JVI Figure table search 04
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kurath, G.
Right arrow Articles by Winton, J. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kurath, G.
Right arrow Articles by Winton, J. R.

 Previous Article  |  Next Article 

Journal of Virology, February 2004, p. 2045-2056, Vol. 78, No. 4
0022-538X/04/$08.00+0     DOI: 10.1128/JVI.78.4.2045-2056.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Complete Genome Sequence of Fer-de-Lance Virus Reveals a Novel Gene in Reptilian Paramyxoviruses

Gael Kurath,1* William N. Batts,1 Winfried Ahne,2 and James R. Winton1

U.S. Geological Survey, Western Fisheries Research Center, Seattle, Washington 98115,1 Institute of Zoology, Fishery Biology, and Fish Disease, University of Munich, D-80539 Munich, Germany2

Received 24 March 2003/ Accepted 4 November 2003


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The complete RNA genome sequence of the archetype reptilian paramyxovirus, Fer-de-Lance virus (FDLV), has been determined. The genome is 15,378 nucleotides in length and consists of seven nonoverlapping genes in the order 3' N-U-P-M-F-HN-L 5', coding for the nucleocapsid, unknown, phospho-, matrix, fusion, hemagglutinin-neuraminidase, and large polymerase proteins, respectively. The gene junctions contain highly conserved transcription start and stop signal sequences and tri-nucleotide intergenic regions similar to those of other Paramyxoviridae. The FDLV P gene expression strategy is like that of rubulaviruses, which express the accessory V protein from the primary transcript and edit a portion of the mRNA to encode P and I proteins. There is also an overlapping open reading frame potentially encoding a small basic protein in the P gene. The gene designated U (unknown), encodes a deduced protein of 19.4 kDa that has no counterpart in other paramyxoviruses and has no similarity with sequences in the National Center for Biotechnology Information database. Active transcription of the U gene in infected cells was demonstrated by Northern blot analysis, and bicistronic N-U mRNA was also evident. The genomes of two other snake paramyxovirus genotypes were also found to have U genes, with 11 to 16% nucleotide divergence from the FDLV U gene. Pairwise comparisons of amino acid identities and phylogenetic analyses of all deduced FDLV protein sequences with homologous sequences from other Paramyxoviridae indicate that FDLV represents a new genus within the subfamily Paramyxovirinae. We suggest the name Ferlavirus for the new genus, with FDLV as the type species.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The first recorded epidemic in reptiles caused by a paramyxovirus occurred in a Swiss serpentarium in 1972 (16), resulting in the isolation of Fer-de-Lance virus (FDLV) from a Fer-de-Lance viper (Bothrops atrox) (16). Since that time, over 40 paramyxovirus-like agents have been isolated from various reptile species and described in the literature (1, 2, 14, 17). These viruses have hemagglutination and neuraminidase activity, and in reptile cell lines they replicate at 28 to 30°C, causing cytopathic effects including syncytia, cytoplasmic inclusion bodies, and lysis. Serological studies indicate that many of these agents are antigenically related (3, 10, 26, 48, 50). Clinically, these viruses are serious pathogens causing epidemics with high mortalities in captive reptile populations in snake farms and zoological collections (14, 27). Infected animals show respiratory signs of severe proliferative pneumonia, sometimes accompanied by nervous disorders and pancreatic necrosis. Although the prevalence of paramyxoviruses in wild reptiles is uncertain, the viruses have been isolated from wild hosts (30) and described as potential agents of emerging infectious disease that may threaten wildlife health (11, 28). Despite this history, there has been no genetic characterization of these viruses until recent publications describing partial polymerase (L), hemagglutinin-neuraminidase (HN), and fusion protein (F) gene sequences (2, 17, 30, 33). The FDLV genome sequence described here is the first complete genome sequence of a paramyxovirus isolated from a reptile host.

Due to its historical precedence as the first paramyxovirus isolated from a reptile, FDLV is considered the archetype of the reptilian paramyxoviruses (14). The virus is listed as an unassigned member of the family Paramyxoviridae in the seventh report of the International Committee on Taxonomy of Viruses (ICTV) (37). The Paramyxoviridae are pleomorphic, enveloped viruses containing a single-stranded negative-sense RNA genome of approximately 15,500 nucleotides (nt). In the most recent ICTV report (37), the family Paramyxoviridae is comprised of two subfamilies: the subfamily Paramyxovirinae contains the genera Respirovirus, Morbillivirus, and Rubulavirus (type species Sendai virus [SeV], measles virus [MeV], and mumps virus [MuV], respectively, and the subfamily Pneumovirinae contains the genera Pneumovirus and Metapneumovirus (type species human respiratory syncytial virus [HRSV] and turkey rhinotracheitis virus [TRTV], respectively. These five established paramyxovirus genera include many highly contagious viruses that cause well-known human and animal diseases. In addition, over the last 25 years, many new paramyxoviruses have emerged and been recognized as agents of severe animal diseases, sometimes with zoonotic links to human disease. These include marine morbilliviruses (42), Hendra virus (HeV) (18, 41), Menangle virus (MenV) (7, 45), and Nipah virus (NiV) (8). Additional new paramyxoviruses of uncertain pathogenicity include Tupaia virus (TPMV) (55), Salem virus (SalV) (49), and Tioman virus (TiV) (9). One common feature of all these newly recognized viruses is replication in mammalian hosts. These new paramyxoviruses have been well characterized genetically (4, 9, 19, 20, 49, 55, 58, 59), and in many cases complete genome sequences are available. They are all members of the subfamily Paramyxovirinae, with six invariant genes in the order 3' N-P-M-F-HN-L 5', indicating the nucleocapsid, phospho-, matrix, fusion, hemagglutinin-neuraminidase, and large polymerase protein genes, respectively. Beyond this basic similarity, however, there is diversity in overall genome length, presence or absence of overlapping open reading frames (ORFs) within these invariant genes, genome terminal sequences, length and sequence of intergenic and untranslated regions, transcription regulatory signals, hexamer phasing positions of transcription units, and editing strategies for P gene expression. This variation, along with phylogenetic analyses, has resulted in the recent establishment of the new genera Henipavirus, including HeV and NiV (8, 19, 57, 59), and Avulavirus, including avian rubulaviruses with Newcastle disease virus (NDV) as the type species (12, 51). Thus, there are now five recognized genera in the subfamily Paramyxovirinae.

Amid these efforts to broaden and improve the taxonomy within the mammalian and avian Paramyxoviridae, we now present the first complete genome sequence of a paramyxovirus from a reptile host. This sequence has revealed a novel transcription unit potentially encoding a 19.4-kDa protein between the N and P/V genes and a mosaic of other features that suggest that reptile paramyxoviruses will likely comprise another new genus within the subfamily Paramyxovirinae.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Genome sequencing. The FDLV strain was obtained from the American Type Culture Collection (ATCC VR-895) and propagated in the IgH2 iguana heart cell line (ATCC CCL-108; Rockville, Md.) at 28°C (2). Construction of a random-primed cDNA library to purified viral genomic RNA has been described previously (2). Clones with significant alignment to paramyxovirus sequences were identified by searches of the National Center for Biotechnology Information (NCBI) database. Primers for reverse transcription (RT)-PCR amplification across gaps not covered by the cDNA clones were designed by using terminal sequences of clones containing N, P, F, HN, and L gene sequences. PCR amplification and sequencing of products were as described (2).

Sequences of the 3' and 5' genomic termini were determined by rapid amplification of cDNA ends (5' RACE; Gibco BRL) with primers designed from sequences near the FDLV genomic termini. For 3'-end analysis, total RNA from FDLV-infected IgH2 cells was used as a template for amplification with three FDLV N gene-specific primers (GSP) that bind to the N gene mRNA or antigenome of FDLV: GSP1, 5'-CCCATTGCAAAACTCCA-3'; GSP2, 5'-AACTCCATATCATCGGGTATG-3'; and GSP3, 5'-AGTATGACCATGTACGGTGC-3'. FDLV genomic RNA extracted from infected cell culture supernatant was used to amplify the 5' end of the genome by using three FDLV L gene primers: GSP1, 5'-TAGTTAGCAAGATAGCAC-3'; GSP2, 5'-CGAGACTAGGGACTAACTGG-3'; and GSP3, 5'-TTAAAGAACAGTAACCCTGC-3'. PCR products were purified from a low-melt agarose gel (Wizard kit; Promega) and sequenced.

