![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Journal of Virology, December 2004, p. 12877-12887, Vol. 78, No. 23
0022-538X/04/$08.00+0 DOI: 10.1128/JVI.78.23.12877-12887.2004
Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland
Received 24 May 2004/ Accepted 8 July 2004
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SH, G, and SH/G deletion mutants were readily recovered and were found to replicate efficiently during multicycle growth in cell culture. Thus, the SH and G proteins are not essential for growth in cell culture. Apart from the absence of the deleted protein(s), the virions produced by the gene deletion mutants were similar by protein yield and gel electrophoresis protein profile to wild-type HMPV. When administered intranasally to hamsters, the G and SH/G mutants replicated in both the upper and lower respiratory tracts, showing that HMPV containing F as the sole viral surface protein is competent for replication in vivo. However, both viruses were at least 40-fold and 600-fold restricted in replication in the lower and upper respiratory tract, respectively, compared to wild-type rHMPV. They also induced high titers of HMPV-neutralizing serum antibodies and conferred complete protection against replication of wild-type HMPV challenge virus in the lungs. Surprisingly, G is dispensable for protection, and the G and SH/G viruses represent promising vaccine candidates. In contrast, SH replicated somewhat more efficiently in hamster lungs compared to wild-type rHMPV (20-fold increase on day 5 postinfection). This indicates that SH is completely dispensable in vivo and that its deletion does not confer an attenuating effect, at least in this rodent model. |
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The HMPV genome, which ranges in length from 13,280 to 13,335 nucleotides (nt) for the available complete sequences (1), contains eight genes in the order 3'-N-P-M-F-M2-SH-G-L-5' and encodes nine proteins (1, 29). By analogy to HRSV, the HMPV proteins are the following: N, nucleoprotein; P, phosphoprotein; M, matrix protein; F, fusion protein; M2-1, transcription elongation factor; M2-2, RNA synthesis regulatory factor; SH, small hydrophobic protein of unknown function; G, attachment glycoprotein; and L, viral polymerase. Compared to HRSV, HMPV lacks the nonstructural NS1 and NS2 genes and has the F, M2, SH, and G genes in the order F-M2-SH-G, compared to SH-G-F-M2 for HRSV.
Two genetic subgroups have been described for HMPV which are 80% identical on the nucleotide level and 90% identical on the aggregate amino acid level (1). These values are similar to the 81% nucleotide identity and 88% aggregate amino acid sequence identity observed for the two subgroups of HRSV (A2 and B1 strains; GenBank accession nos. M74568 and AF013254, respectively). It was initially reported that these two HMPV genetic subgroups represented two serotypes (31). However, other studies indicated that the known isolates of HMPV formed a single serotype and that the two genetic subgroups exhibited less antigenic diversity than was the case for the two subgroups of HRSV (15, 22). Members of the Pneumovirinae subfamily (hereafter called pneumoviruses) encode three surface glycoproteins, F, G, and SH (5). The fusion glycoprotein F, which is highly conserved (95% identity between the two HMPV subgroups, compared to 89% between the HRSV subgroups), mediates penetration and syncytium formation. The human and bovine RSV F proteins are synthesized as an F0 precursor that is cleaved intracellularly by a furin-like protease at two closely spaced sites to yield the F1 and F2 subunits (5) plus an intervening peptide (9, 34). The HMPV F0 protein contains a single candidate cleavage site whose structure, R-Q-X-R, does not conform to the furin motif, and HMPV usually has been reported to require exogenous trypsin for growth in vitro. The HRSV F protein is one of the two major neutralization and protective antigens of HRSV, and HMPV F also appears to be an important HMPV neutralization and protective antigen (22, 24).
The pneumovirus G protein is a type II membrane protein, such that its membrane anchor is proximal to the N terminus and its C terminus is oriented externally. The G protein of HRSV has been studied in the greatest detail among the pneumoviruses. Most of its ectodomain has a high content of serine, threonine, and proline residues and numerous O- and N-linked sugars, features that are reminiscent of mucins and might confer an extended, nonglobular structure (5). The HRSV G protein is 289 to 299 amino acids (aa) in length and is highly divergent among HRSV strains (55% identity between the two HRSV antigenic subgroups) (11, 23), a characteristic that is even more pronounced for HMPV G (31 to 35% identity between the two proposed genetic subgroups), with an amino acid length of 217 to 236 (1, 18). The high carbohydrate content of HRSV G results in an apparent molecular mass as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of 84 to 90 kDa compared to 32.6 kDa for the unmodified form calculated from the gene sequence. Remarkably, HRSV G is not essential for virus assembly (27) or growth in vitro. Indeed, a biologically derived, cold-passaged, attenuated HRSV subgroup B mutant (B1 cp-52) was shown to replicate to high titers in cell culture despite the absence of functional SH and G proteins (13). Moreover, recombinant HRSV (rHRSV) from which the G gene alone was deleted was infectious but had an in vitro host range restriction and was highly attenuated in the respiratory tract of mice. Virus could not be recovered from the nasal turbinates, only trace amounts were detected in the lungs, and it was uncertain whether replication had occurred (26, 28). Comparable results were observed for a recombinant bovine RSV (BRSV) lacking its G gene which was fully competent with regard to in vitro growth and strongly attenuated in the respiratory tract of calves (12, 21).
