Mutations in the Cytoplasmic Domain of the Newcastle Disease Virus Fusion Protein Confer Hyperfusogenic Phenotypes Modulating Viral Replication and Pathogenicity

Abstract

The Newcastle disease virus (NDV) fusion protein (F) mediates fusion of viral and host cell membranes and is a major determinant of NDV pathogenicity. In the present study, we demonstrate the effects of functional properties of F cytoplasmic tail (CT) amino acids on virus replication and pathogenesis. Out of a series of C-terminal deletions in the CT, we were able to rescue mutant viruses lacking two or four residues (rΔ2 and rΔ4). We further rescued viral mutants with individual amino acid substitutions at each of these four terminal residues (rM553A, rK552A, rT551A, and rT550A). In addition, the NDV F CT has two conserved tyrosine residues (Y524 and Y527) and a dileucine motif (LL536-537). In other paramyxoviruses, these residues were shown to affect fusion activity and are central elements in basolateral targeting. The deletion of 2 and 4 CT amino acids and single tyrosine substitution resulted in hyperfusogenic phenotypes and increased viral replication and pathogenesis. We further found that in rY524A and rY527A viruses, disruption of the targeting signals did not reduce the expression on the apical or basolateral surface in polarized Madin-Darby canine kidney cells, whereas in double tyrosine mutant, it was reduced on both the apical and basolateral surfaces. Interestingly, in rL536A and rL537A mutants, the F protein expression was more on the apical than on the basolateral surface, and this effect was more pronounced in the rL537A mutant. We conclude that these wild-type residues in the NDV F CT have an effect on regulating F protein biological functions and thus modulating viral replication and pathogenesis.

INTRODUCTION

Newcastle disease virus (NDV) is a highly prevalent avian pathogen that infects essentially all species of birds and is of major economic importance to the poultry industry (1, 2). The disease varies in degree of severity, ranging from an inapparent infection to outbreaks of severe respiratory and neurologic disease that can have 100% mortality. Based on severity of the disease, NDV strains are grouped into lentogenic (nonpathogenic or mildly pathogenic), mesogenic (moderately pathogenic), and velogenic (highly pathogenic) pathotypes (3). NDV belongs to the genus Avulavirus within the family Paramyxoviridae, a family of enveloped, nonsegmented, negative-sense RNA viruses (4).

The envelope of NDV contains two surface glycoproteins, HN and F (5, 6). HN mediates viral attachment by binding to sialic acid cellular receptors, and F protein facilitates viral entry into the cells by fusing the viral envelope with the host cell membrane (7, 8). The virus fusion process involves a series of major coordinated conformational changes in the F protein that bring together and merges the opposing membranes (5, 9).

The NDV F protein (Fig. 1) is synthesized as an inactive precursor, F₀ (66 kDa), that is cleaved posttranslationally by host cell proteases into two disulfide-linked subunits, N-terminal F₂ (12.5 kDa) and C-terminal F₁ (55 kDa) (8, 10). The F cleavage site is a major determinant of NDV tropism and virulence: virulent strains typically have cleavage sites with multiple basic residues that are readily cleaved in most cell types and provide for systemic spread, whereas avirulent stains depend on secreted protease for cleavage and usually are restricted to the mucosal surfaces of the respiratory and enteric tracts. The F₁ subunit contains two hydrophobic domains, the fusion peptide (FP), present at the N terminus and created by protease cleavage, and the transmembrane (TM), domain which is located near the C terminus and anchors the protein in the membrane of the virus or infected cell. The FP initiates the process of fusion by inserting into the target host cell membrane (11). The F₁ subunit has two heptad repeat (HR) motifs: HR1 is immediately C terminal to the FP, and HR2 is immediately N terminal to the TM domain. Upon triggering, HR1 and HR2 domains undergo coordinated conformational changes necessary for fusion (8, 12). The NDV F protein is a class I fusion protein that has structural and functional characteristics that are highly related to those of the F proteins of other paramyxoviruses, including parainfluenza type 5 (PIV5) (previously known as SV5), measles virus, respiratory syncytial virus (RSV), and Nipah and Hendra viruses, and also has general similarity to gp41 of human immunodeficiency virus type 1 (HIV-1), the hemagglutinin (HA) of influenza virus, and GP2 of Ebola virus (11, 13–18).

Fig 1.

Schematic diagram of the NDV F protein and mutations that were introduced into the CT. A linear diagram of the NDV F protein and sequences of the intact WT CT and of progressive deletion mutations (Δ) are shown. Gray boxes: heavy shading, fusion peptide; intermediate shading, heptad repeats (HR); light shading, transmembrane (TM) domain. Alanine substitutions are in bold.

