNA produced in G1S-C-infected cells was greatly reduced, and its level was similar to rA2M2-2; antigenomic RNA levels produced in G9-C- and G1L-A-infected cells were moderately reduced, with a decrease of approximately two- to threefold (Fig. 3A). G9-A had antigenomic RNA slightly higher than rA2. This result suggested that G1S-C might have completely lost the M2-2 function, whereas G9-C and G1L-A might have partially lost M2-2 function. To examine the impact of the inserted M2-2 on the downstream gene expression, levels of NS1 and M2-1 RNAs were examined by Northern blotting. As predicted, there was a significant amount of M2-2/NS1 readthrough RNA produced in G1S-infected cells, resulting in reduction of the NS1 mRNA. The amount of NS1 mRNA produced in G1L-A infected cells was similar to rA2. The M2-1 mRNA in G9-A- and G9-C-infected cells was slightly reduced, and a readthrough M2-2/M2-1 RNA was detected. Insertion of M2-2 at the first position (G1S and G1L) had no effect on the synthesis of M2-1. The level of the N mRNA produced by the recombinant viruses was comparable to rA2. Viral proteins produced in the recombinant virus-infected cells were examined by Western blotting analysis (Fig. 3B). The amount of the N protein was equivalent among different viruses. Consistent with the Northern blotting result, the NS1 protein was reduced in G1S-C-infected cells. Both G9-A and G9-C had reduced M2-1 protein expression.
Aberrant expression of the M2-2 proteins in M2-2 recombinant virus-infected cells. There was an overall higher level of M2-2 mRNA produced in cells infected with the viruses that had the M2-2 inserted at the first gene position compared to those at the ninth position (Fig. 4A). A high level of M2-2/NS1 and M2-2/NS1/NS2 readthrough RNA was produced in G1S-infected cells. G1L that had the GE sequence and the intergenic sequence derived from SH/G inserted between the M2-2 and NS1 junction terminated M2-2 mRNA transcription efficiently, and very little M2-2/NS1 readthrough RNA was detected. Thus, the amount of M2-2 mRNA produced by G1L was higher than that of the G1S viruses. The M2-2 mRNA produced by the G9 viruses was higher than rA2 but slightly lower than most of the G1S viruses. These results confirmed that the virus containing an independent M2-2 transcriptional unit synthesized higher levels of mRNA; in addition, comparison of G1L and G1S demonstrated that the GE sequence and the length of the intergenic sequence affected the gene expression level.
The M2-2 proteins produced in the M2-2 recombinant virus-infected cells were examined by immunoprecipitation using anti-M2-2 antibody (Fig. 4B). Only G9-A (lane 6) produced a protein that migrated identically to that of rA2 and at a level slightly higher than that of rA2; the M2-2 polypeptides detected in other M2-2 recombinant virus-infected cells were heterogeneous in size and amount. G1L-A produced a high level of a single M2-2 band that was smaller than wild-type M2-2 (Fig. 4B). No M2-2 protein was detected in G1S-C-infected cells. Multiple M2-2 bands were detected in cells infected with G1S-A, G1S-B, G1S-D, G9-B, and G9-C. These data indicated that the M2-2 protein produced by most of the M2-2 recombinant viruses did not have the correct protein size, which most likely resulted from different sequence changes in the M2-2 ORF.