The pneumovirus SH protein also is a type II transmembrane glycoprotein (6). The HMPV SH protein is the largest among the pneumoviruses (179 aa for the Canadian HMPV isolate CAN97-83 [HMPV83], versus 175 aa for avian metapneumovirus(MPV) C, 174 aa for AMPV A, 81 aa for BRSV, and 64 aa for HRSV A2) (1, 29). The HMPV SH protein is poorly conserved (59% identity between the two proposed genetic subgroups) and is more divergent than its HRSV counterpart (72% identity between subgroups). In HRSV-infected cells, SH accumulates in several different forms: SH0 (mass, 7.5 kDa) is the full-length nonglycosylated species, SHg (13 to 15 kDa) contains one high-mannose N-linked carbohydrate chain, SHp (21 to 30 kDa) is generated by the addition of polylactosaminoglycan to the N-linked carbohydrate chain, and SHt (4.8 kDa) is a nonglycosylated species that is derived from an internal initiation (17). SH0 and SHp are the predominant forms in the virion. On the basis of chemical cross-linking (6) and global searching of molecular dynamic simulations (14), SH was suggested to form channel-like homopentamers whose minimal transmembrane pore is 1.46 Å. These observations are reinforced by the observation that expression of the SH protein in bacteria increases membrane permeability to small-molecular-weight compounds (20). However, the function of SH in the viral infectious cycle still remains unknown. Recombinant bovine and human RSV from which the SH gene has been deleted (SH) are fully viable in cell culture (4, 12). In vivo, the HRSVSH virus was attenuated in the upper, but not in the lower, respiratory tract of mice and was slightly attenuated at both sites in chimpanzees (4, 33).
The ability to produce infectious rHMPV from full-length cDNA (2) provides a method to investigate the functions of individual HMPV proteins and to develop live-attenuated vaccines. As a first step for functional studies and the development of attenuated HMPV derivatives, we engineered a full-length cDNA copy of the HMPV antigenome to delete the SH, G, or both SH and G genes together in their entirety. Recombinant virus was recovered and analyzed in vitro and by infection in hamsters.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Construction of SH and/or G gene deletion mutant cDNA.
The plasmid pHMPV, which contains the complete consensus antigenomic sequence of HMPV83 (GenBank accession number AY297749 [1]) except for four nucleotide substitutions involved in the NheI site (nt 3038) as described previously (2), was modified to contain an additional BsiWI and BsrGI restriction site in the M2-SH and SH-G intergenic regions, respectively (Fig. 1). To introduce the restriction sites, the NheI/Acc65I fragment of pHMPV, containing the F, M2, SH, and G genes, was subcloned, and a two-step site-directed mutagenesis was done using the QuikChange kit (Stratagene). First, the BsiWI restriction site (nt 5459) was created using two complementary primers CTTAAGTTAGTAAAAACACGTACGAGTGGGATAAGTGACAATG and CATTGTCACTTATCCCACTCGTACGTGTTTTTACTAACTTAAG (mutated nucleotides are in bold, and the BsiWI restriction enzyme site is underlined). Next, the BsrGI restriction site (nt 6099) was created by using the complementary primers GTTTAGTTATTTTAAAATATTTGTACATAGGTAAGTTTCTATGGCAC and GTGCCATAGAAACTTACCTATGTACAAATATTTTAAAATAACTAAAC (mutated nucleotides are in bold, and the BsrGI restriction enzyme site is underlined). The mutated subclone was then digested with NheI/BsiWI and the resulting fragment was cloned into the NheI/Acc65I window of pHMPV to create pHMPV-SH/G, or it was digested with NheI/BsrGI and the resulting fragment was cloned into the NheI/Acc65I window of pHMPV to create pHMPV-G. To create pHMPV-SH, the mutated subclone was digested with BsiWI/BsrGI, religated, and then digested with NheI/Acc65I, and the resulting fragment was cloned in the NheI/Acc65I window of pHMPV. The strategy for these ligations was based on the knowledge that digestion with BsiWI, BsrGI, and Acc65I creates a common overhang, GTAC. The sequence of the NheI/Acc65I fragment of each modified pHMPV was confirmed by nucleotide sequencing performed with an ABI 3100 automatic sequencer, using the Big-Dye Terminator Ready Reaction kit version 1.1 (Applied Biosystems) and specific primers.