NDV F protein has structural features in addition to the cleavage site, FP, and HR that affect the efficiency of the fusion process and thus may influence viral infectivity, replication, and pathogenicity (19–24). While it is well known that structural features in the ectodomain of the F protein can have a major impact on fusion (7, 19), several reports on other type I fusion glycoproteins (those of retrovirus, lentivirus, herpesvirus, and other paramyxoviruses) have also indicated the role of cytoplasmic tail (CT) in regulating viral entry, F protein cleavage, and fusogenicity (21, 25–34). In addition, the CTs of several other viral envelope glycoproteins contain sequence motifs that target the transport of newly synthesized glycoproteins from the endoplasmic reticulum (ER) to different intracellular compartments and to the cell surface (35, 36). In recent years, tyrosine-containing signals, especially Y-X-X-aliphatic/aromatic consensus motifs, in the CTs of viral membrane proteins have been found to be associated with targeted protein delivery (37–39). A second type of signal, a dileucine (LL) motif, has similarly been shown to mediate internalization and targeting of viral glycoproteins to intracellular compartments and to the basolateral surface of polarized epithelial cells (40, 41). Mutagenesis of tyrosine and dileucine motifs in the CTs of several other viral envelope glycoproteins provided evidence that they can affect fusion and infectivity (42–44).

The NDV fusion protein CT is 31 amino acids long (amino acid positions 523 to 553) (Fig. 1) and is highly conserved among different strains of NDV (20, 45–47). It has been previously reported that deletions in the NDV F CT greatly reduced syncytium formation (48). In addition, the NDV F protein CT has tyrosine residues at positions 524 and 527 and a dileucine motif at positions 536 and 537, and their possible roles in fusion and viral pathogenicity have not been known.

In the present study, we investigated the potential role(s) of the NDV F protein CT in viral replication and pathogenicity. Using reverse genetics, we rescued 11 NDV mutant viruses with truncated CTs or with point mutations in the CT involving conserved signals or possible motifs. The mutant viruses were characterized for intracellular processing and surface expression of F and for membrane fusion and replication in vitro and in vivo. Our results showed that truncation of 2 and 4 C-terminal amino acids and substitution of CT tyrosine residues in F protein resulted in hyperfusogenic phenotypes with increased pathogenicity. We further analyzed the significance of tyrosine and dileucine motifs in polarized cells for apical and basolateral transport. Substitution of both the tyrosine residues together resulted in impaired apical expression of F glycoprotein; however, substitution of a single leucine residue moderately reduced basolateral targeting of F glycoprotein.

MATERIALS AND METHODS

Cells and viruses.

The chicken embryo fibroblast DF1 and human epidermoid carcinoma HEp-2 cell lines were grown in Dulbecco's minimal essential medium (DMEM) with 10% fetal bovine serum (FBS) and maintained in DMEM with 5% FBS. The African green monkey kidney Vero cell line was grown in Eagle's minimal essential medium (EMEM) containing 10% FBS and maintained in EMEM with 5% FBS. The modified vaccinia virus strain Ankara (MVA) expressing T7 RNA polymerase was kindly provided by Bernard Moss (NIAID, Bethesda, MD) and propagated in primary chicken embryo fibroblast cells in DMEM with 5% FBS. The moderately pathogenic (mesogenic) NDV strain Beaudette C (BC) and its mutant derivatives were grown in 9-day-old embryonated specific-pathogen-free (SPF) chicken eggs in an enhanced biosafety level 3 (BSL3) containment facility certified by the USDA following the guidelines of IACUC, University of Maryland. After 2 days, the allantoic fluid was harvested and the virus was plaque purified using our standard procedure (46).

Construction of plasmids and recovery of mutant viruses.

The construction of plasmid pNDVfl carrying the full-length antigenome cDNA of NDV strain BC has been described previously (46). The mutations that were introduced into the F protein CT are summarized in Fig. 1. Their introduction was facilitated by the presence of the unique restriction enzyme sites PacI and AgeI located in the untranslated regions (UTRs) flanking the F and HN open reading frames (ORFs) in the NDV cDNA. The PacI-AgeI fragment containing the F-HN gene was mutagenized with primers containing the desired mutations in a two-step procedure. First, two PCR products were generated: one PCR product was generated using a nonmutant forward oligonucleotide that primed upstream of the PacI site combined with a mutation-bearing reverse primer, and the second PCR product was generated using a mutation-bearing forward primer and a nonmutant reverse oligonucleotide that primed downstream of the AgeI site. Second, overlapping PCR was then used to generate the ∼4-kb PacI-AgeI fragment containing the desired mutation, which was cloned into TOPO-XL vector (Invitrogen). The inserts bearing the desired mutation were cloned into the full-length antigenomic cDNA of strain BC. The rule of six was maintained in all of the mutants. All mutant F cDNAs were sequenced in their entirety to confirm the presence of the desired mutations.

Recovery of viruses and confirmation of genetic stability and lack of adventitious mutations.

Plasmid transfection and recovery of mutant NDV mutants were performed as described previously (46). Briefly, HEp-2 cells were transfected with three plasmids individually encoding the N, P, and L proteins (3.0 μg, 2.0 μg, and 1.0 μg per single well of a six-well dish, respectively) and a fourth plasmid encoding the full-length antigenome (5.0 μg) using Lipofectamine (Invitrogen, Carlsbad, CA) and simultaneously infected with vaccinia virus MVA expressing T7 RNA polymerase at a multiplicity of infection (MOI) of 1 PFU/cell. Two days after transfection, the cell culture medium supernatant was harvested and inoculated into the allantoic cavities of 9-day-old SPF embryonated chicken eggs. Recovery of the virus was confirmed by hemagglutination assay using 1% chicken red blood cells (RBCs). The sequences of the F and HN genes in the recovered chimeric viruses were confirmed by reverse transcription-PCR (RT-PCR) and nucleotide sequencing. In cases where virus was not recovered, at least three independent transfections were performed in parallel with the wild-type (WT) cDNA as a positive control before the construct negative was considered for virus recovery.

