In paramyxoviruses, proteolytic processing of the F protein is a prerequisite for the generation of mature infectious virus. There are nine serotypes of APMV within the genus Avulavirus
in the family Paramyxoviridae
. Virulent strains of APMV-1 (NDV) have multibasic cleavage sites that are recognized and cleaved by furin, a ubiquitous intracellular protease whose preferred cleavage sequence is RX(R/K)R↓. These viruses are able to replicate in most cell types and cause systemic infection. The avirulent strains of APMV-1 have one or occasionally two unpaired basic amino acids that lack the furin motif and are cleaved by trypsin-like extracellular proteases. Hence, these viruses are restricted mostly to the respiratory and gastrointestinal tracts, and they require supplementation with exogenous proteases for in vitro
growth. It has been shown that the F protein cleavage site is a major determinant of APMV-1 virulence in chickens (27
). In contrast, the contribution of the F protein cleavage site to the pathogenicity of APMV-2 to -9 is unknown. Each of these APMV serotypes has been isolated from a different avian species. The natural host(s) of these viruses is not clearly defined. APMV-2 is endemic among passerines and causes severe respiratory illness in parrots but only mild respiratory illness in chickens. APMV-5 causes high mortality in budgerigars but is apathogenic in chickens. APMV serotypes 3, 4, 6, 7, 8, and 9 also cause mild or inapparent disease in chickens. It is not known whether the avirulence of APMV-2 to -9 in chickens is due to the F protein cleavage site sequence of these serotypes.
The F protein cleavage site sequences vary widely among the APMV serotypes. The F protein cleavage sites of APMV-4 (DIQPR
↓F) and APMV-7 (LPSSR
↓F) contain a single basic amino acid residue, and that of APMV-2 (K
↓F) has two unpaired basic amino acids. However, in each case, only the R residue in the −1 position matches the furin cleavage sequence. These three serotypes contain phenylalanine at the F1 terminal end and do not require exogenous protease supplementation for growth in cell culture (25
). APMV-5 (KRKKR
↓F) contains five basic residues in the F cleavage site and does not require exogenous protease supplementation for growth in cell culture (34
). The requirement of exogenous protease supplementation for APMV-6 (APEPR
↓L) varies with strains. The other APMV serotypes, APMV-3 (R
↓L), APMV-8 (YPQTR
↓L), and APMV-9 (IR
↓I), require exogenous protease supplementation for in vitro
growth. It is noteworthy that the cleavage sites of APMV-3, -6, and -8 have a leucine instead of a phenylalanine as the first residue of the F1 subunit, which is also found in avirulent strains of APMV-1. The presence of leucine at this position has been associated with reduced cleavability of the APMV-1 F protein (22
The goal of this study was to evaluate the role of amino acid sequence at the F protein cleavage site of APMV-2 in replication, formation of syncytia and plaques, and pathogenicity. In order to mutate the F protein cleavage site, a reverse genetics system for APMV-2 was developed for the first time. This system will be useful to study the function of the APMV-2 structural features and macromolecules in replication and pathogenesis. We generated 12 recombinant APMV-2 F protein cleavage site mutants: 11 mutants containing the F protein cleavage site of naturally occurring APMV serotypes, and 1 mutant where the phenylalanine at the N-terminal end of the F1 subunit was replaced by a leucine residue.
We were able to recover all of the above-mentioned F protein cleavage site mutants without exogenous protease supplementation, suggesting that the mutations did not adversely affect the functions of F protein. Irrespective of the F protein cleavage site or the type of cell line used, none of the APMV-2 F protein cleavage site mutants required exogenous protease supplementation for in vitro growth, resembling the wild-type APMV-2 virus. This was particularly surprising in the case of mutants containing cleavage sites from the APMV-1 avirulent strain, APMV-3, APMV-8, and APMV-9, because these cleavage sites require exogenous protease when present in their respective natural F proteins. This suggests that APMV-2 contains one or more structural features outside the cleavage site that promotes cleavage by intracellular protease. This might also account for the ability of wild-type APMV-2 to replicate even though its F protein cleavage site contains only two basic residues, one of which is consistent with the preferred furin cleavage site.