|
SH, pHMPV-G, or pHMPV-SH/G, 2 µg each of pT7-N and pT7-P, and 1 µg each of pT7-M2-1 and pT7-L support plasmids per well. Transfections were done with Lipofectamine 2000 (Invitrogen) in OptiMEM without trypsin or serum (day 1) and maintained overnight at 32°C. The Lipofectamine transfection medium was removed (day 2) and replaced with Glasgow MEM without trypsin or serum. Trypsin was added on day 3 to a final concentration of 5 µg/ml, and cell-medium mixtures were passaged onto fresh LLC-MK2 cells on day 4 and incubated at 32°C. The recovered rHMPV, rHMPVSH (SH), rHMPVG (G), and rHMPVSH/G (SH/G) viruses then were amplified by two passages in LLC-MK2 cells, and only low-passage stocks were used in all experiments.
Verification of recombinant viral genome structures.
Viral RNA was extracted from each virus stock corresponding to the second passage of rHMPV, SH, G, or SH/G (QIAamp viral RNA kit; QIAGEN). The viral RNA was copied and amplified by reverse transcription-PCR with a positive-sense primer designed to hybridize within the M2 gene (HMPV83 nt 5363 to 5382) and a negative-sense primer designed to hybridize within the L gene (HMPV83 nt 7499 to 7479). PCR products were analyzed on a 1% agarose gel and were also subjected to direct nucleotide sequencing in their entirety.
Northern blot analysis.
LLC-MK2 cells were mock infected or infected with rHMPV, SH, G, or SH/G virus, each at an input multiplicity of infection (MOI) of 3 PFU per cell, incubated for 3 days at 32°C in the presence of 5 µg of trypsin/ml, and processed to purify total intracellular RNA using an RNeasy kit (QIAGEN) according to the manufacturer's instructions. RNA was electrophoresed in a 1% agarose gel in the presence of 0.44 M formaldehyde, transferred to a charged nylon membrane (Hybond-N+; Amersham Biosciences), fixed by UV cross-linking, and analyzed by hybridization with denatured double-stranded 32P-labeled DNA probes generated by random priming from F, SH, or G gene cDNA fragments (Megaprime DNA labeling system; Amersham Biosciences). Radioactivity was detected by analysis with a phosphorimager.
Virus purification by sucrose gradient centrifugation.
LLC-MK2 cells were infected with either HMPV83, rHMPV, SH, G, or SH/G virus (0.1 PFU per cell), incubated for 10 days at 32°C in the presence of 5 µg of trypsin/ml, and harvested by scraping. The cell suspensions were adjusted to contain 50 mM HEPES (pH 7.5) and 0.1 M MgSO4 and clarified by low-speed centrifugation at 2,000 rpm for 20 min at 4°C in a Beckman Coulter Allegra 6R centrifuge. The cell pellet in a small volume of medium was subjected to three cycles of freezing and thawing, the mixture was clarified, the supernatant fluids were combined, and the virus was pelleted by centrifugation at 7,000 rpm in a Beckman JLA rotor overnight at 4°C. The virus pellet was resuspended in OptiMEM containing 50 mM HEPES buffer and 0.1 M MgSO4, layered on a 30-to-60% (wt/vol) discontinuous sucrose gradient, and centrifuged for 90 min at 26,000 rpm at 4°C in an SW28 rotor. The virus-containing band at the 30%-60% interface was collected and diluted in TEN (10 mM Tris-HCl [pH 7.4], 0.1 M NaCl, 1 mM EDTA), and virus particles were pelleted by centrifugation at 25,000 rpm in an SW28 rotor for 90 min at 4°C. Finally, the virus pellet was resuspended in TEN, and the protein yield of each preparation was quantified using the BCA protein assay kit (Pierce).