To assay genetic stability, the recovered CT mutant viruses were passaged in 9-day-old SPF chicken embryos five times. From each passage, total RNAs were isolated from NDV-infected allantoic fluid of 9-day-old SPF chicken embryos, using TRIzol reagent (Invitrogen). RT-PCR was performed using the Thermoscript RT-PCR kit (Invitrogen) with specific forward and reverse primers to amplify the F gene. The amplified cDNA fragments were then sequenced using the BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems Inc.) in an ABI 3130xl genetic analyzer to confirm the presence of the introduced mutations in the recovered viruses. The HN gene from each recovered virus was also sequenced with available primers from our laboratory.

Western blot analysis and domain-selective surface biotinylation and immunoprecipitation.

Expression of F protein was examined by Western blotting. Briefly, Vero cells were infected at a multiplicity of infection (MOI) of 0.01 PFU. The cells were harvested at 24 h postinfection (hpi), the lysed proteins were denatured, reduced, separated by 10% SDS-PAGE, and analyzed by Western blotting using a 1:100 dilution of anti-Fcyt terminal specific antibodies (22). The blot was stripped using stripping buffer (Restore Plus Western blot stripping buffer; Thermo Scientific) and reimmunoblotted using anti-β-tubulin antibodies (Invitrogen).

Madin-Darby canine kidney (MDCK) cells were seeded on 0.4-μm-pore-size Transwell polycarbonate filters (BD Falcon), and polarization of the cell monolayer was tested every day by monitoring the electrical resistance between the upper and lower chambers (World Precision Instruments [WPI]). Cells were infected with recombinant WT (rWT) and mutant viruses. At 48 hpi, cells were washed three times with phosphate-buffered saline (PBS), and either the apical or the basolateral side of the filter membrane was incubated twice for 20 min at 4°C with PBS containing 2 mg S-NHS-biotin ml⁻¹ (Thermo Scientific). After biotinylation, cells were washed with cold PBS containing glycine (0.1 M) to the opposite membranes. After washing the cells once with 0.1 M glycine and three times with PBS, filter membranes were cut out and lysed in 0.5 ml radioimmunoprecipitation assay buffer (1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 0.15 M NaCl, 10 mM EDTA, 10 mM iodoacetamide, 1 mM phenylmethylsulfonyl fluoride [PMSF], and 20 mM Tris-HCl, pH 8.5). Cell lysates were clarified by centrifugation for 20 min at 19,000 × g. Supernatants were immunoprecipitated using anti-F_Nterm antibodies and 50 μl protein A-Sepharose beads (Calbiochem). Following separation by 10% SDS-PAGE and blotting on to nitrocellulose, biotinylated proteins were detected with streptavidin-biotinylated horseradish peroxidase complex and enhanced chemiluminescence (Amersham).

Cell surface expression of the CT mutant viruses.

Cell surface expression of the F proteins of the CT mutant viruses was quantified by flow cytometry. Briefly, DF1 cells were infected with each mutant virus at an MOI of 0.1. After overnight infection, the cells were detached with PBS containing 5 mM EDTA and centrifuged at 500 × g for 5 min at 4°C. Cells were then incubated with rabbit anti-F_Nterm antiserum (1:10 dilution) for 30 min at 4°C (22). Subsequently, cells were washed 3 times with PBS and incubated for 30 min on ice with 1:500 diluted Alexa Fluor 488-conjugated goat anti-rabbit immunoglobulin G antibodies. Cells were analyzed by using a FACSria II apparatus and FlowJo software (Becton Dickinson Biosciences).

Surface immunofluorescence analysis.

MDCK cells were grown on 0.4-μm-pore size filter supports and infected with rWT and mutant viruses. At 48 hpi, infected cells were fixed with 2% paraformaldehyde (PFA) in DMEM for 2 h and then incubated from both sides with rabbit antibodies directed against the NDV F protein overnight at 4°C. The primary antibodies were detected using Alexa Fluor 555-conjugated secondary antibodies (Invitrogen) for 1.5 h at 4°C. To visualize cell junctions, cells were permeabilized for 10 min with 1:1 methanol and acetone and stained with a monoclonal antibody against E-cadherin (BD Biosciences), Alexa Fluor 488-conjugated secondary antibodies (Invitrogen), and DAPI (4′,6′-diamidino-2-phenylindole) (BD Biosciences). Filters were cut out from their supports, mounted onto microscope slides in antifade solution (Invitrogen), and analyzed using a Zeiss Axiovert200M microscope or with a confocal laser scanning microscope (Zeiss LSM510).

Fusion assay of the CT mutant viruses.