In the complete genome sequence, 85.6% of the nucleotide sites were determined in multiple cDNA clones or PCR fragments, and 14.4% of the sequence was determined by the sequence of a single cDNA clone. Thus, a minor portion of the genome sequence may contain quasispecies sequence microheterogeneity relative to the consensus of the original FDLV template population. Sequence data were analyzed with MacVector 6.0 and AssemblyLIGN 1.0.9 software (Oxford Molecular Group). Phylogenetic analyses were carried out with the neighbor-joining and parsimony programs in the PAUP* version 4.0b software package (53). Consensus trees were derived from 1,000 bootstrapped replicates of the data sets, and bootstrap values above 70 were considered significant (23).

Analysis of P/V mRNA editing. Confluent IgH2 cells were infected with FDLV at a high multiplicity of infection and allowed to replicate at 28°C for 3 days. Polyadenylated mRNA was extracted (Micro-Fast Track 2.0; Invitrogen) and copied in a RT reaction mixture (13) containing antisense primer 5'-ACACTCTCCGCAGACGCA-3' and 200 U of Moloney murine leukemia virus RT (USB) enzyme. The cDNA was treated with 0.7 mg per ml of RNase A for 15 min at 37°C and heated to 95°C for 2 min before being added to a 50-µl single-round PCR containing 50 pmol of the antisense primer described above, 50 pmol of sense primer (5'-CCGGTGCAACAGCAGGA-3'), and 2.5 units of proofreading polymerase Pfu Turbo (Stratagene). Reaction conditions were as described (2), with an annealing temperature of 50°C and a final extension step at 72°C for 7 min. The 336-bp product was purified (StrataPrep PCR kit; Stratagene) and cloned by using the Perfectly Blunt pSTBlue-1 Cloning kit (Novagen). Clones were screened by the rapid colony PCR procedure and sequenced as above.

Northern blot analysis of U gene transcription. Total RNA (ToTALLY RNA; Ambion Inc.) or mRNA (Micro-Fast Track 2.0) was extracted from IgH2 cells 2 to 3 days postinfection with a high multiplicity of FDLV infection or from uninfected IgH2 cells (3 x 106 cells each). Radioactive 32P-labeled single-stranded RNA probes specific for FDLV U ("unknown") and N gene sequences were created with the Lign'Scribe kit protocols (Ambion Inc.). Initial templates for the probes were PCR products representing a 382-bp region of the U gene (nt 1676 to 2057 of the FDLV genome) and a 369-bp region of the N gene (nt 535 to 903). Single-round PCRs were performed as described above using 5'-ATGTCACACCCCCATCAA-3' and 5'-CCACCCTGTTATCAACCA-3' primers for the U gene and 5'-GGTCAAACAAGGCTGATG-3' and 5'-TCTGAAAGAAACAGGTGAG-3' primers for the N gene. Denaturing gel electrophoresis, blotting, hybridization at 68°C, and autoradiography were as described (NorthernMax; Ambion).

Determination of U gene sequences from other snake paramyxoviruses. Snake paramyxovirus isolates Gono-GER85 and Biti-CA98 were originally from Germany and California, respectively, and have been described (2). These viruses were grown in IgH2 cells at 28°C and used as templates for a first round of RT-PCR (2) with the sense primer 5'-CTATAGGCCAGAAAATGG-3' and antisense primer 5'-CTATTGAAGAGTTGCTTGAA-3' to amplify a 1,450-bp region surrounding the U gene. A second-round PCR was performed with primers 5'-GGTCATAGCTGGCATGTC-3' and 5'-AAGGCTGCTTGCTTCGAAA-3' to yield a 777-bp amplification product. The second-round primers were used to sequence the purified PCR products (StrataPrep; Stratagene).

Nucleotide sequence accession number. The complete FDLV genome sequence has been deposited in the GenBank database under the accession number AY141760.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cloning of FDLV genomic RNA. Several initial attempts to amplify regions of purified FDLV genomic RNA by using degenerate primers designed to anneal with conserved sequences from mammalian and avian paramyxoviruses failed to generate products by RT-PCR (data not shown). Therefore, a randomly primed cDNA library was generated with FDLV genomic RNA as the template. Twenty-five clones were identified that had insert sequences with significant similarity to paramyxovirus N, P, F, HN, or L gene sequences in the NCBI database. The complete sequences of these clones represented 57% of the final FDLV genome sequence. A total of eight RT-PCR fragments, comprising 35% of the final sequence, were then generated and analyzed to fill the gaps between the cDNA clones. Rapid amplification of cDNA ends was used to determine 704 nt at the 3' genome terminus and 519 nt at the 5' terminus.

FDLV transcription regulatory signals and genomic termini. The complete FDLV genome is 15,378 nt in length. This length is very similar to the uniform size range that has been described for most paramyxovirus genomes (37, 38), and it is distinct from the unusually long genomes described recently for HeV and NiV (19, 59). The FDLV genome length is evenly divisible by six, so it is consistent with the rule of six described for members of the subfamily Paramyxovirinae (5, 34).

The FDLV genome sequence contained six putative gene junction sequences that were highly conserved within FDLV and with known gene junctions of other paramyxoviruses (Fig. 1). The FDLV transcription start signals consisted of a 10-base sequence that was strictly conserved for five of the seven genes, and only a single base varied for the other two genes. The consensus of these FDLV transcription start signals was most similar to the consensus start signal for HeV, but similarity was evident with the gene start signals for all Paramyxovirinae. Transcription stop signals for FDLV genes consisted of an 11-base conserved sequence that ended with a 4- to 6-base poly(A) tract (mRNA-sense), as found in all paramyxovirus gene stop signals. Within FDLV, four of the seven gene stop sequences were strictly conserved, and the other three varied at 1, 2, or 4 bases. The consensus FDLV stop signal was identical to that of HeV, and it was highly conserved with the stop signals for many other members of the Paramyxovirinae. The intergenic regions were all trinucleotides that were CCT or CTT. These regions are very similar to the highly conserved trinucleotide intergenic regions found in members of the Henipavirus, Respirovirus, and Morbillivirus genera, and they differ from the variable-length intergenic regions found in the Rubulavirus, Avulavirus, and Pneumovirus genera.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 1. FDLV genome organization (A) and conserved gene junction sequences (B). The horizontal line in panel A indicates the 15,378-nt single-stranded, minus-sense RNA genome, and shaded blocks indicate the position and length of the seven FDLV genes: N, U, P/V, M, F, HN, and L. Transcription regulatory signals (B) are shown in mRNA-sense. Signal sequences of the seven FDLV genes were used to derive the FDLV consensus sequence, which is aligned with consensus sequences from the type species of the genera HeV (59), SeV (38), MeV (GenBank accession no. AF266290), MuV (GenBank accession no. AF314561), NDV (38), and HRSV (GenBank accession no. M11486), and with MenV (4), listed in order of similarity to the FDLV consensus. In the consensus sequences, uppercase letters indicate strict conservation, lowercase letters a, c, g, and t indicate no more than two exceptions, and lowercase letters used by the International Union of Pure and Applied Chemistry indicate more variable sites (as in reference 38). In the NDV consensus sequence, the asterisk indicates a nucleotide deletion (38).

 
The FDLV genome begins with a 55-nt 3' leader sequence before the transcription start site for the N gene, and there is a 36-nt 5' trailer following the L gene transcription stop site (Fig. 2A). The length of the 3' leader is exactly conserved with all other known paramyxoviruses (38). The 5' trailer length is within the 25- to 58-nt range seen in the Respirovirus, Morbillivirus, Rubulavirus, and Henipavirus genera. The FDLV leader and trailer sequences are exactly complementary for the first 18 nt, and if a 2-base gap is allowed in the alignment, then 30 out of the 36 bases in the 5' trailer can form complementary base pairs with the 3' leader (Fig. 2A). The sequence of the first 12 to 13 nt at the 3' ends of paramyxovirus genomes has been found to be genus specific, and it is considered critical as a promoter element for the viral RNA-dependent RNA polymerase (38). The 3' terminal sequence of FDLV is highly conserved with the Henipavirus genus and also with the Respirovirus and Morbillivirus genera (Fig. 2B).