Protein electrophoresis and Western blot assay. Aliquots of sucrose-purified virus (10 µg for Coomassie blue staining or 1.5 µg for Western blot analysis) were reduced and denatured by heating at 99°C for 5 min and analyzed on a 4-to-20% polyacrylamide bis-Tris gel (NuPAGE Novex bis-Tris; Invitrogen). The separated proteins were stained with a Coomassie blue staining kit (Invitrogen) or electrotransferred onto a polyvinylidene difluoride membrane (Invitrolon; Invitrogen). The membrane was blocked for 2 h at room temperature in Western blocking reagent (Roche) and then incubated with a rabbit antiserum raised either against peptide G1-17, representing aa 1 to 17 of the HMPV83 G N terminus, against peptide G203-219, representing aa 203 to 219 of G (the C terminus), or against peptide SH82-96, representing aa 82 to 96 of SH. HMPV F was detected using a hyperimmune serum produced by immunization of hamsters with rHPIV1 expressing HMPV83 F from an added gene (22). Bound antibodies were visualized by binding with horseradish peroxidase-conjugated goat anti-rabbit or anti-hamster immunoglobulin G antibodies (Kirkegaard & Perry Laboratories, Gaithersburg, Md.) followed by chemiluminescence (SuperSignal West Pico chemiluminescent substrate; Pierce). The Coomassie blue-stained gel and Western blot autoradiographies were scanned using a densitometer (Personal Densitometer SI; Molecular Dynamics) to quantify HMPV proteins.
Determination of replication and protective efficacy of SH, G, and SH/G viruses in hamsters.
Groups of 6-week-old Golden Syrian hamsters were inoculated intranasally, under light methoxyflurane anesthesia, with 0.1 ml of L15 medium containing 5.7 log10 TCID50 of HMPV83, rHMPV, SH, or SH/G virus or 5.2 log10 TCID50 of rHMPV or G virus. On days 3 and 5 postinfection, the lungs and nasal turbinates were harvested from six animals from each group, and the virus titer of individual specimens was quantified by serial dilution of tissue homogenates, as described above. To determine the level of immunogenicity and protective efficacy of the recombinants, additional hamsters in groups of six were infected with 5.7 log10 TCID50 of rHMPV, SH, or SH/G virus or 5.2 log10 TCID50 of G virus in a 0.1-ml inoculum or with L15 medium diluent. Sera were collected 2 days prior to infection and 27 days postinfection. On day 28 postinfection, hamsters were challenged by intranasal administration of 5.7 log10 TCID50 of HMPV83 in a 0.1-ml inoculum. Nasal turbinates and lungs were harvested 3 days later, and the titer of virus in each tissue homogenate was determined as described above.
HMPV serum neutralization assays. The titers of HMPV-neutralizing antibodies were determined using an endpoint dilution neutralization assay as described previously (22). Briefly, 75 µl of MEM containing approximately 240 TCID50 of HMPV83 was mixed with an equal volume of MEM containing serial twofold dilutions of hamster serum that had been heat inactivated at 56°C for 30 min. The virus-antibody mixture was incubated at 37°C for 1 h, and 50 µl of this mixture was then transferred to each of two wells of a 96-well plate containing LLC-MK2 cells. After 1 h at 37°C the virus-antibody mixture was removed, and the monolayers were washed twice and incubated at 32°C for 8 days in OptiMEM supplemented with trypsin. Infected cultures were detected by immunoperoxidase staining with polyclonal rabbit anti-HMPV serum, as described previously (2). The neutralization titer was defined as the highest dilution of antibody at which half of the cultures were negative for infection.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SH/G), 165 nt (the SH-L intergenic region of SH), and 122 nt (the M2-G intergenic region of G), which are intermediate in size between the naturally occurring G-L (190 nt) and SH-G (124 nt) intergenic regions (Fig. 1). For the genes in the vicinity of the deletions, each one that was preceded or followed in wild-type HMPV by either a short or long intergenic region was similar in the gene deletion mutants (e.g., G was always preceded by and followed by long intergenic regions, SH was always preceded by a short intergenic region and followed by a long one, and L was always preceded by a long intergenic region). The HMPV antigenomes expressed by the respective plasmids were 12,695 (SH), 12,475 (G), and 11,835 nt (SH/G) in length, compared with the 13,335 nt for the parental rHMPV. rHMPVs were readily recovered by transfection of BSR T7/5 cells with each respective full-length plasmid together with the four support plasmids encoding the HMPV N, P, M2-1, and L proteins (2, 3). The recombinant viruses were amplified by two passages in LLC-MK2 cells, and two-passage stocks were used in all experiments. Viral RNA from each gene deletion mutant was subjected to reverse transcription-PCR using primers specific to the end of the M2 gene and the beginning of the L gene. In each case, the product obtained was consistent with the expected deletion and was dependent on the reverse transcription step and, hence, was derived from RNA rather than from contaminating DNA (data not shown). Furthermore, each fragment was completely sequenced, confirming the expected structure.