A second phenotype was the ability to form syncytia. Wild-type APMV-1, -3, and -5 produce syncytia in infected cells, whereas APMV-2, -4, and -6 to -9 cause single-cell infection. Similarly to wild-type APMV-2, the rAPMV-2 and rAPMV-2 (F-L) mutants caused single-cell infections and did not form syncytia or plaques. Surprisingly, however, all of the other F protein cleavage site mutants produced syncytia and plaques in DF1, Vero, and MDCK cells, even if the cleavage site involved (e.g., APMV-4, -6, and -9) does not confer syncytium or plaque formation in its native virus. In particular, the cleavage sites of APMV-4, -6, -7, and -8, which contain only a single basic amino acid, and the cleavage sites of avirulent APMV-1 and -9, which contain two basic unpaired amino acids, also caused syncytium formation in the absence of exogenous protease when present within the APMV-2 F protein. One possibility for these results could be an alteration of conformation of APMV-2 F protein around the cleavage site due to change in amino acid, leading to more efficient recognition by intracellular proteases and hence cleavage and exposure of the amino-terminal F1 subunit for cell-to-cell fusion. Alternatively, the F protein does not need much help from HN to mediate fusion, and it may not require as much cleaved F to get the virus to fuse with cells and subsequently cause infection. This could also explain the difference in syncytium formation due to the good cleavage mutants having sufficient F protein present to push past a threshold needed to induce cell-cell fusion. It is known that cell-cell fusion differs from virus-cell fusion in the amount of fusion activity needed to get the two pairs of membranes to merge. To verify that the amino acid modifications at the F protein cleavage site indeed led to cleavage of F0 protein, pulse-chase experiments with the parental and a few selected syncytium-forming mutant viruses were performed. The results showed that the F protein of parental rAPMV-2 is cleaved relatively slowly but that the efficiency of cleavage is greatly enhanced in the mutants that were examined. Further, our growth kinetics results showed that syncytium formation increased the replication of the mutant viruses 10-fold, suggesting that the enhanced growth was probably due to increased cell-to-cell spread of the virus.
Thus, incorporation of the cleavage sites of the other APMV serotypes into APMV-2 resulted in growth that was independent of added protease, was increased 10-fold, and conferred the ability to form syncytia and plaques. The effects were similar for the various cleavage sites irrespective of the number of basic residues or the presence of phenylalanine versus leucine at the N terminus of the F1 subunit. These results led to the expectation that virulence in vivo
would be enhanced, based on the well-known example of APMV-1, for which the sequence at the F protein cleavage site is a major determinant of virulence (27
). Surprisingly, however, all of the viruses retained the avirulent phenotype of the APMV-2 parent. In all cases, the MDT values of APMV-2 F protein cleavage site mutants were more than 168 h and the ICPI values were zero, similar to that of the wild-type APMV-2. This suggests that the F protein cleavage site does not play an important role in the virulence of APMV-2.
In addition, the effect of number of basic amino acids at the F protein cleavage site on pathogenesis was studied in 2-week-old chickens. We chose five mutants that varied from one another in number of basic amino acid residues at the F protein cleavage site, i.e., rAPMV-2 (type 4), rAPMV-2, rAPMV-2 (type 3), rAPMV-2 (type 1v), and rAPMV-2 (type 5), with 1, 2, 3, 4, and 5 basic amino acids, respectively, at the F protein cleavage site. Our results did not show any significant difference in viral replication and tissue tropism among the F protein cleavage site mutants, suggesting no direct correlation between the number of basic amino acids at the F protein cleavage site and the pathogenicity of APMV-2. Again, these results are consistent with the interpretation that the F protein cleavage site is not a major determinant of virulence in APMV-2.
We have shown previously that altering the F protein cleavage site of an avirulent strain to that of a neurovirulent strain of NDV did not convert the avirulent strain into a neurovirulent strain after a natural route of infection (27
). Although the biological activities of the fusion protein and growth characteristics of the virus in vivo
improved over those for the avirulent parental strain, the complete spectrum of virulence phenotype could not be achieved by modifying the F cleavage site. Further, it has been shown that other proteins of NDV, such as the HN, V, NP, P, and L proteins, are responsible for virulence along with the F protein (12
). Therefore, it is possible that all of these proteins along with the F protein might contribute to the virulence of APMV-2, which requires further investigation.
In conclusion, we found that replacement of the F protein cleavage site of APMV-2 with that of any of the other APMV serotypes was associated with replication independent of added protease, the ability to form syncytia and plaques, and increased viral replication. It may be that these substitutions caused a conformational change leading to increased efficiency of F protein cleavage and function. However, none of these changes increased the virulence or changed the tropism of the virus. Thus, the cleavage and function of the F protein do not appear to be important factors in the virulence of APMV-2.