Syncytium formation was quantified as described by Kohn (49). Briefly, Vero cells in 6-well plates were infected with each virus at an MOI of 0.1. Cells were maintained in 5% minimal essential medium (MEM) at 37°C under 5% CO₂. At 24 hpi, the medium was removed and the cells were washed with PBS, fixed with methanol for 20 min at room temperature, and stained with hematoxylin-eosin. The fusion index of each mutant virus was calculated by observing 10 fields per well in duplicate. The fusion index is the ratio of the total number of nuclei to the number of cells in which these nuclei are present (i.e., the mean number of nuclei per cell).

Multicycle growth in DF1 cells.

DF1 cells in duplicate wells of six-well plates were infected with each virus at an MOI of 0.01. After 1 h of adsorption, the cells were washed with PBS and overlaid with DMEM containing 5% FBS at 37°C. The medium was collected and replaced with an equal volume of fresh medium at 8-h intervals until 64 hpi. Virus titers were quantified by plaque assay on DF1 cells.

Virus purification.

The wild-type and hyperfusogenic mutant viruses were harvested from allantoic fluid, clarified by centrifugation at 5,000 rpm for 5 min, and then overlaid with 30% (wt/vol) sucrose and subjected to centrifugation at 100,000 × g for 3 h at 4°C, using a Beckman SW 25 rotor. Viral proteins were resolved by SDS-PAGE, and the gel was then stained with Coomassie brilliant blue. Purified viruses were further analyzed by Western blotting using a 1:100 dilution of anti-Fcyt antibodies. The blots were stripped using stripping buffer and again blotted with anti-NP polyclonal antibodies at a 1:200 dilution. Amounts of F and NP proteins were quantified by densitometry.

MDT and ICPI.

The pathogenicity of the F cytoplasmic mutant viruses was determined by the mean death time (MDT) test in 9-day-old embryonated chicken eggs and by the intracerebral pathogenicity index (ICPI) test in 1-day-old SPF chicks. All studies were conducted under enhanced biosafety level 3 (EBSL3) conditions at the University of Maryland (22).

Statistical analysis.

Statistically significant differences in data from different recombinant virus groups were evaluated by one-way analysis of variance (ANOVA). Growth kinetics for various groups were analyzed using correlation of XY pairs (Pearson) and P value (two tailed). P values below 0.05 were regarded as being significant for all analyses. Experiments were repeated a minimum of three times. Statistical analysis for mean and standard deviation of data and one-way ANOVA was done by using Prism 5.0 computer software (GraphPad Software Inc., San Diego, CA).

RESULTS

Construction and recovery of F CT mutant viruses.

In the present study, we investigated the importance of the cytoplasmic tail (CT) of the NDV F protein in viral replication and pathogenesis using reverse genetics. We created constructs with mutations in the NDV F protein CT using a cDNA encoding the full-length antigenome cDNA of the moderately virulent (mesogenic) NDV strain BC (46). Six mutants were constructed, involving progressive deletion of 2, 4, 6, 12, 18, or 30 amino acids from the C terminus (Fig. 1). However, of these six mutants, only two viable viruses were recovered, with 2- and 4-amino-acid deletions (rΔ2 and rΔ4, respectively). This implied that deletion of 6 amino acids or more from the CT was lethal for the production of infectious NDV. Since the C-terminal 4 amino acids of the CT were dispensable for virus replication, we constructed four more mutants in which these 4 residues were individually replaced by alanine (Fig. 1). We were able to recover all four of these mutants, designated rM553A, rK552A, rT551A, and rT550A.

Several viral membrane proteins have been shown to contain tyrosine (YXXϕ)- or dileucine (LL)-based targeting motifs involved in protein trafficking in both the secretory and endocytic pathways. The NDV F protein CT contains two tyrosine residues (Y) at positions 524 and 527 and one dileucine (LL) motif at positions 536 and 537. These residues are highly conserved between NDV isolates. To investigate possible roles of these Y and LL motifs in regulating apical sorting and fusion, we mutated each of the two tyrosine residues singly to alanine and doubly to alanine and mutated each of the residues in the dileucine singly to alanine and doubly to dialanine (Fig. 1). We were able to rescue viruses with substitution of tyrosine to alanine (rY524A, rY527A, and rYY524-527AA) and single substitution of leucine to alanine (rL536A and rL537A).

Intracellular processing and expression of the F proteins of the CT mutant viruses.

In order to investigate the effects of CT deletions and point mutations on F protein synthesis and processing, the rWT and mutant CT viruses were used to infect Vero cells and were analyzed by Western blotting using Fcyt-specific antibody which was raised against 30-amino-acid-long C-terminal synthetic peptides of the F protein (22). The F₀ protein of each of the mutant viruses was expressed and cleaved to F₁-F₂, although the extent of cleavage varied among these mutants (Fig. 2). In the rΔ2, rΔ4, rM553A, rK552A, rT551A, and rT550A mutants, both F₀ and F₁ were detectable; however, there was reduced processivity in rM553A, rK552A, rT551A, and rT550A, resulting in a higher percentage of F₀ than of F₁. In contrast, in the rY524A and rY527A mutants, there was a higher percentage of F₁ than F₀, suggesting enhanced processivity of the F protein. In case of rYY524-527AA, rL536A, and rL537A, the intracellular expression of F₀ and F₁ was less than that for rWT.

Fig 2.