Sequence variations of the M2-2 gene in M2-2 recombinant viruses. The M2-2 gene sequences of all the obtained M2-2 recombinant viruses were examined by sequencing of the cDNAs amplified from the viral RNA. Consistent with the previous protein analysis result, only G9-A had the correct M2-2 sequence; all the other viruses had sequence changes clustered in three different regions (Table 1). The first region was A/U rich and located between the second and the third AUG as found in M2-2 of G1L-A and G1S-A. G1L-A had a single A nucleotide inserted at position 16, and thus M2-2 could only be translated from the third initiation codon, resulting in a protein smaller than M2-2 (Fig. 4B, lane 1). G1S-A had nucleotide U-17 replaced by A, resulting in a substitution of Ile-6 by Lys. The second mutation region was at nt 100 of the M2-2 gene as shown for G1S-C, which had an A nucleotide inserted at nt 100 that resulted in the frameshift of the M2-2 ORF. Therefore, the M2-2 protein was not detected in G1S-C-infected cells (Fig. 4B, lane 4). The third highly mutated region occurred in an A/U-rich region at nt 213 to 216. The sequence downstream of nt 213 contained mixed sequences for G1S-B, G1S-D, G9-B, and G9-C, indicating that these viruses contained mixed virus populations despite three rounds of virus plaque purification. G1S-B and G1S-D had 1 nt inserted at position 213, and G9-B contained 3 or 6 nt inserted at position 213; G9-C had three to six U or A nucleotides inserted at the same position. Insertion of 1 or 2 nt at position 213 would cause the frameshift of M2-2 ORF and result in the smallest-sized M2-2 protein produced in cells infected with G1S-B, G1S-D, G9-B, and G9-C (Fig. 4B, lanes 3, 5, 7, and 8).
To eliminate the possibility that the inserted M2-2 might cause mutations in the other RSV genomes, the NS2, N, P, and M2-1 genes were sequenced from all the obtained M2-2 recombinant viruses. No mutations were identified in these genes, indicating that the mutation that occurred in the M2-2 was the result of the M2-2 overexpression instead of its ability to cause an increased mutation rate in the RSV genome.
The effect of M2-2 gene sequence variations on M2-2 function. To further determine the nature of the mixed mutations in the M2-2 gene of the recombinant viruses, the M2-2 cDNAs amplified from the viral RNAs of M2-2 recombinant viruses were cloned into the pPCR2.1 vector, and approximately 10 cDNA clones from each RT-PCR DNA product were sequenced. A few of the representative M2-2 cDNAs were further cloned into the pCITE2a vector such that the M2-2 gene was under the control of the T7 promoter. The functions of these M2-2 genes were analyzed using the minigenome assay as described previously (38). Consistent with the sequencing result using RT-PCR DNA as a template, G1L-A had a nucleotide inserted at nt 16, and the protein could only be made from the third AUG codon, resulting in a shortened M2-2 protein (Table 2). This protein almost completely lost its M2-2 transcriptional inhibition function; only a residual activity (1%) could be detected. The G1S-A contained three different sequence changes. One species (5 out of 13 clones) had a point mutation at nt 17 from U to A, which resulted in the substitution of Lys for Ile-6 (I6K). This protein maintained 81% of M2-2 activity. The other two types of mutations in the M2-2 gene of G1S-A had either 1 or 2 nt deleted between the second and the third AUG codons such that the protein could only be produced from the third AUG, and thus some of the M2-2 in G1S-A-infected cells migrated faster on the polyacrylamide gel (Fig. 4B, lane 2). Similar to G1L-A, these two deletion mutations had very little M2-2 inhibition function as assayed in the minigenome system.
To further determine the function of M2-2 produced from each of the three AUG codons, the M2-2 cDNA was mutagenized by removing two of the three alternative AUG codons. As shown in Table 2, the M2-2 protein initiated from either the first (AUG1) or the second (AUG2) AUG only had slightly reduced activity at a level of 94% and 93% of wild-type M2-2, respectively. The protein produced from the third AUG (AUG3) was not functional as determined by the minigenome assay. To determine the protein sequence change in the N terminus of M2-2 on protein function, the individual amino acids between the first and the third AUG codon were replaced by the Ala codon, and the protein function was analyzed in vitro. Substituting Ala for Thr-2, Pro-4, and Lys-5 had little impact on the in vitro activity of M2-2 (Table 2); only I6A reduced the protein function by 20%, which was similar to the result obtained with G1S-A#1 that had the I6K mutation. Deletion of Pro-4 or Lys-5 between the second and third AUG slightly reduced M2-2 activity; only deletion of the Ile-6 codon eliminated M2-2 function. This result indicated that the Ile-6 is a residue critical to the M2-2 function.