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 2. FDLV genomic termini. Complementarity between the FDLV 3' and 5' genomic termini is shown in panel A, with shading indicating complementary base pairs and dashes indicating deletions in the alignment. Comparisons of the FDLV 3' terminal 20 nt with 3' terminal sequences of other paramyxoviruses is shown in panel B, with shading indicating identical conserved residues. Lowercase letters indicate bases that do not form complementary base pairs between the 3' and 5' genomic ends of each virus (note that no genomic trailer sequence is available for MenV). Numbers of bases matching the FDLV 3' 20-nt sequence are shown to the right of each virus terminal sequence, followed by the GenBank accession number source for each sequence. Abbreviations: PIV3, parainfluenza virus 3; CDV, canine distemper virus; SV5, simian virus 5; VSV, vesicular stomatitis virus.

 
FDLV genome organization. The gene junction sequences of FDLV defined seven discrete, nonoverlapping transcription units. Analysis of the ORFs indicated that five of the transcription units (the first and the fourth through seventh) each contained a single large ORF capable of encoding a protein that could be identified by database similarity searches as one of the known invariant paramyxovirus proteins: N, M, F, HN, and L. The second transcription unit encoded a small protein that had no discernible similarity to any sequence in the NCBI database. The gene and its putative protein product were designated U. The ORF in the third transcription unit was less than half the length of the gene and encoded a protein with similarity to paramyxovirus V proteins. A second ORF in an overlapping reading frame encoded a protein with similarity to the carboxyl terminus of paramyxovirus P proteins. Thus, the third ORF is the paramyxovirus P/V gene. The FDLV genome was thus determined to contain seven genes in the order 3' N-U-P/V-M-F-HN-L 5' (Fig. 1). The presence of an additional gene between the N and P/V genes is a unique feature not seen in any other paramyxovirus.

A summary of the features of the seven FDLV genes and their deduced proteins is provided in Table 1. The total coding percentage for the FDLV genome is 92%, which is exactly the percentage found in all Paramyxovirinae genera except henipaviruses, which have 82 to 84% (59). Subunit hexamer phasing positions for the start sites of the six FDLV genes that are common to all Paramyxovirinae are shown in Table 2. The pattern of the FDLV gene start sites is unique relative to the patterns for other paramyxoviruses, which have been shown to be genus specific within the Paramyxovirinae (19, 34, 59). The FDLV pattern is most similar to that of the morbilliviruses MeV and Rinderpest virus, matching at four out of the six start sites. As has been noted previously (34), position 5 is not used as the start site for any paramyxovirus gene, including those of FDLV.


View this table:
[in this window]
[in a new window]
  		TABLE 1. FDLV genome transcription units and deduced protein products

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Subunit hexamer phasing positions for gene start sites and P editing sites of selected Paramyxovirinae

 
FDLV structural genes. The N protein of paramyxoviruses is an abundant virion protein that encapsidates the viral RNA to form the nucleocapsid. The FDLV N gene transcription unit contains a single large ORF coding for a 471-amino-acid (aa) protein with a calculated size of 51.3 kDa (Table 1). This size is slightly smaller than the currently known range of paramyxovirus N proteins, which is 489 to 553 aa (38). The first AUG codon for this ORF is in strong context for translation initiation (35), and the next AUG is 327 nt downstream, suggesting that the first AUG is the most likely start site. Pairwise alignments indicated that the FDLV N protein has relatively low levels of amino acid identity with the N proteins of viruses in the Paramyxovirinae subfamily, ranging from 21 to 27%, and it has even less identity with HRSV in the Pneumovirus subfamily (Table 3). Alignment of the FDLV N protein sequence with N proteins of 16 other paramyxoviruses clearly showed that the amino-terminal 85% of the protein (aa 1 to 398) was relatively well conserved, while the carboxy-terminal 15% (aa 399 to 471) had almost no similarity (data not shown). This pattern is consistent with previous descriptions in which the C-terminal 20% of paramyxovirus N proteins is poorly conserved and contains most of the phosphorylation and antigenic sites (38). Within the conserved domain of the N proteins, the Y260 and F324 residues that have been reported to be conserved in all Paramyxovirinae are present in FDLV as Y253 and F317, respectively. These amino acids are important for RNA binding and N-N protein self-assembly, respectively (40). The FDLV F317 is also the start of a larger conserved motif found near the middle of all paramyxovirus N proteins, F-X4-Y-X4-SYAMG (38). In FDLV this motif is present with a single amino acid difference, as F-X4-Y-X4-SFAMG.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Pairwise percent amino acid identities of FDLV proteins with analogous proteins in other paramyxoviruses

 
The P protein is an invariant structural protein of paramyxovirus virions that is a component of the active polymerase complex. The V protein is an accessory protein that is expressed from an alternative reading frame in the P gene and has a role in blocking host interferon defense mechanisms (22, 44, 47). The first AUG codon in the 1,433-nt P/V gene is in strong context for translation initiation and starts an ORF that encodes a 227-aa protein. Although the deduced protein of this ORF is poorly conserved with other paramyxovirus proteins overall (Table 3), the carboxy-terminal 69 aa are extremely well conserved and easily identified as the cysteine-rich carboxy-terminal region of a paramyxovirus V protein (38, 54). The P genes of nearly all members of the Paramyxovirinae contain a conserved AnGn (mRNA-sense) RNA editing site, where the viral polymerase inserts nontemplated G residues during mRNA synthesis to access alternative reading frames in the P mRNA (38, 54). In FDLV there is a conserved AnGn site that is predicted to facilitate the insertion of G residues at nt 502 of the P mRNA (Fig. 3). This position is a subunit hexamer phasing position 2 for the FDLV RNA editing site (Table 2). The actual editing site sequence will be described in a separate section below. Insertion of two G residues at this site accesses a longer reading frame to encode a 429-aa P protein (Fig. 3). The FDLV V and P proteins are, thus, amino coterminal for the first 158 aa, and they differ in the carboxy-terminal regions encoded after the RNA editing site. Insertion of a single G residue at the editing site is predicted to encode a protein which is truncated by a stop codon 2 aa after the editing site. This sequence is analogous to the sequence of the I proteins of rubulaviruses (43, 54) and the W proteins of respiroviruses (38). Thus, for the FDLV P gene the primary transcript of the genome encodes the V protein, and edited mRNAs are predicted to express P and I proteins (Fig. 3), which is identical to the expression strategy of rubulaviruses. This strategy is fundamentally different from that of the respiroviruses, morbilliviruses, henipaviruses, and avulaviruses, all of which express P proteins as the primary transcript and V and W/D/I proteins via edited mRNAs (38).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 3. FDLV P gene expression strategy and RNA editing site sequence, compared with other paramyxovirus P gene RNA editing sites. The horizontal scale at the top indicates the 1,433-nt P gene mRNA, and bars below indicate ORFs accessed in the three possible reading frames by ribosomal choice and RNA editing. In the editing site sequences, spacing between the 6 nt upstream, An, and Gn elements is to facilitate visual comparison (21, 38). Accession numbers for sequences not previously noted are as follows: for parainfluenza virus 2, X57559; for parainfluenza virus 4, M55976; for SalV, AF237881.

 
As reported for other paramyxovirus P proteins, the FDLV P protein is poorly conserved within the family, sharing only 12 to 18% amino acid identity (Table 3). At 429 aa, the size of the FDLV P protein is similar to the P proteins of rubulaviruses and avulaviruses, and it is shorter than the P proteins of respiroviruses, morbilliviruses, and henipaviruses (20, 56, 59). Features shared in common among most paramyxovirus P proteins are an acidic pI and a large number of potential phosphorylation sites for serine- or threonine-specific kinases (38). Accordingly, the FDLV P protein has a pI of 5.2, and there are 11 consensus recognition sites for casein kinase II and 6 sites for protein kinase C (32).

The complete FDLV V protein shares 13 to 25% amino acid identity with other paramyxovirus V proteins and is closest to rubulavirus V proteins (Table 3). In contrast, the V-specific carboxy-terminal region encoded after the editing site (aa 159 to 227) shares 34 to 51% amino acid identity with other paramyxoviruses, and the highest identity is with henipaviruses (Table 3). The V protein carboxy-terminal domain is the most conserved region among all paramyxovirus proteins. In the FDLV V protein carboxy-terminal domain, all seven cysteines known to be conserved among other paramyxovirus V proteins are present, forming the zinc finger domain (38, 54). Also in this domain, the invariant H-R-R-E motif is strictly conserved in the FDLV V protein, and the conserved W-C-N-P is present with a single amino acid change to F-C-N-P (38).