Western blot analysis of F CT mutants expressed in Vero cells. Vero cells were infected with virus at an MOI of 0.01 PFU. After 24 hpi, the cells were collected and processed to prepare cell lysates. The samples were denatured, reduced, and subjected to Western blot analysis using anti-F_cyt specific antibodies. The positions of F₀ and F₁ are indicated by arrows in the left margins. The immunoblot was stripped using stripping buffer and again immunostained using monoclonal antibody against β-tubulin as a cellular control.

The relative levels of expression of the F CT mutant viruses were measured by flow cytometric analysis of DF1-infected cells using anti-F_Nterm rabbit antiserum. The results showed that the percentages of cells expressing F protein on the cell surface were indistinguishable for the rWT and CT mutant viruses. However, the level of F protein expression per cell, measured by mean fluorescence intensity, varied considerably (Table 1). After 12 hpi, the deleted CT mutant F protein expression exceeded that of rWT by 15% to 52%, depending on the mutant. The highest levels of expression were observed with the two tyrosine mutants, rY527A and rY524A (52% and 45% increases, respectively). In contrast, in the case of the tyrosine double mutant rYY524-527AA, there was a reduction of fluorescence intensity by 42%, followed by moderate reduction of intensity in leucine single mutants rL536A and rL537A (11% and 18%, respectively). These data suggest that all the mutants are capable of spreading under these conditions.

Table 1.

Cell surface expression of the F proteins of CT mutant viruses^a

Virus	% of positive cells (mean ± SD)b	Relative mean fluorescence intensityc
rWT	99 ± 1.8	1.00
rΔ2	99 ± 1.2	1.30
rΔ4	99 ± 1.5	1.25
rM553A	98 ± 2.0	1.19
rK552A	95 ± 1.2	1.15
rT551A	99 ± 2.0	1.20
rT550A	99 ± 2.0	1.08
rY524A	99 ± 1.8	1.45*
rY527A	99 ± 1.2	1.52*
rYY524-527AA	96 ± 1.5	0.58
rL536A	95 ± 1.2	0.89
rL537A	92 ± 2.4	0.82
None (mock-infected cells)		0.01

^aCell surface expression of the F protein was determined by flow cytometry. DF1 cells were infected with each mutant virus at an MOI of 0.1. Surface expression of the F proteins was assessed by flow cytometry at 12 hpi with rabbit anti-F_Nterm antiserum followed by anti-rabbit Alexa Fluor 488-conjugated antibodies. Surface immunofluorescence was quantitated by fluorescence-activated cell sorter (FACS) analysis. Uninfected DF1 cells were used as negative controls.

^bValues shown are results from three independent experiments.

^c*, P < 0.05.

Our results indicate that CT mutants retained the ability to be synthesized and transported to the cell surface and indeed showed that most of the mutants did so more efficiently than rWT. The single tyrosine mutant proteins cleaved more efficiently than wild-type F protein, which might be enhancing the transport of the F protein through the Golgi apparatus and therefore increasing cell surface expression, whereas in double tyrosine and single leucine mutants, the cleavage of F protein was reduced, which resulted in decreased cell surface expression. However, in this study cytoplasmic tail antiserum was used for immunoblotting, and it is possible that the apparent variation in the quantity of F protein produced in some mutants might be due to loss of a major epitope.

Fusion activity and cytopathic effect (CPE) of the CT mutant viruses in Vero cells.

To investigate possible effects of the CT mutations on the fusion activity of F protein, mutant viruses were examined in Vero cells (Fig. 3). The CT deletion/substitution and single tyrosine mutant viruses exhibited increased (13% to 48% higher) fusion indices than the rWT virus, with the exception of the rT550A mutant, which was essentially identical to rWT, and the rT551A mutant, whose fusion index was only marginally increased. The most efficient fusion was observed with the rY527A and rY524A mutants (48% and 36% increases, respectively, compared to rWT). Substantial increases also were observed with the rΔ2 and rΔ4 mutants (27% and 13%) and the rM553A and rK552A mutants (32% and 27%, respectively). In contrast, in tyrosine double mutant rYY524-527AA there was a reduction in fusion of 32%, and in rL536A and rL537A there were reductions in fusion activity of 8% and 12%, respectively. These data indicate that the CT of F protein is capable of modulating cell-cell fusion mediated by rWT F glycoprotein.

Fig 3.

Comparison of the fusogenicity and CPE of rWT and the CT mutant viruses in Vero cells. Relative levels of fusion obtained for the CT mutants compared to rWT are shown. Vero cells were infected with the indicated viruses at an MOI of 0.1, fixed at 24 hpi, and stained with hematoxylin-eosin. The fusion index was calculated as the ratio of the total number of nuclei in multinuclear cells to the total number of nuclei in the field. Ten fields were counted per condition. Fusion levels were normalized to WT at 100%. Data shown are averages of three independent experiments. Error bars refer to standard errors of the means (SEM) (P < 0.0001, reflects comparison of all groups through ANOVA).

Replication kinetics of the CT mutant viruses.