The second most frequently occurring insertion mutations at nt 213 to 216 were also evaluated for their effect on M2-2 function (Table 3). The M2-2 cDNA amplified from the viral RNA of G9-C was cloned into the expression vector and shown to have 3 to 6 nt inserted at nt 213. G9-C#1 (two out of nine clones) had an insertion of three U nucleotides, resulting in an insertion of a Phe residue that had little effect on M2-2 function. However, a second Phe residue insertion from the inserted six U nucleotides in G9-C#2 reduced M2-2 function by 38% (Table 3). G9-C#3 also had six nucleotides (AAUUUU) inserted at nt 213 that resulted in substitution of N71K in addition to the insertion of Ile and Phe. G9-C#3 had only 32% of M2-2 activity. The fourth version of the G9-C mutation accounting for 50% of the clones was the insertion of the four U nucleotides that resulted in the shift of the M2-2 ORF (G9-C#4). However, the G9-C#4 M2-2 protein still had 76% of M2-2 function, which was higher than G9-C#3, which maintained the M2-2 ORF. Amino acid sequence comparisons of these four G9-C-derived insertion mutations indicated that F72 might be critical to M2-2 function. To test this hypothesis, a mutant that lacked the C-terminal 18 residues (18#1) was made by inserting a U nucleotide at position 213 that would result in the shift of the M2-2 ORF. Interestingly, this M2-2 had almost no activity (8%) compared to G9-C#4 (73%). In contrast, 18#2, which had the four-U nucleotide insertion and one Phe residue added at residue 73, had activity (58%) significantly higher than 18#1 (8%). 18#3 with Ile-73 had no M2-2 activity, indicating that F73 was important in maintaining some of the M2-2 activity. These results demonstrated that introduction of specific mutations at several selected regions without completely abolishing M2-2 activity was important for replication of these M2-2 recombinant viruses that had the M2-2 gene overexpressed.
The requirement of the C-terminal amino acid residues on M2-2 function was analyzed by introduction of premature stop codons (Table 3). A single amino acid deletion from the C-terminal M2-2 resulted in a 32% reduction in its activity (C1). A much greater loss of M2-2 activity was observed when two (C2; 48%), four (C4; 17%), or eight (C8; 4%) amino acids were deleted.
Effect of M2-2 overexpression on RSV infection. To examine whether the M2-2 protein supplied in trans would have any inhibitory effect on RSV infection, replication of RSV in cells transiently expressing M2-2 was studied. To distinguish the transiently expressed M2-2 from the virus-encoded M2-2 produced during RSV infection, the M2-2 gene under the control of the T7 promoter in pCITE2a vector was tagged with the ICP4 sequence at its C terminus. The ICP4-tagged M2-2 inhibited RSV minigenome synthesis similarly to M2-2. Thus, the ICP4 tag had no adverse effect on M2-2 function. Three hours after transfection of the BSR T7/5 cells with the ICP4-tagged M2-2 plasmid, the cells were superinfected with RSV A2 at an MOI of 10 and incubated overnight at 35°C. The infected cells were examined by immunofluorescence microscopy (Fig. 5). The influenza virus A/AA/6/60 NP plasmid (29) was used as a control. The transiently expressed M2-2 or NP proteins are shown in red and RSV infected cells are shown in green. RSV was able to replicate in the cells expressing influenza NP but not in the cells producing M2-2. These data indicated that the M2-2 protein expressed at high levels inhibited RSV replication.
DISCUSSION
The amount of M2-2 produced in RSV-infected cells was relatively low and had a slower synthesis kinetics compared to other viral proteins (2). It is thus speculated that the overlapping configuration of M2-2 with the M2-1 ORF might impose a strict regulation on M2-2 gene expression. Additionally, it was found that the M2-1 protein had a negative feedback on the transcription termination of the preceding F gene (23), which may exert an additional control on the level of the M2-1 and M2-2 produced in the infected cells. The exogenously expressed M2-2 strongly inhibited RSV replication in the transfection and superinfection assay. When the M2-2 protein was overexpressed through the change of its position, virus was unable to tolerate a high level of M2-2 protein, and most of the recombinant viruses had mutations in the M2-2 gene, resulting in reduced M2-2 gene function that enabled their replication better than rA2M2-2. The fact that most of the M2-2 mutations were not lethal to M2-2 gene function further demonstrated an important role of M2-2 in RSV replication.
