Sapporolike viruses

It is now clearly established that human enteric caliciviruses which display the classic surface structure have a fundamentally different genome organization to the NLVs (Liu et al 1995). They are also phylogenetically distinct from NLVs and have been assigned to a separate genus (Berke et al 1997, Noel et al 1997). The classic caliciviruses were first described in the UK (Madeley & Cosgrove 1976), but it was not until some years later that the prototype Sapporo virus was identified (Terashima et al 1983). The 'interim' classification system allowed differentiation of these viruses by their distinctive morphology but there are also important epidemiological differences (Caul & Appleton 1982). Infections are mainly found in children under 4 years of age suggesting that immunity develops early and is life-long. In contrast to the NLVs, infections caused by the SLVs are seldom associated with large outbreaks, diagnosis is by electron microscopy and positive specimens usually occur as single sporadic cases of gastroenteritis.

The first complete SLV sequence was determined for 'Manchester virus' (Liu et al 1995) with a genome length of 7431 nucleotides. The first open reading frame of Manchester virus starts at nucleotide 12 and contains the characteristic motifs of the 2C-like NTPase, 3C-like protease and RNA polymerase seen in the NLVs. However, the major difference between the genome of Manchester virus and the NLV genomes (Fig. 2) is that the capsid structural protein gene is in the same frame as ORF1 and is contiguous with the RNA-dependent RNA polymerase, giving rise to a single large polyprotein that covers most of the genome. This type of genomic organization is also found in the caliciviruses that infect lagomorphs (Clarke & Lambden 1997). Like all other caliciviruses the SLVs have repeated nucleotide sequence motifs located at the genome 5' terminus and at the start of the capsid open reading frame. Whilst SLVs cannot be grown in cell culture it is assumed that expression of the capsid protein can occur through two routes, either by cleavage from the polyprotein or by direct expression from a separate subgenomic RNA. Preliminary work using a full-length genomic clone of Manchester virus in an in vitro transcription/translation system has shown that the polyprotein undergoes proteolytic cleavage liberating the capsid protein.

Although no volunteer studies have been performed to regenerate SLVs for detailed biochemical studies, it has been possible to purify virions directly from stool samples for further characterization. These studies have shown that the Sapporo virions are comprised of a single major capsid protein of 62kDa (Terashima et al 1983). Attempts to produce the capsid protein in heterologous expression systems have met with mixed fortunes. The Sapporo virus capsid protein is exported to the cell culture supernatant when expressed in insect cells using recombinant baculoviruses but formation of VLPs appears to be dependent on the length and nature of 5' leader sequences (Jiang et al 1999). In our hands expression of the Manchester virus capsid protein from a recombinant baculovirus produces abundant capsid protein in the cell culture supernatant but we have not been able to obtain VLPs.

Computer analysis of the Manchester virus genome sequence predicts a typical calicivirus short 3' terminal ORF of 165 amino acids. This is frame shifted —1 relative to main ORF (ORF1) and encodes a 17.8 kDa basic protein, hydrophilic in nature. A second short ORF (encoding 161 amino acids) is also predicted to overlap the capsid region of the genome but is in a different reading frame. This overlapping ORF is found in several SLV isolates but its significance remains unknown.

The availability of the genome sequence has led to the development of oligonucleotide primers for further amplification of related SLV sequences and

FIG. 3. An unrooted phylogenetic tree comparing the amino acid sequences of calicivirus structural capsid proteins. The tree shows the relationship of the 'Norwalk-like' calicivirus genus with the 'Sapporo-like' virus genus. Accession numbers (in parentheses) for caliciviruses are as follows: Manchester SLV (X86560), Houston DCC SLV (U95643), Parkville SLV (U73124), Houston 90 SLV (U95644), Stockholm (AF194182), Bristol 98 SLV (AJ249939), London/92 SLV (U95645), Porcine enteric calicivirus (AF182760), Southampton NLV (L07418), Norwalk virus (M87661), Desert Shield NLV (U04469), Bovine Jena virus (AJ011099), Lordsdale NLV (X86557), Hawaii NLV (U07611).

FIG. 3. An unrooted phylogenetic tree comparing the amino acid sequences of calicivirus structural capsid proteins. The tree shows the relationship of the 'Norwalk-like' calicivirus genus with the 'Sapporo-like' virus genus. Accession numbers (in parentheses) for caliciviruses are as follows: Manchester SLV (X86560), Houston DCC SLV (U95643), Parkville SLV (U73124), Houston 90 SLV (U95644), Stockholm (AF194182), Bristol 98 SLV (AJ249939), London/92 SLV (U95645), Porcine enteric calicivirus (AF182760), Southampton NLV (L07418), Norwalk virus (M87661), Desert Shield NLV (U04469), Bovine Jena virus (AJ011099), Lordsdale NLV (X86557), Hawaii NLV (U07611).

there are now a number of partial sequences available. Immune electron microscopical studies suggested that there are a number of SLV 'serotypes'. This observation has been supported by phylogenetic analysis of the accumulating SLV sequences (Berke et al 1997, Noel et al 1997). It is now clear that the human SLVs can also be divided into two discreet genogroups based on sequence comparison of both RNA polymerase and capsid regions of the genome (Fig. 3). Within these two genogroupings there are distinctive but clearly related 'types' of virus (e.g. Parkville, Houston/90).

0 0

Post a comment