The multistep growth kinetics and magnitudes of replication of the CT mutant viruses were determined in DF1 cells (Fig. 4). The CT deleted/substituted mutant viruses and single tyrosine mutants replicated exponentially until ∼40 hpi, after which replication was at a plateau. The magnitudes of replication were similar for rWT and the rT550A and rT551A viruses (Fig. 4A) but were substantially higher for the other CT mutant viruses. The highest viral titers were seen with the tyrosine mutant viruses rY524A and rY527A, followed closely by the rΔ2, rM553A, rK552, and rΔ4 viruses. For example, the titer of the rY527 virus was 1.0 log₁₀ higher than that of rWT at 16 hpi and 2.0 log₁₀ higher than that of rWT at 40 hpi (Fig. 4B). Interestingly, the doubly substituted tyrosine had a reduced replication rate and had the maximum titer reduced by 1 log₁₀, whereas the rL536A and rL537A mutants replicated at a lower titer of 0.5 log₁₀ until 48 hpi (Fig. 4B), after which the replication rates were similar to that of wild-type virus.

Fig 4.

Comparison of the multicycle growth kinetics and CPE of rWT and the CT mutant viruses in DF1 cells. (A) Comparison of multicycle growth kinetics of rWT, rΔ2, rΔ4, rM553A, rK552A, rT551A, and rT550A. Cells were infected with each virus at an MOI of 0.01, and cell culture medium supernatant aliquots were harvested at 8-h intervals until 64 hpi. The virus titers in the aliquots were determined by plaque assay in DF1 cells (P = 0.0001). (B) Comparison of multicycle growth kinetics of rWT, rY524A, rY527A, rYY524-527AA, rL536A, and rL537A. Data shown are averages of three independent experiments with standard errors (P = 0.002).

Surface distribution of F glycoproteins in infected polarized MDCK cells.

To study the roles of tyrosine and dileucine motifs in the transport of NDV F glycoprotein in polarized epithelial cells, MDCK cells were grow on polycarbonate filters, infected with rWT or the tyrosine and leucine mutants, and subjected to domain-specific surface biotinylation assay as described in Materials and Methods. In rWT-infected cells, F glycoprotein is abundantly expressed on both the apical and basolateral surfaces (Fig. 5 panel A). Similar results were seen in single tyrosine mutants, where F glycoproteins were found on both the apical and basolateral surfaces, suggesting that substitution of single tyrosine residue did not have any significant effect on transporting the F glycoprotein to either of the cell surfaces. In contrast, substitution of both the tyrosine residues reduced apical surface expression of F glycoprotein, and there was also a reduction in F protein expression on the basolateral surface compared rWT and single tyrosine residue mutant F protein expression (Fig. 5). This indicates that in the context of virus infection, abundant expression of NDV F protein on apical and basolateral surfaces is prevented by mutation of both the tyrosine residues. However, in leucine mutant-infected MDCK cells, a higher percentage of F glycoproteins was expressed apically (Fig. 5B). These data indicate that the dileucine motif in F glycoprotein might be influencing the basolateral targeting of F protein.

Fig 5.

Surface-selective biotinylation of wild-type or mutant F CT glycoproteins in polarized MDCK cells. MDCK cells were grown on permeable filter supports and infected with viruses at an MOI of 0.01 PFU. After 48 hpi, cells were surface biotinylated from either the apical (Ap) or the basolateral (Bl) side. After cell lysis, proteins were immunoprecipitated with F-specific antibodies and were analyzed by SDS-PAGE using peroxidase-conjugated streptavidin. (A) Cell lysates of rY524A, rY527A, and rYY524-527AA compared with rWT. (B) Cell lysates of rL536A and rL537A compared with rWT.

NDV F protein in single tyrosine mutants is expressed predominantly on apical and basolateral surfaces of polarized epithelial cells.

To further evaluate the spread of single tyrosine mutants and the distribution of these mutant F proteins on apical and basolateral surfaces, polarized MDCK cells were grown on filters and were infected with wild-type and mutant viruses from the apical side. At 48 hpi the samples were fixed with 2% PFA and immunostained with anti-F_Nterm antiserum and Alexa Fluor 555-conjugated secondary antibodies. Cell junctions were visualized by costaining with E-cadherin-specific monoclonal antibodies. As shown by confocal analysis of the distribution of F protein (Fig. 6A), fluorescent signals were detected more at the apical surface in single-tyrosine-mutant-infected MDCK cells than in cells with wild-type F protein. In infected MDCK cells, there was disruption of the cell junctions, which might be a result of viral replication and fusion of adjacent cells. Confocal horizontal sections through the apical to basal surfaces in MDCK cells infected with the rY527A mutant showed extensive expression of F protein on the apical surface compared to that in cells infected with the wild type (Fig. 6B). In addition, the fluorescent signals for F protein in rY527A mutant-infected MDCK cells were not reduced in the central and basal parts compared to those in wild-type-infected cells; rather, there was more expression on the basolateral surface (vertical sections shown in Fig. 6B).

Fig 6.

Surface distribution of wild-type and single tyrosine mutant F proteins in polarized MDCK cells. MDCK cells were grown on filter supports until full polarization was reached and then infected with the wild type or single tyrosine mutants. (A) At 48 hpi, cells were fixed with 2% PFA and subsequently stained with anti-F specific antiserum and Alexa Fluor 555-conjugated secondary antibodies. After permeabilization with methanol-acetone, cell junctions were visualized with anti-E cadherin monoclonal antibodies and with Alexa Fluor 488-conjugated secondary antibodies and costained with DAPI. (B) Confocal horizontal sections through the apical part of the cell monolayer to the bottom, showing the surface distribution of wild-type F protein (i) and rY527A F protein (ii) on the apical and basolateral sides.