Although three tandem AUG codons are present at the 5' end of the M2-2 gene, only the first and the second AUG codons can be used to make a functional M2-2 protein. When two of the alternative AUG initiation sites in the M2-2 gene were removed, any one of the three AUG sites could be used by ribosomes, but the protein initiated from the third AUG was not functional as assayed in the minigenome system. G1L-A that had the M2-2 protein initiated from the third AUG replicated better than rA2M2-2, and the level of the antigenomic RNA detected in G1L-A-infected cells was also higher than that of rA2M2-2, indicating that M2-2 produced from the third AUG might have some residual function in vivo, at least when overexpressed. Because most of the inserted M2-2 gene was mutated in the recombinant viruses, it was difficult to correlate the level of active M2-2 protein with virus replication in this study.
There is a good correlation between the function of M2-2 and the amount of genome and antigenome RNA synthesized in the infected cells. As shown previously, complete removal of the M2-2 ORF significantly reduced the level of the genomic and antigenomic RNA produced in the infected cells (6, 27). G1S-C could only make the N-terminal 33 amino acids of M2-2 and likely lacked any M2-2 function. Its replication in vitro was very similar to rA2M2-2, and a very low level of antigenomic and genomic RNA was produced in the infected cells. The other M2-2 recombinant viruses had antigenomic and genomic RNA produced at a level intermediate between rA2 and rA2M2-2 and replicated better than rA2M2-2 but worse than rA2 in HEp-2 cells. Thus, the reduced growth of these M2-2 recombinant viruses in HEp-2 cells reflected their reduced M2-2 function. The level of RNA and protein synthesis of the M2-2-shuffled recombinant viruses was slightly lower in HEp-2 cells than Vero cells, but their patterns were very similar in these two cell lines. Further studies are needed to elucidate the mechanism by which the M2-2 protein affects virus replication in HEp-2 cells.
It has been shown that transcription of nonsegmented negative-strand RNA viruses is obligatorily sequential (1, 4). The genes located at the 3' promoter proximal position were produced more abundantly than those positioned distal to the promoter (16). Thus, the NS1 protein was the most abundantly produced protein in RSV-infected cells. The NS1 and NS2 proteins have been shown to be interferon antagonists (8, 34, 35), and the removal of either of these two proteins resulted in poor virus replication in HEp-2 cells (30, 39). Insertion of the M2-2 gene positioned before the NS1 gene could thus affect the production of the NS1 and NS2 proteins. As expected, much less NS1 was produced in cells infected with G1S virus that had a less efficient GE sequence and a short intergenic sequence between M2-2 and NS1, but the level of NS1 produced in G1L-A-infected cells was normal because the GE sequence and the intergenic sequence derived from the SG/G junction in G1L-A terminates M2-2 efficiently. However, virus replication appeared to be impaired mainly by the reduction of M2-2 function rather than the reduction of NS1. Some of the M2-2 molecules produced by the G1S-A recombinant had an I6K mutation (78% activity), and G1S-A grew better than G1S-C that had no M2-2 function. Insertion of M2-2 into the F-M2 junction also reduced the level of M2-1 produced in the infected cells because of insufficient termination of the M2-2 transcription. Nevertheless, the reduced M2-1 protein did not have a significant effect on virus replication. Only G9-B and G9-C that had the impaired M2-2 function replicated poorly in HEp-2 cells; G9-A containing the correct M2-2 gene replicated as efficiently as wild-type rA2 in both Vero and HEp-2 cells, despite its reduced M2-1 protein synthesis and increased levels of the M2-2 protein and antigenomic RNA produced in the infected cells.