The FDLV P/V gene does not contain an ORF analogous to the C ORF of respiroviruses and morbilliviruses in an overlapping reading frame upstream of the editing site. However, there is a smaller ORF upstream of the editing site (nt 215 to 373 of the 1,433-nt P gene), that overlaps the large P/V ORF in the + 1 reading frame (Fig. 3). This small ORF begins with an AUG that is not in optimal context for initiation, but the third codon is also an AUG, but in strong context. The ORF potentially encodes a 53-aa protein with a predicted Mr of 6.1 kDa and a strongly basic pI of 13.4. The protein is arginine rich and contains no acidic amino acids. A similar ORF encoding a protein of 65 aa with a pI of 11.9 has been described in a similar location within the HeV P/V/C gene (58). This HeV ORF was designated SB, for small basic protein, and we therefore refer to the analogous ORF in FDLV as SB.

The M protein of paramyxoviruses is the most abundant structural protein in virions, and it associates with membranes and with the hydrophobic tails of viral F and HN proteins (38). The first AUG codon in the FDLV M gene is in strong context for translation initiation, and it begins an ORF that encodes a 351-aa protein (Table 1). The deduced FDLV M protein has a basic pI of 9.7, which is typical for other M proteins. By pairwise amino acid identity, it is closest to the M proteins of respiroviruses, morbilliviruses, henipaviruses, and TPMV, with which it shares 34 to 37% identity (Table 3). There is less identity to rubulavirus, avulavirus, MenV, and TiV M proteins, and there is no identity with the M protein of the pneumovirus HRSV.

Paramyxovirus F proteins are type I integral membrane proteins that mediate viral penetration into host cells by fusion of the viral envelope with the cell plasma membrane. F proteins are synthesized as inactive precursors that are cleaved by cellular proteases to generate F1 and F2 fragments that are linked by disulfide bonds to create the active form of the protein. The FDLV F ORF encodes a 545-aa protein that is closest in sequence to that of TPMV, followed by respirovirus and henipavirus F proteins (Table 3). Hydrophobicity and transmembrane analyses confirm that the FDLV F protein has features of a type I integral membrane protein. These features include a putative signal sequence at the N-terminal 21 aa and a transmembrane domain of approximately 28 aa near the carboxy terminus (aa 492 to 520), followed by a carboxy-terminal hydrophilic tail of 24 aa (data not shown). There is also a 25-aa hydrophobic domain at aa 111 to 136 that is likely to represent the fusion peptide because it follows immediately after a multibasic consensus cleavage site, R-E-K-R, at aa 107 to 110. Morbilliviruses, rubulaviruses, avulaviruses, pneumoviruses, and some respiroviruses (human parainfluenza virus 3) have a conserved multibasic F protein cleavage site with the consensus sequence R-X-R/K-R that is required for cleavage by the cellular protease furin (24). The presence of this consensus site in the FDLV F protein suggests that it is also cleaved by furin, as has been suggested (17). The FDLV F protein has three potential N-linked glycosylation sites, two in the subunit F1 and one in F2.

Paramyxovirus HN proteins are type II integral membrane proteins on the virion surface that mediate attachment to cell-surface receptor molecules. They are the major antigenic determinants of paramyxoviruses. The FDLV HN gene contains an ORF beginning with an AUG in strong initiation context, encoding a 564-aa protein. The FDLV HN is closest to respirovirus HN proteins and most distant from morbillivirus H proteins (Table 3). Analyses of hydrophobic and transmembrane domains indicate that the FDLV HN protein has features of a type II membrane protein. There is a hydrophobic 25-aa putative transmembrane domain near the amino terminus (aa 26 to 50) following an amino-terminal cytoplasmic tail of 25 aa (data not shown). There are two potential N-linked glycosylation sites, and the acidic pI of 4.9 is within the typical range of 4.8 to 5.5 for HN proteins of other Paramyxovirinae (38). The FDLV HN protein contains the sialic acid binding site motif N-R-K-S-C-S (aa 246 to 251) that is conserved in respiroviruses and rubulaviruses (29) but is not present in morbilliviruses, henipaviruses, or MenV. This finding suggests the probability that FDLV uses a sialic acid-containing cellular receptor similar to respiroviruses and rubulaviruses. The globular head domain of the FDLV HN protein also contains the same six of the seven neuraminidase active-site residues found to be conserved in HN proteins of respiroviruses and rubulaviruses (39). These features are consistent with the known neuraminidase activity of FDLV virions (14).

The polymerase proteins of Paramyxoviridae, like the polymerases of all viruses in the superfamily Mononegavirales, are large proteins that form a complex with N and P proteins to carry out multiple enzymatic functions involved in viral transcription and replication. The L gene of FDLV is 6,640 nt long and has a single large ORF starting with an AUG codon in strong initiation context. This ORF encodes a 2,181-aa protein, which is similar in length to all other paramyxovirus L proteins (38). The full-length FDLV L protein is closest to respiroviruses, with 41 to 43% amino acid identity, followed by morbilliviruses, henipaviruses, and TPMV. Rubulaviruses, avulaviruses, and TiV are notably more distant. Within the L proteins of Mononegavirales, a consistent pattern of conserved and variable sequence domains and subdomains has been described (31, 46). These domains include major conserved domains I to VI, more highly conserved subdomains A to D within domain III, and the highly variable hinge region between conserved domains II and III. Pairwise alignments of the FDLV L protein with L proteins of other paramyxoviruses show a pattern that generally conforms to this pattern (Table 4). Domain II is the best conserved of the major domains, and domain VI is the least conserved, showing little identity above the level seen throughout the entire L protein. The hinge region is uniformly low in amino acid identity, as expected. Within domain III, subdomain IIIA is the most conserved, while subdomain IIID is not conserved above the overall level of identity for domain III. Within domain III, the highly conserved G-D-N-Q sequence that is present in nearly all Mononegavirales is present in FDLV. This sequence is unlike the variant G-D-N-E found in HeV and NiV (19, 59) and in TPMV (GenBank accession no. AF079780). The FDLV domain VI contains a putative ATP-binding site motif, K-X21-A-E-G-S-G, that is identical to the ATP-binding site sequences of rubulaviruses and avulaviruses. This sequence differs from the ATP-binding site sequences of respiroviruses, morbilliviruses, henipaviruses, and pneumoviruses by one to two residues (19).


View this table:
[in this window]
[in a new window]
  		TABLE 4. Pairwise percent amino acid identities of FDLV L protein domains and subdomains with those of other paramyxoviruses

 
Novel U gene structure, transcription, and diversity. The FDLV U gene is 604 nt long and contains a 501-nt ORF starting at nt 24. The first AUG codon is in poor context for initiation, and there is a second AUG 72 nt downstream that is in strong context. It has not been established which codon is the actual initiator of translation. Depending on whether the first or second AUG codon is used, the U ORF encodes a protein of 167 or 143 aa, with predicted sizes of 19.4 or 16.6 kDa. Both putative proteins are basic, with pIs of 9.6 to 10.0. They are unusually rich in tryptophan (3.6 to 4.2% relative to 0.47 to 1.1% in the other FDLV proteins). Regardless of which initiation codon is used, the U protein has a single N-linked glycosylation site and three potential phosphorylation sites for serine- or threonine-specific kinases (32). Hydrophobicity and transmembrane analyses show that although most of the U protein is hydrophilic, there is a 14-aa hydrophobic domain near the carboxy terminus, followed by a 25-aa hydrophilic tail (data not shown). Thus, the U protein may be a small transmembrane protein, but this has not been experimentally tested. The U protein has no similarity to protein sequences currently in the NCBI database. There is also a 330-nt ORF that overlaps much of the beginning of the U gene ORF, but it extends 14 nt upstream of the N gene transcription termination signal, so it could only be expressed from a bicistronic mRNA. There is only one other AUG codon in the U gene, 67 nt downstream of the transcription start. This codon is in poor initiation context, but if it were expressed from the U mRNA by ribosomal choice, it would encode an 82-aa protein with a calculated Mr of 9.6 kDa and a pI of 5.2.