Multicycle growth in vitro of the gene deletion mutants.
The SH, G, and SH/G viruses were compared to wild-type rHMPV with regard to the efficiency of multicycle replication in LLC-MK2 cells following infection with an input of 0.01 PFU per cell (Fig. 2). As previously shown, biologically derived HMPV83 and rHMPV replicated with similar efficiencies in vitro (2). rHMPV containing the deletion of a single gene, either SH or G, replicated as well as or somewhat more efficiently than rHMPV. The final titers reached by SH and G viruses were 1.7 x 107 and 1.1 x 107 PFU per ml, respectively, compared to 7.8 x 106 PFU per ml for rHMPV. Thus, these genes are not essential for replication in vitro, and the loss of the SH and G genes seems to marginally improve the viral replicative fitness in vitro, particularly in the case of the SH virus. rHMPV lacking both SH and G replicated with an efficiency reduced 5.5-fold on day 8 postinfection compared to that of rHMPV, but the final titer reached by SH/G was similar to that of rHMPV, 5.3 x 106 versus 7.8 x 106 PFU per ml.
|
SH, G, and SH/G viruses were shorter in length, reflecting the gene deletions. The Northern blotting performed with the F probe showed that there were reduced levels of vRNA as well as F mRNA for the gene deletion mutants, particularly for the SH/G mutant. This was somewhat variable between different experiments performed with the same RNA preparations and likely reflected nonhomogenous RNA transfer in this series of Northern blot assays.
|
SH virus and SH-L and M2-SH-L for G virus. Analysis of proteins in purified virions of the gene deletion viruses. Since the HMPV SH and G proteins are presumed to be structural proteins, by analogy to HRSV, it was of interest to compare purified virions of the gene deletion viruses and wild-type HMPV by gel electrophoresis, direct protein staining, and Western blot analysis. LLC-MK2 cells were infected with HMPV and, 10 days postinfection, the medium supernatants were harvested and viruses were concentrated and purified by sucrose gradient centrifugation. In one purification, we assayed the titer of infectious particles at different stages in the process and found that the total number of PFU was essentially undiminished by the purification process, suggesting that the particles remained largely intact. Ten micrograms of each sucrose-purified virus protein was separated on a 4 to 20% polyacrylamide gel and visualized by Coomassie blue staining (Fig. 4). The major protein bands visualized by staining were identified as the N, P, and M2-1 proteins by Western blot analysis of lysates of BSR T7/5 cells transfected with T7-driven plasmids expressing either N, P, or M2-1 and analysis with a rabbit antiserum against purified HMPV83 virions (data not shown). The SH, G, and F proteins could not be unequivocally detected, due to low abundance or inefficient staining with Coomassie blue, as is characteristic of glycoproteins. Overall, the pattern of viral proteins detected by direct staining was very similar for each of the gene deletion viruses (Fig. 4), with minor differences that appeared to represent experimental variability. Densitometer quantification of the N, P, M, and M2-1 proteins indicated that they were incorporated in approximately the same proportion in each of the different viruses (data not shown), and the sum of the four proteins was approximately the same for each virus (Fig. 4). This indicated that that deletion of the SH and/or G genes did not result in a dramatic change in virion composition or yield.
|
G, and absent as expected in the SH and SH/G viruses. The 23-kDa species (designated SH0) was the appropriate size to be the complete nonglycosylated SH protein (calculated mass of 20.6 kDa), while the other two species (designated SHg1 and SHg2) presumably are glycosylated forms, by analogy to HRSV. Based on its electrophoretic difference relative to SH0, SHg1 might contain one or two N-linked carbohydrate side chains: the predicted sequence of the HMPV83 SH protein contains three potential Asn-X-Ser/Thr acceptor sites in the predicted ectodomain (1). Based on its high molecular mass and diffuse pattern, SHg2 might correspond to the polylactosaminoglycan-containing form that has been identified for HRSV. These three forms also were detected by Western blotting with a rabbit serum raised against peptide SH1-15 representing aa 1 to 15 of the HMPV83 SH N terminus (data not shown). The amount of SH protein in G virions was consistently low compared to that in HMPV83 and rHMPV, amounting to 26% of that in the wild-type viruses (mean of three independent experiments). This was reminiscent of results with HRSV, where the incorporation of SH protein into virions of a G-deletion mutant was 25% that of wild type (25).