Except for G1S-C that had a nucleotide insertion at nt 100, resulting in complete loss of M2-2 function, most of the viruses had mixed mutations in the M2-2 gene. The attempts to isolate viruses with homogenous M2-2 gene populations by several additional rounds of plaque purification were not successful, despite the fact that RSV lacking the M2-2 ORF could be obtained. This indicated that M2-2 produced from the M2-2 gene of different mutations might have resulted in a more balanced M2-2 function in the infected cells. Interestingly, most of the mutations were clustered in two regions that were rich in the A and U sequences, as also seen in the RSV G gene (21). The nucleotide insertions or deletions were possibly generated by polymerase stuttering or slippage errors in these regions (7, 18). The NS2, N, P, and M2-1 genes in each of the M2-2 recombinant viruses did not contain any mutations. In addition, the recombinant RSV with the lacZ gene inserted at the 3' promoter proximal position reported previously by our group did not contain any mutation in the lacZ gene (10). These results indicated that the mutations occurred in the inserted M2-2 gene in these recombinant RSV mutants were due to the overexpression of the M2-2 protein instead of a global phenomenon caused by the inserted gene.
Here we showed that the RSV minigenome assay was very useful in screening amino acid residues in M2-2 critical to RSV replication. Deletion of the first five amino acids individually had a minimal effect on its function. The only residue in the N-terminal M2-2 critical to its in vitro function was Ile-6. Deletion of Ile-6 abolished M2-2 function, but replacement of Ile-6 by Ala or Lys was tolerated and resulted in a slight reduction in its M2-2 activity. Unexpectedly, G9-C#4, which lacked the C-terminal 18 amino acids, still maintained most of the M2-2 function as measured by the minigenome assay, which was much better than the clone with four or eight amino acids deleted from its C terminus. The addition of a Phe at position 73 was the accidental result of the insertion of a UUU codon. The M2-2 does not have a Phe at its C-terminal 18 amino acids; this Phe-73 may have performed some function that enabled the protein to have a better conformation compared to the M2-2 molecules with shorter deletions.
The mechanism by which M2-2 influences the viral RNA synthesis process remains to be determined. Most likely, M2-2 needs to interact with one of the viral proteins that are involved in RNA transcription and replication. We examined the interaction of the ICP4-tagged M2-2 and hemagglutinin-tagged L protein by coimmunoprecipitation, but the results were not conclusive (data not shown). Recently, it was found that RSV M2-1 was detected in the nucleus of the infected cells, and it interacted with Rel A protein to induce the activation of nuclear factor kappa B (33). More studies need to be done to investigate whether M2-2 functions through its interaction with a viral protein(s) or a host factor(s).
ACKNOWLEDGMENTS
We thank the tissue culture facility of MedImmune Vaccine for providing tissue culture cells, Chengjun Mo for helping with immunofluorescence microscopy, Winnie Chan for technical assistance, members of Hong Jin's group for assistance and discussions, and Bin Lu, Richard Spaete, and George Kemble for review of the manuscript.
REFERENCES
Abraham, G., and A. K. Banerjee. 1976. Sequential transcription of the genes of vesicular stomatitis virus. Proc. Natl. Acad. Sci. USA 73:1504-1508.
Ahmadian, G., P. Chambers, and A. J. Easton. 1999. Detection and characterization of proteins encoded by the second ORF of the M2 gene of pneumoviruses. J. Gen. Virol. 80:2011-2016.
Ahmadian, G., J. S. Randhawa, and A. J. Easton. 2000. Expression of the ORF-2 protein of the human respiratory syncytial M2 gene is initiated by aribosomal termination-dependent reinitiation mechanism. EMBO J. 19: 2681-2689.
Ball, L. A., and C. N. White. 1976. Order of transcription of genes of vesicular stomatitis virus. Proc. Natl. Acad. Sci. USA 73:442-446.
Barik, S. 1992. Transcription of human respiratory syncytial virus genome RNA in vitro: requirement of cellular factor(s). J. Virol. 66:6813-6818.
Bermingham, A., and P. L. Collins. 1999. The M2-2 protein of human respiratory syncytial virus is a regulatory factor involved in the balance between RNA replication and transcription. Proc. Natl. Acad. Sci. USA 96:11259-11264.