Northern blot analysis of RNA from FDLV-infected cells demonstrated active transcription of U gene mRNA at the anticipated size of approximately 600 nt (Fig. 4). Simultaneous analysis of N gene mRNA transcription as a positive control identified the expected N mRNA at approximately 1.5 kb, and due to the nearly identical sizes of the U- and N-specific radioactive probes, the difference in intensity on the two blots suggested that U mRNA was substantially more abundant than the N mRNA on a molar basis. A transcript of approximately 2.2 kb was also observed in both blots, indicating synthesis of bicistronic N-U readthrough mRNA. Additional, very faint bands at 3.7, 5.1, and 6.7 kb suggested small amounts of additional readthrough products, starting with the N gene and extending up to pentacistronic mRNAs (data not shown).



View larger version (59K):
[in this window]
[in a new window]
  		FIG. 4. Northern blot analysis of U gene mRNA transcription. Preparations of mRNA from mock-infected IgH2 cells (lanes M) and FDLV-infected IgH2 cells (lanes F) were run on a denaturing 1.2% agarose gel, transferred to a BrightStar-Plus membrane (Ambion Inc.), and hybridized with 32P-labeled single-stranded minus-sense RNA probes specific for the FDLV U gene (left panel) or the FDLV N gene (right panel). Migration positions of nonradioactive RNA size markers on the same gel are indicated on the left in kilobases. The weak hybridization of the U probe to a transcript at the N mRNA size was unexpected and may be due to short regions of similarity (maximum identity, 14 of 15 contiguous nt) between the U probe and N mRNA sequences (data not shown).

 
To assess the presence and genetic diversity of U genes in other snake paramyxoviruses, FDLV sequences were used to design RT-PCR primers for amplification of the entire U gene transcription unit, along with short flanking sequences of the N and P genes. Two additional snake paramyxovirus isolates, Gono-GER85 and Biti-CA98, were selected for characterization because partial HN and L gene sequences suggested that they represent two viral genotypes different from FDLV (2). The isolates were obtained from the host species Gonosoma oxycephala and Bitis atropos, which are members of two different major families of snakes. RT-PCR amplification of genomic RNA from Gono-GER85 and Biti-CA98 generated products of the expected size for the U gene. Sequence analysis confirmed that these two additional snake paramyxoviruses also have U genes in their genomes. Each U gene contained an intact transcription unit of 604 nt bounded by transcription start and stop sequences identical to those of the FDLV U gene (Fig. 1). The flanking N gene transcription stop site, P gene start site, and trinucleotide intergenic regions were also identical to those of FDLV (Fig. 1).

The U genes of Gono-GER85 and Biti-CA98 each contained a single ORF with a 3' untranslated region (UTR) of 89 nt, but the 5' UTRs upstream of the ORFs were 14 and 23 nt, respectively. The ORFs thus encoded proteins of 167 and 164 aa, respectively, with predicted sizes of 19.2 to 19.3 kDa and basic pI values of 9.6 to 9.9. The nucleotide sequences of the Gono-GER85 and Biti-CA98 ORFs had 11 and 16% nucleotide divergence from FDLV, respectively, and they had 15% nucleotide divergence from each other. An alignment of the deduced U proteins of FDLV, Gono-GER85, and Biti-CA98 is shown in Fig. 5. The alignment shows that the shorter ORF in Biti-CA98 is due to the absence of the methionine codon that starts the ORFs of the other two viruses. The Biti-CA98 ORF starts three codons later, with an AUG codon that is absent in FDLV and Gono-GER85. The first AUG codon in each of these three U ORFs is in suboptimal context for intitation; the second AUG in each ORF is in strong context, and it is conserved in all three viruses (Fig. 5). The alignment also shows that the region between the first and second AUGs of these three proteins has significantly more amino acid divergence (46 to 57%) than the regions downstream of the second AUG (6 to 13%). This pattern of greater constraint after the second AUG codon suggests that it may be the actual start of translation for these ORFs, but this has not been tested. If U gene translation initiates at the second conserved AUG codon, then all three genes would encode proteins of 143 aa, with predicted sizes of 16.6 to 16.8 kDa and basic pI values of 9.3 to 9.6. These proteins have 6 to 13% amino acid diversity, and if the U proteins start with the first AUG codon, they diverge by 11 to 19%. In either case, FDLV is more closely related to Gono-GER85 than to Biti-CA98, as indicated previously by characterization of partial HN and L gene sequences (2). The level of nucleotide divergence observed among these three snake paramyxovirus U genes (11 to 16%) was similar to the divergence previously observed for their partial HN genes (11 to 18%) and partial L gene sequences (12 to 20%) (2), indicating that the U genes are not more variable than other snake paramyxovirus genes. A virus-specific protein with an apparent size of 16 to 17 kDa has been observed in Western blots of purified Gono-GER85 virion proteins probed with polyclonal antisera against the virus (J. Franke and W. Ahne, unpublished results). Although the identity of this protein has not yet been investigated, its size suggests it may be either the U protein or the P gene I protein (Table 1).



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 5. Alignment of the putative U protein sequences from three different snake paramyxovirus species. The top line presents the full-length amino acid sequence of the deduced FDLV U protein, and in the two lines below, dots indicate amino acids identical to the FDLV sequence in the U protein sequences of Gono-GER85 (Gono) and Biti-CA98 (Biti). Only differences from the FDLV sequence are shown. Note that the Biti-CA98 sequence does not start at the same position as the sequences of the other two genes. The arrow indicates the conserved second methionine residue in all three sequences, which may be the actual start of translation (see text).

 
P gene RNA editing site and editing activity. Comparisons of paramyxovirus RNA editing site sequences and experimental mutagenesis studies have indicated that insertion of G residues during RNA editing occurs by a stuttering process and that a short cis-acting sequence of 6 nt just upstream of the AnGn site modulates the overall frequency of mRNA editing and the number of nucleotides inserted (21). The FDLV editing site and its upstream regulatory sequence are shown in Fig. 3B, along with the editing sites of several other paramyxoviruses. The FDLV editing site is unique among all paramyxoviruses in having a U residue within the An run. This is significant because present RNA editing models hypothesize that RNA editing occurs when the nascent mRNA realigns by slipping back one or two positions along a "slippery" sequence containing the An run, resulting in hybrids that are nearly as stable as the nonslipped hybrid (21). For FDLV, slippage of nascent mRNA by one or two positions would result in hybrids with mismatched A:A and U:U base pairs that are not stable. Thus, the FDLV editing site does not appear to be as slippery as other paramyxovirus editing site sequences. The 6 bases upstream of the FDLV AnGn sequence are also unique among paramyxoviruses, so that although FDLV uses the same editing strategy as rubulaviruses, the editing site is not discernibly more similar to rubulaviruses than to other paramyxovirus genera.

To confirm the location and activity of the putative FDLV editing site, mRNA from FDLV-infected cells was amplified by RT-PCR using primers flanking the editing site in a protocol specific for mRNA-sense templates. The resulting PCR product was cloned, and a 240-nt sequence surrounding the putative editing site was determined for 37 individual clones. Relative to the six Gs encoded directly from the genomic sequence, 24 clones had no insertion, 2 clones had a single G insertion, 10 clones had a two-G insertion, and 1 clone had a three-G insertion. This finding confirms the location of the FDLV editing site and the capacity of the FDLV P gene to encode V, P, and I proteins by RNA editing. Table 5 shows a comparison of the relative proportions of mRNA encoding each of the three possible protein products in infected cells for paramyxoviruses that use a rubulavirus-like editing strategy (i.e., encode V protein as the primary transcript and edit by insertion of two Gs to encode P protein). Among paramyxoviruses that encode P proteins from the primary transcript, SeV generates an average of 62% P and 31% V mRNA in infected cells (56), and MeV edits to generate approximately equal proportions of P and V mRNA both in cell culture and in vivo (6).