|
To detect HMPV F protein in purified virions, Western blots were incubated with hamster hyperimmune serum obtained by immunization with rHPIV1 expressing HMPV83 F as an added gene. Under nonreducing, nondenaturing conditions, similarly abundant amounts of F-specific signals were detected (Fig. 5D), spanning over a high-molecular-mass range of 120 to 220 kDa and higher. These might represent dimeric and trimeric forms of F. Additionally, a more discrete band was detected at a molecular mass of about 58 to 60 kDa, consistent with the expected molecular mass of the predicted disulfide-linked F1 and F2 subunits (or their uncleaved F0 precursor). When analyzed under reducing conditions (Fig. 5E), F-specific signals were detected at a molecular mass of about 47 kDa, which would be the expected size of the F1 subunit. This showed that the 58-to-60-kDa band as well as the high-molecular-mass putative F multimers detected under nonreduced conditions consist of disulfide-linked F1 and F2 rather than uncleaved F0. Using the F-specific hyperimmune serum, we were not able to detect any low-molecular-weight protein that could be identified as F2 (data not shown).
Replication, immunogenicity, and protective efficacy following intranasal administration to hamsters.
Golden Syrian hamsters were infected intranasally either with 5.7 log10 TCID50 of HMPV83, rHMPV, SH, or SH/G or with 5.2 log10 TCID50 of rHMPV or G (Table 1). The G virus was administered at a lower dose because we wished to use the same low-passage stock that had been used for sequencing and most of the other analyses, and it was of lower titer. Thus, an additional group of animals were administered rHMPV in parallel at the same lower dose for direct comparison. Six animals from each group were sacrificed 3 and 5 days later, the nasal turbinates and lungs were harvested, and viral titers were determined (Table 1). Recombinantly derived rHMPV replicated in vivo with an efficiency similar to that of its biologically derived parent, HMPV83: the titer of rHMPV was 20- and 3-fold lower in the nasal turbinates and lungs, respectively, on day 3, but there was no significant difference on day 5. Thus, rHMPV appears to have wild-type-like growth properties in vivo as well as in vitro and represents a suitable starting point for developing attenuated derivatives as candidate vaccines. Also, the two different doses of rHMPV used in this study were similar in terms of the level of replication in the upper and lower respiratory tracts, allowing comparison between G virus and the other viruses that were inoculated at a higher dose.
|
SH virus replicated somewhat better than its wild-type counterpart, especially in the lungs, where a 20-fold increase was detected on day 5 postinfection compared to wild-type rHMPV. In contrast, the G and SH/G viruses were strongly attenuated. In the lungs, replication of the G and SH/G viruses was significantly reduced on day 3 (at least 40- and 100-fold, respectively) compared to the corresponding dose of wild-type rHMPV, but it was not significantly reduced on day 5 postinfection. Virus was recovered from the lungs of every animal on each day. In the upper respiratory tract, replication of the G and SH/G viruses was not detected in 33 and 17% of the animals, respectively, on day 3 but was detected in all animals on day 5. The level of replication of the G and SH/G viruses in the upper respiratory tract on both days was reduced more than 600- and 300-fold, respectively.