Incorporation of F protein in hyperfusogenic viruses.

To examine the amount of incorporation of the F proteins into hyperfusogenic viruses, the wild-type and hyperfusogenic mutant viruses were partially purified from infected allantoic fluid by sucrose gradient centrifugation, analyzed by Coomassie blue staining, and then quantified by Western blotting. We examined the level of F protein incorporation in different mutant viruses by determining the ratio of the F protein to another viral protein, NP (Fig. 7). Compared to the wild-type virus F protein band, there were 8% and 6% increases in the relative intensities of F protein bands in the rΔ2 and rΔ4 mutant viruses. The highest intensity of the F protein band was seen in the rY527A mutant (25% increase compared to that for rWT), followed by 20% in rY524A virions.

Fig 7.

Incorporation of the F protein into viral particles. (A) Ultracentrifuge-purified viruses from infected allantoic fluid samples were separated by electrophoresis, and the gel was then stained with Coomassie brilliant blue. Lanes: 1, rWT; 2, rΔ2; 3, rΔ4; 4, rY524A; 5, rY527A. (B) Purified virus samples were analyzed by Western blotting using anti-Fcyt antibodies (top) and anti-NP polyclonal antibodies (bottom). (C) Relative intensities of NP and F protein in wild-type and mutant viruses as measured by densitometry. *, P < 0.05 for rY524A and rY527A.

Pathogenicity of the CT mutant viruses in embryonated chicken eggs and 1-day-old chicks.

We evaluated the effect of the CT mutations on viral pathogenicity using two standard pathogenicity assays, namely, the mean embryo death time (MDT) assay and the intracerebral pathogenicity index (ICPI) test. MDT values were determined in 9-day-old embryonated chicken eggs (Table 2). NDV strains are categorized into three pathotypes on the basis of their MDT values: velogenic (less than 60 h), mesogenic (60 to 90 h), and lentogenic (greater than 90 h). The MDT value of the rT550A mutant (59 h) was essentially identical to that of rWT virus. The MDTs of the other mutants were reduced to various extents compared to that for rWT, which is suggestive of modest increases in virulence. The greatest differences were observed with the MDTs of the rY527A (51.2 h), rY524A (52 h), and rΔ2 (54 h) mutants, which had values that were up to 15% less than that of rWT virus. The tyrosine double mutant rYY524-527AA showed a moderate decrease in MDT (68 h) compared to that of rWT. The other viruses had intermediate values.

Table 2.

Pathogenicity of the CT mutant viruses in embryonated chicken eggs and 1-day-old chicks

Virus	MDT (h)a	ICPI scoreb
rWT	60	1.51
rΔ2	54	1.68*
rΔ4	56	1.65
rM553A	55	1.60*
rK552A	54.8	1.53
rT551A	56	1.53
rT550A	59	1.53
rY524A	52	1.70*
rY527A	51.2	1.78*
rYY524-527AA	68	1.30
rL536A	62	1.56*
rL537A	64	1.50

^aThe mean embryo death time (MDT) is the mean time for the minimum lethal dose of virus to kill all of the inoculated embryos. Pathotype definition: virulent strains, <60 h; intermediate virulent strains, 60 to 90 h; avirulent strains, >90 h.

^bIntracerebral pathogenicity index (ICPI) = [(total number of sick chicks at each observation × 1) + (total number of dead chicks at each observation × 2)]/80 observations. ICPI values for velogenic strains approach the maximum score of 2.00, whereas lentogenic strains give values close to 0. Values are means of three independent experiments. *, P < 0.05.

The pathogenicity of the CT mutant viruses also was evaluated by the ICPI test in 1-day-old chicks (Table 2). Velogenic strains give values approaching 2.0, whereas lentogenic strains give values close to 0. The differences in ICPI values compared to that of rWT virus (1.51) were greatest with the rY527A (1.78), rY524A (1.70), and rΔ2 (1.68) viruses, whose values were up to 18% greater than that of rWT virus. The ICPI values of the CT mutants were increased compared to that of rWT virus (except for the rYY524-527AA [1.30] and leucine mutants), which is indicative of increased pathogenicity, although the increases were modest. The ICPI values of the other viruses were intermediate. The results of the ICPI test were consistent with the results of the MDT test, and in particular, the two tyrosine mutants rY527A and rY524A and the deletion mutant rΔ2 were the most virulent.