View this table:
[in this window]
[in a new window]
  		TABLE 5. P gene mRNA editing frequencies in cell culture infections with Paramyxovirinae having rubulavirus-like RNA editing mechanisms (i.e., encoding the V protein on the primary non-edited transcript)

 
Among the 37 clones sequenced to characterize the FDLV RNA editing frequency, 1 clone contained a deletion of a single A residue 59 nt upstream of the RNA editing site. In an independent experiment, one clone was found with an insertion of 20 As at this same site (data not shown). Other than the editing site itself, there were no other sequence differences found in any of the other clones. The A deletion and insertion occurred within a run of five As in the sequence 5'-AUGG(A)5C-3' (mRNA-sense), which is similar to the consensus FDLV transcription stop signal 5'-UAAG(A)5C-3'. Thus, these variants may be the result of an aberrant polyadenylation reaction, as suggested by Cataneo et al. (6), who observed an insertion of 19 residues after a stretch of five identical residues in an untranslated portion of the MeV F mRNA.

Phylogenetic analyses. The deduced amino acid sequences of the full-length N, P, M, F, HN, and L genes of FDLV were used to infer phylogenetic relationships with other Paramyxoviridae by using Ebola virus as an outgroup. Representative trees illustrating the relationships observed are shown in Fig. 6. The N protein phylogeny shows FDLV separate from all other known paramyxovirus species and genera, branching from near the trunk of the tree between the Rubulavirus, Henipavirus, and Morbillivirus genera. The P, M, and F protein trees are similar in showing FDLV branching from the trunk, but the identities of the nearest branches vary (data not shown). In the HN and L protein trees, by both neighbor-joining and parsimony analyses, FDLV appears in a position basal to the Respirovirus genus but close to the trunk of the trees (Fig. 6B). Thus, phylogenetic analyses of HN and L suggest that, although distant, FDLV is most closely related to the respiroviruses. However, trees of the different proteins are not consistent, indicating variously that FDLV is most distant from respiroviruses (N and P proteins), rubulaviruses and avulaviruses (M, F, and L proteins), or morbilliviruses (HN protein). All of the trees clearly show FDLV as an independent branch not closely related to any known Paramyxovirus genus or species.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 6. Phylogenetic analyses of N and L proteins of FDLV and other Paramyxovirinae. Trees shown are neighbor-joining distance trees with branch lengths accurately representing genetic distance. Bootstrap values above 70 from 1,000 resampled data sets were overlaid on the trees, and Ebola virus, from the family Filoviridae, was used as an outgroup. Viral species within the six established Paramyxovirus genera are indicated in shaded ovals. MenV and TiV were recently described as probable new members of the genus Rubulavirus (4, 49) and were thus included within the genus oval when they fell within a cluster defined by accepted species in the genus.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Despite the global prominence of reptiles as one of the five major classes of vertebrates, they have received relatively little attention as viral hosts. The viruses described from reptiles include members from six DNA virus families and from nine RNA virus families including Paramyxoviridae, Rhabdoviridae, Bunyaviridae, Retroviridae, Caliciviridae, Togaviridae, Picornaviridae, Flaviviridae, and Reoviridae (14). However, the seventh report of the ICTV includes only 12 DNA viruses and 9 RNA viruses from reptiles (37), and the majority of the viruses are not assigned to any genus. The first full-length genome sequence determined for a reptile virus was that of an endogenous retrovirus from pythons, which was reported in 2002 and could not be classified into any of the established retrovirus genera (25). To the best of our knowledge, the FDLV genome described here is the second complete genome sequence of any virus from a reptile host.

The FDLV genome consists of seven genes that potentially encode 10 proteins, including the six invariant proteins found in all paramyxoviruses, three accessory proteins (V, I, and SB) from the P gene, and the unique U protein. Characterization of the FDLV genome sequence has revealed several features in common with all Paramyxoviridae studied to date. The genome organization contains the six invariant paramyxovirus genes in the standard order found in all paramyxovirus species (37, 38). The sizes of the FDLV genes and deduced proteins are all either within the range for known Paramyxoviridae or slightly smaller. The pattern of amino acid conservation among FDLV proteins and homologous proteins in other paramyxoviruses, from most conserved to least, is V-carboxyl domain > L > M, F > N, and HN > V > P (Table 3). This order is exactly the relative order of conservation described among proteins of known Paramyxoviridae in the most recent ICTV report (37). In addition, the 55-nt 3' leader sequence length, complementarity between the 3' and 5' genomic termini, transcription start and stop sequences, and pattern of conserved and divergent domains within the L protein are features FDLV shares with all other paramyxoviruses. Within the Paramyxoviridae family, the FDLV gene order, genome length in a multiple of 6 nt, lack of an M2 protein, evidence for expression of multiple proteins from the P gene locus, conserved V-carboxyl terminal domain, and uniformly low amino acid identities with HRSV proteins (Table 3) indicate conclusively that FDLV is not in the subfamily Pneumovirinae but is a member of the subfamily Paramyxovirinae.

Within the subfamily, various features of the FDLV genome comprise a mosaic with regard to similarity to known genera. The 3' terminal genomic sequence and transcription start and stop sequences are most similar to henipaviruses, but the genome length and 5' trailer size are unlike henipaviruses. The trinucleotide intergenic regions are similar to henipaviruses, morbilliviruses, and respiroviruses. The ATP-binding motif in the L protein is identical to that of rubulaviruses and avulaviruses. The ICTV states that amino acid sequence relationships are the basis of the placement of viruses into genera within the family Paramyxoviridae (37). On this basis, none of the FDLV proteins shares enough amino acid identity with known paramyxovirus proteins to be considered within any of the known genera (Table 3). The FDLV N and P proteins in the 3' portion of the genome share highest amino acid identity with henipaviruses and morbilliviruses, respectively, while the M, F, HN, and L proteins in the 5' region of the genome share highest identity with respiroviruses (Table 3). Within the P gene, the accessory V protein has highest identity to rubulaviruses, but the conserved V-carboxyl domain is most like henipaviruses (Table 3). Phylogenetic analyses of the FDLV proteins also indicate that FDLV is not consistently more closely related to any known paramyxovirus genus or species than to others (Fig. 6). The HN and L protein phylogenies suggested that FDLV was slightly closer to respiroviruses than to other genera. Despite the overall lack of relatedness between the proteins of FDLV and rubulaviruses, the FDLV P gene coding capacity and expression strategy are identical to those of the rubulaviruses. This feature is reminiscent of SalV, which was recently described as follows: "The noted similarity to the rubulaviruses is based only on a common editing mechanism and evidence for any direct phylogenetic relationship is unsupported at this time" (49). The fact that this statement is also true for FDLV, and that FDLV and SalV are not closely related, suggests that perhaps a switch between a respirovirus-like and a rubulavirus-like editing mechanism has occurred independently more than once during the evolution of the Paramyxoviridae.

Other features of the FDLV P gene raised two interesting observations. Lamb and Kolakofsky (38) have recently proposed a nomenclature for the amino-terminal modules of Paramyxovirinae P proteins upstream of the editing site. In this nomenclature, respirovirus and morbillivirus P proteins all have a P-amino 1 module, typified as being approximately 300 aa, with an acidic pI of about 5.5. Rubulaviruses and all other paramyxoviruses are suggested to have a P-amino 2 module, which is shorter and basic in nature, with a length of about 170 aa and pI of about 10.5. The FDLV P protein amino-terminal module is 156 aa, but it has an acidic pI of 4.8. An analysis of several other paramyxoviruses revealed that NDV, HeV, NiV, TPMV, TiV, MenV, and SalV all have acidic P protein amino-terminal modules with pI values of 4.3 to 5.7, similar to FDLV (data not shown). Thus, a basic pI is not a consistent feature if these viruses are all considered to have P-amino 2 modules.

The second interesting observation involved the FDLV P gene SB ORF. As noted by Wang et al. (58), the P gene SB ORF described in HeV, and now also in FDLV, is similar to ORFs reported to encode small basic proteins of 55 to 67 aa in the P genes of two rhabdoviruses (36, 52) and the filovirus Marburg (15). The protein product of one of the rhabdovirus SB ORFs has been detected in virus-infected cells (52). Examination of other paramyxovirus P gene sequences for the potential to encode a protein of 30 to 100 aa with a pI over 10.0 identified 22 potential SB ORFs in a total of 13 different paramyxovirus species (data not shown). Based on size and position within the P gene, the potential SB ORFs most similar to the FDLV SB were present in HeV, MeV, simian virus 5 (SV5), and NDV. Each of these viruses has an ORF upstream of the P gene-editing site that encodes a protein of 50 to 65 aa, with predicted pIs of 11.9 to 13.4. Thus, potential to encode an SB protein is a relatively ubiquitous feature of P genes of viruses in the Mononegavirales superfamily.