Bilsel, P. A., and S. T. Nichol. 1990. Polymerase errors accumulating during natural evolution of the glycoprotein gene of vesicular stomatitis virus Indiana serotype isolates. J. Virol. 64:4873-4883.
Bossert, B., S. Marozin, and K. K. Conzelmann. 2003. Nonstructural proteins NS1 and NS2 of bovine respiratory syncytial virus block activation of interferon regulatory factor 3. J. Virol. 77:8661-8668.
Buchholz, U. J., S. Finke, and K. K. Conzelmann. 1999. Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter. J. Virol. 73:251-259.
Cheng, X., M. Munoz, H. Zhou, and H. Jin. 2002. Expression of beta-galactosidase by recombinant respiratory syncytial viruses for microneutralization assay. J. Virol. Methods 105:287-295.
Cheng, X., H. Zhou, R. S. Tang, M. G. Munoz, and H. Jin. 2001. Chimeric subgroup A respiratory syncytial virus with the glycoproteins substituted by those of subgroup B and RSV without the M2-2 gene are attenuated in African green monkeys. Virology 283:59-68.
Collins P. L., R. M. Chanock, and B. R. Murphy. 2001. Respiratory syncytial virus, p. 1443-1485. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Collins, P. L., M. G. Hill, J. Cristina, and H. Grosfeld. 1996. Transcription elongation factor of respiratory syncytial virus, a nonsegmented negative-strand RNA virus. Proc. Natl. Acad. Sci. USA 93:81-85.
Collins, P. L., M. G. Hill, and P. R. Johnson. 1990. The two open reading frames of the 22K mRNA of human respiratory syncytial virus: sequence comparison of antigenic subgroups A and B and expression in vitro. J. Gen. Virol. 71:3015-3020.
Collins, P. L., and B. R. Murphy. 2002. Respiratory syncytial virus: reverse genetics and vaccine strategies. Virology 296:204-211.
Collins, P. L., and G. W. Wertz. 1983. cDNA cloning and transcriptional mapping of nine polyadenylylated RNAs encoded by the genome of human respiratory syncytial virus. Proc. Natl. Acad. Sci. USA 80:3208-3212.
Collins, P. L., and G. W. Wertz. 1985. The envelope-associated 22K protein of human respiratory syncytial virus: nucleotide sequence of the mRNA and a related polytranscript. J. Virol. 54:65-71.
Cunningham, P. R., C. J. Weitzmann, and J. Ofengand. 1991. SP6 RNA polymerase stutters when initiating from an AAA... sequence. Nucleic Acids Res. 19:4669-4673.
Fearns, R., and P. L Collins. 1999. Model for polymerase access to the overlapped L gene of respiratory syncytial virus. J. Virol. 73:388-397.
Fearns, R., and P. L. Collins. 1999. Role of the M2-1 transcription antitermination protein of respiratory syncytial virus in sequential transcription. J. Virol. 73:5852-5864.
Garcia-Barreno, B., A. Portela, T. Delgado, J. A. Lopez, and J. A. Melero. 1990. Frame shift mutations as a novel mechanism for the generation of neutralization resistant mutants of human respiratory syncytial virus. EMBO J. 9:4181-4187.
Grosfeld, H., M. G. Hill, and P. L. Collins. 1995. RNA replication by respiratory syncytial virus (RSV) is directed by the N, P, and L protein; transcription also occurs under these conditions but requires RSV superinfection for efficient synthesis of full-length mRNA. J. Virol. 69:5677-5686.
Hardy, R. W., S. B. Harmon, and G. W. Wertz. 1999. Diverse gene junctions of respiratory syncytial virus modulate the efficiency of transcription termination and respond differently to M2-mediated antitermination. J. Virol. 73:170-176.
Harmon, S. B., A. G. Megaw, and G. W. Wertz. 2001. RNA sequences involved in transcriptional termination of respiratory syncytial virus. J. Virol. 75:36-44.
Huang, Y. T., P. L. Collins, and G. W. Wertz. 1985. Characterization of the 10 proteins of human respiratory syncytial virus: identification of a fourth envelope-associated protein. Virus Res. 2:157-173.