In order to evaluate immunogenicity and protective efficacy, the gene deletion viruses and their wild-type recombinant parent were administered intranasally to additional hamsters. Serum samples were taken 27 days later and analyzed in a neutralization assay (Table 2). The highly attenuated G and SH/G viruses induced high titers of HMPV-neutralizing serum antibodies, which were less than sixfold reduced compared to those induced by rHMPV. On day 28 postimmunization, the animals were challenged intranasally with wild-type HMPV83, and the animals were sacrificed 3 days later. Nasal turbinates and lungs were harvested, and virus titers were determined by limiting dilution (Table 2). Challenge virus could not be detected in the nasal turbinates or lungs of any of the animals that had previously been infected with the rHMPV or SH viruses. The animals that had been infected with the G or SH/G virus also were protected from challenge virus replication in the lungs. In the nasal turbinates, only 40% of the animals that had been immunized with SH/G had detectable challenge virus, and the mean titer of challenge virus was reduced more than 3,000-fold. Challenge virus was detected in the nasal turbinates of all of the animals that had been immunized with the G virus, but the mean titer was reduced by more than 300-fold compared to challenge virus replication in the mock-immunized group. Thus, despite their strongly attenuated nature, the G and SH/G viruses were immunogenic and highly protective against HMPV challenge.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The amino acid sequence of the HMPV G protein deduced from the nucleotide sequence of its gene suggests that it is a type II mucin-like glycosylated protein with similarities to the G protein of HRSV, representing a separate genus in subfamily Pneumovirinae. The size of the HMPV G protein determined here by Western blot analysis suggested that it is heavily glycosylated, approximately to the same extent as HRSV G. The HMPV G protein detected in purified virions with antibodies specific to its C terminus had a high molecular mass of 80 to 100 kDa, compared to a value of 23.7 kDa calculated for its unmodified polypeptide backbone deduced from the gene sequence. Western blot analysis of HMPV virions with antibodies specific to the N terminus of G detected a ladder of C-terminally truncated forms of G that might represent proteolytic degradation products due to the trypsin in the medium. The finding that C-terminally truncated forms purify with virions, whereas no N-terminally truncated fragments were found, is consistent with the idea that HMPV G indeed is anchored in the membrane by its N-terminally proximal hydrophobic domain. The apparent sensitivity of HMPV G to proteolysis at multiple sites throughout its backbone would be consistent with an exposed, extended, nonglobular structure.
The SH protein of HMPV is substantially longer (179 versus 64 aa) than that of HRSV, the pneumovirus SH protein that has been characterized in the greatest detail. However, the HMPV SH protein has similar characteristics to that of HRSV, including a high percentage of threonine and serine residues and a similar hydrophilicity profile (1, 29). Western blot analysis of the SH protein present in HMPV virions provided evidence of multiple forms that appeared to correspond to some of those of HRSV. These include a candidate to be the complete unglycosylated SH protein (SH0, 23 kDa), a candidate to be the N-glycosylated SH protein (SHg1, 25 to 30 kDa), and a candidate to be a more extensively glycosylated form (SHg2, 80 to 220 kDa and higher). Moreover, in the absence of the G protein, the HMPV SH protein was found to be less efficiently incorporated in virion particles (26% of that in the wild-type virus), which also was seen for the HRSV G deletion mutant (25). Recent data showed that the HRSV F, G, and SH glycoproteins can form an oligomeric complex within the HRSV-infected cell (7). It may be that, within this proposed complex, interactions between G and SH occur such that the absence of G results in a less efficient incorporation of SH at budding sites. To date, the function of the pneumovirus SH protein remains unclear, but the ability to delete HMPV SH with no detrimental effect on replication in vitro or in vivo indicates that it does not play an irreplaceable role in attachment or entry. The HRSV SH protein appears to assemble into homopentamers, and a channel-like activity is suggested by the increased membrane permeability to low-mass compounds observed when the SH protein is expressed in bacteria (6, 14, 20). Thus, HRSV SH protein appears to have characteristics of a viroporin, which modifies membrane permeability (8). By analogy with HRSV, the HMPV SH protein also is a candidate to be a viroporin, which would be the longest known member to date. However, its permeabilization capacity remains to be proven.
In cell culture, rHMPV viruses containing the deletion of a single gene, either SH or G, replicated as well as or somewhat more efficiently than the wild-type virus, indicating that these genes are not essential for replication in vitro. The loss of SH in particular seemed to improve the replicative fitness. This may reflect a growth advantage due to a shorter length of the genome and/or loss of a transcriptional unit increasing the expression of the downstream gene. It also is possible that the absence of the SH or G protein might somehow improve growth. For example, it was found for HRSV that deletion of the SH gene enhanced the rate of virion entry, cell-to-cell fusion, and plaque size, suggesting an inhibitory effect of the SH protein on F protein activity (25). Effects due to ablating expression of the SH and G proteins versus changing the genome length and number of genes will be evaluated in the future by constructing SH and G viruses in which the deleted gene is replaced by a stuffer that is identical with regard to intergenic and gene start and gene end signals and which contains a foreign sequence of identical size that does not encode a protein. For the present, we are interested in the expeditious development and characterization of possible vaccine candidate viruses, where a stuffer gene would be unacceptable for further development for clinical testing. The present results show that the HMPV F protein alone is sufficient to mediate attachment and fusion in the absence of the other surface proteins. If HMPV G protein has a functional role for attachment, as postulated, it would not be the sole attachment protein. Moreover, HMPV G and SH are not required for the efficient assembly or release of progeny virus, as also is the case for bovine and human RSV (12, 13).