Unique features of FDLV not found in any other Paramyxovirus genera include the pattern of the hexamer phasing positions of the gene start sites and the sequence of the RNA editing site, both of which have been described as being genus specific within the family Paramyxoviridae. The most notable unique feature is the presence of the novel U gene between the N and P genes. The U gene was shown to be present, with 11 to 16% nucleotide diversity, in three different snake paramyxoviruses that appear to represent different viral species. This finding suggests that the U gene is most likely a feature common to all reptilian paramyxoviruses. Due to this presence of the novel U gene and the lack of similarity with known Paramyxovirus genera as detailed above, we suggest that the snake paramyxoviruses should comprise a new genus within the subfamily Paramyxovirinae, in the family Paramyxoviridae. We suggest the name Ferlavirus for this new genus, to indicate Fer-de-Lance virus as the type species.


   ACKNOWLEDGMENTS
 
This research was funded, in part, by the U.S. Department of Defense Legacy Resource Management Program Grant number DACA87-00-H-00013 through the Smithsonian Institution.

The authors are grateful for the assistance of Donald Nichols and Elaine Lamirande at the National Zoological Park and Robert Mason at Oregon State University, who provided both technical and intellectual support for this work. Julia Franke is also greatly appreciated for providing unpublished data on viral proteins and for valuable comments on the manuscript.