Hubenthal-Voss, J., R. A. Houghten, L. Pereira, and B. Roizman. 1988. Mapping of functional and antigenic domains of the alpha 4 protein of herpes simplex virus 1. J. Virol. 62:454-462.
Jin, H., X. Cheng, H. Z. Zhou, S. Li, and A. Seddiqui. 2000. Respiratory syncytial virus that lacks open reading frame 2 of the M2 gene (M2-2) has altered growth characteristics and is attenuated in rodents. J. Virol. 74: 74-82.
Jin, H., D. Clarke, Z.-Y.H. Zhou, et al. 1998. Recombinant human respiratory syncytial virus (RSV) from cDNA and construction of subgroup A and B chimeric RSV. Virology 251:206-214.
Jin, H., B. Lu, H. Zhou, et al. 2003. Multiple amino acid residues confer temperature sensitivity to human influenza virus vaccine strains (FluMist) derived from cold-adapted A/Ann. Arbor/6/60. Virology 306:18-24.
Jin, H., H. Zhou, X. Cheng, R. Tang, M. Munoz, and N. Nguyen. 2000. Recombinant respiratory syncytial viruses with deletions in the NS1, NS2, SH, and M2-2 genes are attenuated in vitro and in vivo. Virology 273: 210-218.
Kuo, L., H. Grosfeld, J. Cristina, M. G. Hill, and P. L. Collins. 1996. Effects of mutations in the gene-start and gene-end sequence motifs on transcription of monocistronic and dicistronic minigenomes of respiratory syncytial virus. J. Virol. 70:6892-6901.
Moudy, R. M., W. M. Sullender, and G. W. Wertz. 2004. Variations in intergenic region sequences of human respiratory syncytial virus clinical isolates: analysis of effects on transcriptional regulation. Virology 327: 121-133.
Reimers, K., K. Buchholz, and H. Werchau. 2005. Respiratory syncytial virus M2-1 protein induces the activation of nuclear factor kappa B. Virology 331:260-268.
Schlender, J., B. Bossert, U. Buchholz, and K. K. Conzelmann. 2000. Bovine respiratory syncytial virus nonstructural proteins NS1 and NS2 cooperatively antagonize alpha/beta interferon-induced antiviral response. J. Virol. 74:8234-8242.
Spann, K. M., K. C. Tran, B. Chi, R. L. Rabin, and P. L. Collins. 2004. Suppression of the induction of alpha, beta, and lambda interferons by the NS1 and NS2 proteins of human respiratory syncytial virus in human epithelial cells and macrophages. J. Virol. 78:4363-4369.
Sutherland, K. A., P. L. Collins, and M. E. Peeples. 2001. Synergistic effects of gene-end signal mutations and the M2-1 protein on transcription termination by respiratory syncytial virus. Virology 288:295-307.
Tang, R. S., N. Nguyen, X. Cheng, and H. Jin. 2001. Requirement of cysteines and length of the human respiratory syncytial virus M2-1 protein for protein function and virus viability. J. Virol. 75:11328-11335.
Teng, M. N., and P. L. Collins. 1998. Identification of the respiratory syncytial virus proteins required for formation and passage of helper-dependent infectious particles. J. Virol. 72:5707-5716.
Teng, M. N., and P. L. Collins. 1999. Altered growth characteristics of recombinant respiratory syncytial viruses which do not produce NS2 protein. J. Virol. 73:466-473.
Teng, M. N., S. S. Whitehead, A. Bermingham, M. St. Claire, W. R. Elkins, B. R. Murphy, and P. L. Collins. 2000. Recombinant respiratory syncytial virus that does not express the NS1 or M2-2 protein is highly attenuated and immunogenic in chimpanzees. J. Virol. 74:9317-9321.
Yu, Q., R. W. Hardy, and G. W. Wertz. 1995. Functional cDNA clones of the human respiratory syncytial (RS) virus N, P, and L proteins support replication of RS virus genomic RNA analogs and define minimal trans-acting requirements for RNA replication. J. Virol. 69:2412-2419.(Xing Cheng, HyunJung Park)