When inoculated intranasally into hamsters, recombinantly derived HMPV replicated less efficiently on day 3, but not on day 5 postinfection, compared to its biologically derived HMPV83 parent. The recombinant virus was designed and confirmed to be identical to the consensus sequence of the biological virus, apart from four nucleotide substitutions involved in creating the NheI marker site, but the slight difference might be explained by the fact that the clinical isolate HMPV83 has not been plaque purified and thus might contain minority variants that contribute disproportionately to growth in vivo. However, the difference was not great, and rHMPV appeared to have wild-type-like growth properties in vivo as well as in vitro (2) and thus represents a suitable starting point for developing attenuated derivatives as candidate vaccines. The replication of the SH virus in hamsters was not significantly reduced in the upper respiratory tract and, surprisingly, was increased 20-fold in the lower respiratory tract on day 5 postinfection. Moreover, it was equivalent to its wild-type parent in the ability to induce neutralizing serum antibodies and to confer complete protection against a subsequent wild-type HMPV infection. Therefore, the HMPV SH gene appeared to be completely dispensable for growth in vivo in hamsters. This may represent an unusual situation in which deletion of a viral gene actually improves growth, but this will need to be verified in additional experiments, including with a model closer to the natural host. This clearly differs from the situation with HRSV, where virus that lacked the SH gene was reduced 10-fold in replication in the upper respiratory tract of mice (4) and was moderately attenuated in the respiratory tract of chimpanzees (33).
The HMPV mutants lacking G, G and SH/G, were reduced in replication by at least 600- and 40-fold in the upper and lower respiratory tracts of hamsters, respectively, compared to that with rHMPV on day 3 postinfection. In this experiment, the G and SH/G viruses were recovered 3 days postinfection from the upper respiratory tract of 67 and 83% of the animals, respectively, and from the lower respiratory tract of 100% of the animals. Since each animal shed virus on day 5 postinfection, both G and SH/G unambiguously replicated in all infected animals. This is somewhat different from the situation with the G-deletion rHRSV mutant in mice, where virus was not recovered from the upper respiratory tract and a very low titer was recovered from the lower respiratory tract of 60% of the animals, indicating that rHRSV G is overattenuated in mice and might not replicate at all since the trace amount that was recovered might have represented input virus (28). Similarly, it was not clear that the BRSV G deletion mutant was able to replicate in the bovine respiratory tract (21). Thus, HMPV provides a clear demonstration that the F protein alone as a virion surface protein is sufficient for replication in vivo, albeit at a reduced level.
Immunization of animals with either the G or SH/G virus induced a high titer of HMPV-neutralizing serum antibodies, even though these viruses were highly attenuated. Importantly, the animals were completely protected in the lungs against a subsequent HMPV83 challenge, and only a moderate level of challenge virus replication was detected in the nasal turbinates. Thus, despite their strongly attenuated phenotype in vivo, the G and SH/G viruses were immunogenic and highly protective against HMPV challenge and represent promising vaccine candidates. It was interesting that deletion of the G gene yielded a promising, replication-competent vaccine candidate, since the corresponding deletion in HRSV yielded a virus that was overattenuated in mice and humans (13, 28). An rBRSV lacking the G gene may not have been competent for replication in vivo but still was able to protect calves, its natural host, against a subsequent infection (21), but our experience with attenuated human respiratory viruses is that replication is necessary in order for the virus to be satisfactorily immunogenic. With HRSV, deletion of the G protein from vaccine virus also is considered undesirable, because G is one of the two major neutralization and protective viral antigens. However, the contribution of HMPV G to neutralization and protective efficacy may be relatively minor (unpublished data). Furthermore, the HMPV F protein alone expressed from recombinant vectors induced a high level of neutralizing antibodies and protective immunity against homologous and heterologous HMPV strains (22, 24). Thus, the G and SH/G viruses represent vaccine candidates with promising levels of attenuation and protective efficacy that can be studied further in nonhuman primates such as African green monkeys and should be prepared for clinical evaluation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
), 
), 
), or 
). Supernatant aliquots (0.5 ml out of a total medium volume of 2 ml per well) were taken on the indicated days postinfection and replaced by an equivalent volume of fresh medium containing 5 µg of trypsin/ml. The samples were flash-frozen and analyzed later by plaque assay. Each time point was represented by two wells, and each virus titration was done in duplicate. Means are shown. The standard errors were calculated, but the bars are not shown because the errors were very small and the bars were obscured by the symbols.