   FOOTNOTES
 
* Corresponding author. Mailing address: USGS Western Fisheries Research Center, 6505 NE 65th St., Seattle, WA 98115. Phone: (206) 526-6583. Fax: (206) 526-6654. E-mail: gael_kurath{at}usgs.gov. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Ahne, W. 1991. Viral infections of reptiles, p. 1-12. In 4th International Colloquium on Pathology and Medicine of Reptiles and Amphibians. German Veterinary Association, Giessen, Germany.
  2. Ahne, W., W. N. Batts, G. Kurath, and J. R. Winton. 1999. Comparative sequence analyses of sixteen reptilian paramyxoviruses. Virus Res. 63:65-74.[CrossRef][Medline]
  3. Ahne, W., and W. J. Neubert. 1989. Antigenic relationship between three pleomorphic RNA viruses isolated from different snakes. Herpetopathologia 1:65-72.
  4. Bowden, T. R., M. Westenberg, L. F. Wang, B. T. Eaton, and D. B. Boyle. 2001. Molecular characterization of Menangle virus, a novel paramyxovirus which infects pigs, fruit bats, and humans. Virology 283:358-373.[CrossRef][Medline]
  5. Calain, P., and L. Roux. 1993. The rule of six, a basic feature for efficient replication of Sendai virus defective interfering RNA. J. Virol. 67:4822-4830.[Abstract/Free Full Text]
  6. Cattaneo, R., K. Kaelin, K. Baczko, and M. A. Billeter. 1989. Measles virus editing provides an additional cysteine-rich protein. Cell 56:759-764.[CrossRef][Medline]
  7. Chant, K., R. Chan, M. Smith, D. E. Dwyer, and P. Kirkland. 1998. Probable human infection with a newly described virus in the family Paramyxoviridae. Emerg. Infect. Dis. 4:273-275.[Medline]
  8. Chua, K. B., W. J. Bellini, P. A. Rota, B. H. Harcourt, A. Tamin, S. K. Lam, T. G. Ksiazek, P. E. Rollin, S. R. Zaki, W.-J. Shieh, C. S. Goldsmith, D. J. Gubler, J. T. Roehrig, B. Eaton, A. R. Gould, J. Olson, H. Field, P. Daniels, A. E. Ling, C. J. Peters, L. J. Anderson, and B. W. J. Mahy. 2000. Nipah virus: a recently emergent deadly paramyxovirus. Science 288:1432-1435.[Abstract/Free Full Text]
  9. Chua, K. B., L. F. Wang, S. K. Lam, G. Crameri, M. Yu, T. Wise, D. Boyle, A. D. Hyatt, and B. T. Eaton. 2001. Tioman virus, a novel paramyxovirus isolated from fruit bats in Malaysia. Virology 283:215-229.[CrossRef][Medline]
  10. Clark, H. F., F. S. Lief, D. P. Lunger, D. Water, P. Leloup, D. Foelsch, and R. W. Wyler. 1979. Fer-de-Lance virus (FDLV): a probable paramyxovirus isolated from a reptile. J. Gen. Virol. 44:405-418.[Abstract/Free Full Text]
  11. Daszak, P., A. A. Cunningham, and A. D. Hyatt. 2000. Emerging infectious diseases of wildlife-threats to biodiversity and human health. Science 287:443-449.[Abstract/Free Full Text]
  12. de Leeuw, O., and B. Peeters. 1999. Complete nucleotide sequence of Newcastle disease virus: evidence for the existence of a new genus within the subfamily Paramyxoviridae. J. Gen. Virol. 80:131-136.[Abstract]
  13. Emmenegger, E. J., R. M. Troyer, and G. Kurath. 2003. Characterization of the mutant spectra of a fish RNA virus within individual hosts during natural infections. Virus Res. 96:15-25.[CrossRef][Medline]
  14. Essbauer, S., and W. Ahne. 2001. Viruses of lower vertebrates. J. Vet. Med. B. 48:403-475.[CrossRef]
  15. Feldmann, H., and H.-D. Klenk. 1997. Marburg and Ebola viruses. Adv. Virus Res. 47:1-53.
  16. Foelsch, D. W., and P. Leloup. 1976. Fatale endemische Infektion in einem Serpentarium. Tierarztl Prax. 4:527-536.[Medline]
  17. Franke, J., S. Essbauer, W. Ahne, and S. Blahak. 2001. Identification and molecular characterization of 18 paramyxoviruses isolated from snakes. Virus Res. 80:67-74.[CrossRef][Medline]
  18. Halpin, K., P. Young, and H. Field. 1996. Identification of likely natural hosts for equine morbillivirus. Comm. Dis. Intell. 20:476.
  19. Harcourt, B. H., A. Tamin, T. G. Ksiazek, P. E. Rollin, L. J. Anderson, W. J. Bellini, and P. A. Rota. 2000. Molecular characterization of Nipah virus, a newly emergent paramyxovirus. Virology 271:334-349.[CrossRef][Medline]
  20. Harcourt, B. H., A. Tamin, K. Halpin, T. G. Ksiazek, P. E. Rollin, W. J. Bellini, and P. A. Rota. 2001. Molecular characterization of the polymerase gene and genomic termini of Nipah virus. Virology 287:192-201.[CrossRef][Medline]
  21. Hausmann, S., D. Garcin, C. Delenda, and D. Kolakofsky. 1999. The versatility of paramyxovirus RNA polymerase stuttering. J. Virol. 73:5568-5576.[Abstract/Free Full Text]
  22. He, B., R. G. Paterson, N. Stock, J. E. Durbin, R. K. Durbin, S. Goodbourn, R. E. Randall, and R. A. Lamb. 2002. Recovery of paramyxovirus simian virus 5 with a V protein lacking the conserved cysteine-rich domain: the multifunctional V protein blocks both interferon-ß induction and interferon signaling. Virology 303:15-32.[CrossRef][Medline]
  23. Hillis, D. M., and J. J. Bull. 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst. Biol. 42:182-192.[CrossRef]
  24. Hosaka, M., M. Nagahama, and W.-S. Kim. 1991. Arg-X-Lys/Arg-Arg motif as a signal for precursor cleavage catalyzed by furin within the constitutive secretory pathway. J. Biol. Chem. 266:12127-12130.[Abstract/Free Full Text]
  25. Huder, J. B., J. Boni, J.-M. Hatt, G. Soldati, H. Lutz, and J. Schupbach. 2002. Identification and characterization of two closely related unclassifiable endogenous retroviruses in pythons (Python molurus and Python curtus). J. Virol. 76:7607-7615.[Abstract/Free Full Text]
  26. Jackson, O. F., and J. R. Needham. 1983. Rhinitis and virus antibody titres in chelonians. J. Small Anim. Pract. 24:31-36.
  27. Jacobson, E. R. 1986. Viruses and viral associated diseases of reptiles. Acta Zool. Pathol. Antiv. Maint. Reprod. Rept. Capt. 79:73-90.
  28. Jacobson, E. R. 1993. Implications of infectious diseases for captive propagation and introduction programs of threatened/endangered reptiles. J. Zoo. Wild. Med. 24:245-255.
  29. Jorgensen, E. D., P. L. Collins, and P. T. Lomedico. 1987. Cloning and nucleotide sequence of Newcastle disease virus hemagglutinin-neuraminidase mRNA: identification of a putative sialic acid binding site. Virology 156:12-24.[CrossRef][Medline]
  30. Junquiera de Azevedo, I. L., A. R. B. Prieto da Silva, E. Carmona, and P. L. Ho. 2001. Characterization of a paramyxovirus from a Fer de Lance viper (Bothrops jararaca): partial nucleotide sequence of the putative fusion protein. Arch. Virol. 146:51-57.[CrossRef][Medline]
  31. Kamer, G., and P. Argos. 1984. Primary structural comparison of RNA-dependent polymerases from plant, animal, and bacterial viruses. Nucleic Acids Res. 12:7269-7282.[Abstract/Free Full Text]
  32. Kennelly, P. J., and E. G. Krebs. 1991. Consensus sequences as substrate specificity determinants for protein kinases and protein phosphatases. J. Biol. Chem. 266:15555-15558.[Free Full Text]
  33. Kindermann, J., A. Kubber-Heiss, P. Kerschbaumer, and N. Nowotny. 2001. Phylogenetic analysis of the L and HN gene of ophidian paramyxoviruses. Arch. Virol. 146:1021-1035.[CrossRef][Medline]
  34. Kolakofsky, D., T. Pelet, D. Garcin, S. Hausmann, J. Curran, and L. Roux. 1998. Paramyxovirus RNA synthesis and the requirement for hexamer genome length: the rule of six revisited. J. Virol. 72:891-899.[Free Full Text]
  35. Kozak, M. 1991. Structural features in eukaryotic mRNAs that modulate the initiation of translation. J. Biol. Chem. 266:19867-19870.[Free Full Text]
  36. Kretzschmar, E., R. Peluso, M. J. Schnell, M. A. Whitt, and J. K. Rose. 1996. Normal replication of vesicular stomatitis virus without C proteins. Virology 216:309-316.[CrossRef][Medline]
  37. Lamb, R. A., P. L. Collins, D. Kolakofsky, J. A. Melero, Y. Nagai, M. B. A. Oldstone, C. R. Pringle, and B. K. Rima. 2000. Family Paramyxoviridae, p. 549-56l. In M. H. V. van Regenmortel, C. M. Fauquet, D. H. L. Bishop, E. B. Carstens, M. K. Estes, S. M. Lemon, J. Maniloff, M. A. Mayo, D. J. McGeoch, C. R. Pringle, and R. B. Wickner (ed.), Virus taxonomy: classification and nomenclature of viruses. Seventh report of the International Committee for Taxonomy of Viruses. Academic Press, San Diego, Calif.
  38. Lamb, R. A., and D. Kolakofsky. 2001. Paramyxoviridae: the viruses and their replication, p. 1305-1340. In D. M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams and Wilkins, Philadelphia, Pa.
  39. Langedijk, J. P. M., F. J. Daus, and J. T. van Oirschot. 1997. Sequence and structure alignment of Paramyxoviridae attachment proteins and discovery of enzymatic activity for a morbillivirus hemagglutinin. J. Virol. 71:6155-6167.[Abstract]
  40. Meyers, T. M., A. Pieters, and S. A. Moyer. 1997. A highly conserved region of the Sendai virus nucleocapsid protein contributes to the NP-NP binding domain. Virology 229:322-335.[CrossRef][Medline]
  41. Murray, K., P. Selleck, P. Hooper, A. Hyatt, A. Gould, L. Gleeson, H. Westbury, L. Hiley, L. Selvey, and B. Rodwell. 1995. A morbillivirus that caused fatal disease in horses and humans. Science 268:94-97.[Abstract/Free Full Text]
  42. Osterhaus, A. D., R. L. de Swart, H. W. Vos, P. S. Ross, M. J. Kenter, and T. Barrett. 1995. Morbillivirus infections of aquatic mammals: newly identified members of the genus. Vet. Microbiol. 44:219-227.[CrossRef][Medline]
  43. Paterson, R. G., and R. A. Lamb. 1990. RNA editing by G-nucleotide insertion in mumps virus P-gene mRNA transcripts. J. Virol. 64:4137-4145.[Abstract/Free Full Text]
  44. Patterson, J. B., D. Thomas, H. Lewicki, M. A. Billeter, and M. B. A. Oldstone. 2000. V and C proteins of measles virus function as virulence factors in vivo. Virology 267:80-89.[CrossRef][Medline]
  45. Philbey, A. W., P. D. Kirkland, A. D. Ross, R. J. Davis, A. B. Gleeson, R. J. Love, P. W. Daniels, A. R. Gould, and A. D. Hyatt. 1998. An apparently new virus (family Paramyxoviridae) infectious for pigs, humans, and fruit bats. Emerg. Infect. Dis. 4:269-271.[Medline]
  46. Poch, O., B. M. Blumberg, L. Bougueleret, and N. Tordo. 1990. Sequence comparison of five polymerases (L proteins) of unsegmented negative-strand RNA viruses: theoretical assignment of functional domains. J. Gen. Virol. 71:1153-1162.[Abstract/Free Full Text]
  47. Poole, E., B. He, R. A. Lamb, R. E. Randall, and S. Goodbourn. 2002. The V proteins of simian virus 5 and other paramyxoviruses inhibit induction of interferon-ß. Virology 303:33-46.[CrossRef][Medline]
  48. Potgieter, L., R. Sigler, and R. Russell. 1987. Pneumonia in Ottoman vipers (Vipera xanthia xanthia) associated with parainfluenza 2-like virus. J. Wildl. Dis. 23:355-360.[Abstract]
  49. Renshaw, R. W., A. L. Glaser, H. Van Campen, F. Weiland, and E. J. Dubovi. 2000. Identification and phylogenetic comparison of Salem virus, a novel paramyxovirus of horses. Virology 270:417-429.[CrossRef][Medline]
  50. Richter, G. A., B. L. Homer, S. A. Moyer, D. S. Williams, G. Scherba, S. J. Tucker, B. J. Hall, J. C. Pedersen, and E. R. Jacobson. 1996. Characterization of paramyxoviruses isolated from three snakes. Virus Res. 43:77-83.[CrossRef][Medline]
  51. Seal, B. S., D. J. King, and R. J. Meinersmann. 2000. Molecular evolution of the Newcastle disease virus matrix protein gene and phylogenetic relationships among the paramyxoviridae. Virus Res. 66:1-11.[CrossRef][Medline]
  52. Spiropoulou, C. F., and S. T. Nichol. 1993. A small highly basic protein is encoded in overlapping frame within the P gene of vesicular stomatitis virus. J. Virol. 67:3103-3110.[Abstract/Free Full Text]
  53. Swofford, D. L. 1998. PAUP* v 4: phylogenetic analysis using parsimony (*and other methods). Sinauer Associates, Sunderland, Mass.
  54. Thomas, S. M., R. A. Lamb, and R. G. Paterson. 1988. Two mRNAs that differ by two nontemplated nucleotides encode the amino coterminal proteins P and V of the paramyxovirus SV5. Cell 54:891-902.[CrossRef][Medline]
  55. Tidona, C. A., H. W. Kurz, H. R. Gelderblom, and G. Darai. 1999. Isolation and molecular characterization of a novel cytopathic paramyxovirus from tree shrews. Virology 258:425-434.[CrossRef][Medline]
  56. Vidal, S., J. Curran, and D. Kolakofsky. 1990. Editing of the Sendai virus P/C mRNA by G insertion occurs during mRNA synthesis via a virus-encoded activity. J. Virol. 64:239-246.[Abstract/Free Full Text]
  57. Wang, L. F., and B. T. Eaton. 2002. Henipavirus (Paramyxoviridae), p. 641-644. In C. A Tidona and G. Darai (ed.), The Springer index of viruses. Springer-Verlag, Berlin, Germany.
  58. Wang, L. F., W. P. Michalski, M. Yu, L. I