DISCUSSION

The CTs of the several paramyxovirus F proteins have been shown to play an important role in modulating membrane fusion and hence are a significant determinant in the replication of these viruses (26, 50, 51). Various studies on the CTs of paramyxovirus F proteins and other virus envelope proteins have revealed effects on syncytium, fusion pore formation, oligomerization, protein folding, fusion promotion, and infectivity (25, 48, 50–55). In addition, the CTs of several viral enveloped proteins have been shown to harbor critical residues required for intracellular trafficking, virus assembly, and budding (37, 39, 56–58). It was shown earlier that truncations in the CT of the NDV F protein were inhibitory to membrane fusion (48). In the present study, we have evaluated the effects of truncations and point mutations in the CT of F protein on NDV replication and pathogenesis. We attempted to recover viruses with C-terminal deletions of up to 30 amino acids from the 31-amino-acid CT of the NDV F protein, maintaining rule of six. However, we were able to recover only virus with the two smallest deletions, of 2 and 4 amino acids. Thus, while infectious NDV can readily tolerate deletions of up to 4 amino acids in the F protein CT tail, longer deletions apparently were lethal. Deletion of the first 2 and 4 amino acids from the CT resulted in mutant virus that had a hyperfusogenic phenotype, with moderate levels of increase in F protein cleavability and expression compared to those for the wild type. These deletion mutants also exhibited increased viral growth in vitro and in vivo and increased virulence in 1-day-old chicks. This was especially evident with the rΔ2 virus. In the cases of HIV-1, simian immunodeficiency virus (SIV), and herpes simplex virus 1 (HSV-1), truncations of the C-terminal CT of envelope proteins also resulted in increased cell fusion (28, 32, 33, 54). For RSV, deletion or substitution of the F protein CT severely impaired virus growth in cell culture, and it has been suggested that the deleted CT might be affecting the F protein conformation change, leading to premature fusion (29). However, in measles virus, deletion of CT resulted in enhanced cell-to-cell fusion without a significant effect on virus replication (51), whereas in SV5, the F protein CT is absolute necessary for virus propagation (59). In this study, deletion of 2 and 4 amino acids from CT resulted in enhanced cell-to-cell fusion, which might have affected the increased replication rates of viruses in cell culture.

We further investigated the role of these C-terminal four amino acids in the fusion phenotype and found moderate increases in fusogenicity, viral replication, and pathogenicity when they were individually replaced with alanine. The mechanism by which truncations or point mutations of the C-terminal four amino acids of CT mediated these effects is unclear. In purified virions, we found moderately increased in incorporation of mutant F proteins into virus particles (Fig. 7). However, it is possible that other mechanisms might be present, such as interaction of F with viral proteins M and HN, which modulate the fusion process. For example, in PIV5, the F CT was implicated in regulating fusion pore formation (59). We also note that NDV F protein processing is increased in these mutant viruses (Fig. 2). In future studies it would be interesting to determine how the deletion of CT can affect the processing through the Golgi apparatus.

The CT of NDV F harbors two tyrosine residues and one dileucine motif that are highly conserved among strains and have the potential to be signals involved in processing and transport. Our experiments revealed that in wild-type-infected MDCK cells, F proteins were abundantly present on both the apical and basolateral surface. Substitution of single tyrosine residues did not reduce the expression of F protein on either of the surfaces; rather, we found that they were overexpressed on the basolateral surface compared to wild-type virus F protein. In contrast, substitution of both the tyrosine residues resulted in preferential transport of F protein to the basolateral surface. Similar results were found with measles virus, where viruses were released from the apical membrane but the F protein was transported mostly to the basolateral surface (60). These results suggest the functional redundancy of tyrosine residues in F glycoprotein signaling to the apical surface; however, this did not lead to changes in expression of basolateral sorting. We further noted that in single leucine substitution, the F protein is accumulated more on the apical surface and this is more pronounced for the L537A substitution, suggesting that the dileucine motif might play the predominant role in basolateral sorting of NDV F protein. However, it is yet to be determined how this affects the budding of NDV from the apical and/or basolateral surface.

The significance of polarization-based signals for the course of disease was supported by previous work on Sendai virus (61). Our inability to recover the LL motif mutants even after several attempts suggests the importance of both the leucine residues in the LL motif for F protein function and viability of the virus. The two recovered tyrosine mutants rY524A and rY527A exhibited hyperfusogenic phenotypes with increased replication in vitro and increased pathogenesis in vivo. Indeed, of the mutants characterized in the present study, these phenotypes were most pronounced for the tyrosine mutants. In contrast, substitution of double tyrosine residues and single leucine residues moderately decreased the fusion process and the virus replication in cell culture.

The most widely used tyrosine-based motif is YXXϕ (where Y is tyrosine, X is any amino acid and ϕ) is an amino acid with a bulky hydrophobic group). In the NDV F CT, Y524 and Y527 are both present in the motif YLMY, and Y527 also is present in the motif YKQK. Thus, neither of these conforms to the YXXϕ motif. Therefore, other structural and functional domains of F protein such as the transmembrane domain and N-linked glycosylation might influence apical/basolateral sorting. In addition, the transport of NDV F protein in polarized epithelial cells may be affected by the presence of HN and other viral proteins.

It also is reasonable to suggest that increased fusion played an important role in the observed increased replication and pathogenicity. These results show that the NDV F protein has features that restrain the fusogenic phenotype. Since mutations to the terminal residues and to the tyrosine resides were well tolerated and presumably could readily occur and be selected for in nature, there apparently is not a selective advantage for these mutations in nature. This study has increased our understanding of NDV virulence mediated by the F protein but also has raised new questions about the mechanism by which the CT restrains fusion. These hyperfusogenic viruses may be useful in developing NDV as a better vaccine vector and as an oncolytic agent.