Search

The 2A protein appears to be diverse among picornaviruses, in contrast to the other non-structural proteins, which have homologous structures and functions. In enteroviruses and rhinoviruses, 2A is a trypsin-like protease involved in protein processing and in shut-off of host-cell macromolecular synthesis. The aphthovirus and cardiovirus 2A is associated with an unusual processing event at the 2A/2B junction. It is shown here that the 2A protein of several diverse picornaviruses, the human parechoviruses, Aichi virus and avian encephalomyelitis virus, possess previously unrecognized conserved motifs and are likely to have a common function. Moreover, these motifs, a conserved histidine and flanking amino acids, an asparagine-cysteine dipeptide and a putative transmembrane domain, are characteristic of a family of cellular proteins, at least two of which are involved in the control of cell growth. These observations have important implications for an understanding of picornavirus genome structure and evolution, as well as pointing to possible functions of 2A in these viruses.

Picornaviruses have been divided into five genera until recently, when a sixth genus, Parechovirus, was defined. Human parechovirus type 1 (HPeV1; formerly echovirus 22) was the first recognized member of this genus and preliminary sequence analysis of echovirus 23 [now renamed human parechovirus type 2 (HPeV2)] suggested that it is also a parechovirus. Here we describe the complete nucleotide and predicted amino acid sequences of HPeV2, which indicate a close relationship to HPeV1 throughout the genome. Sequence covariance in the 5' untranslated region allows a prediction of the secondary structure, which indicates that these parechoviruses have a type 2 internal ribosome entry site, most closely related to that of cardioviruses. Overall, HPeV2 has 87.9% amino acid identity with HPeV1, most divergence being seen in regions of the capsid proteins that probably define antigenic sites. The N-terminal sequence extension to VP3, seen only in parechoviruses, is highly basic in both viruses, but has a variable sequence, suggesting that it does not have a sequence-specific role. There is an RGD motif near the C terminus of VP1, in an analogous location to that in HPeV1 which is believed to be functionally significant. The results confirm that both viruses are parechoviruses and give insights into the molecular features of this genus.

We have analysed, by PCR using consensus primers followed by sequencing, 12 human rhinoviruses (HRVs) in a genomic region including that corresponding to the immunogenic site NIm-II. Together with published information, 21 sequences are available for comparison. In the region analysed, which encodes 112 amino acids, the majority (18) of the serotypes exhibited at least 70% amino acid identity to one another and some serotypes are very closely related. These include HRV-36, -58 and -89, known to exhibit antigenic cross reactivity, which were shown to differ at only three amino acid positions. Three serotypes, HRV-3, -14 and -72, share at least 84% identity with one another but are less than 66% identical to the majority group. Interestingly, membership of these two molecular clusters correlates with the groupings determined by sensitivity to antivirus drugs, suggesting that they reflect a fundamental division of HRVs. In contrast, there is no correlation with receptor grouping, since the majority group contains members belonging to both HRV receptor groups.

Enterovirus type 70 (EV70) is the causative agent of acute haemorrhagic conjunctivitis and may also give rise to a rare neurological complication closely resembling poliomyelitis. The complete nucleotide sequence of the genome of EV70 has been determined from cDNA cloned in Escherichia coli. The genome consists of a 5' non-coding region of 726 nucleotides (nt), a long open reading frame of 6582 nt and a 3' non-coding region of 82 nt prior to the poly(A) tract. Comparison of the nucleotide sequence and the predicted amino acid sequence of the polyprotein with those published for other enteroviruses reveals sufficiently high similarity to predict antigenic regions and polyprotein cleavage sites. The P1 region of EV70 is as similar to those of the entero- as to those of the rhinoviruses, whereas the P2 and P3 regions are more closely related to the coxsackie B and swine vesicular disease viruses than other entero- or rhinoviruses.

We have cloned and sequenced the complete genome of a strain of poliovirus type 3 (23127) isolated during an outbreak of poliomyelitis in Finland. The genome is 7435 nucleotides long excluding the 3' poly(A) stretch and is 95.5% homologous at the amino acid level to the previously sequenced type 3 strain, P3/Leon/37. The most striking feature of the presented sequence is the extent of amino acid substitutions relative to P3/Leon/37 and other type 3 strains in areas of known antigenic importance. The major antigenic determinant for virus neutralization (site 1), located at residues 89 to 100 of VP1, has three amino acid substitutions and there are six substitutions in site 3, a composite site made up of sequences from VP1 and VP3. The variation in these regions probably accounts for the observed unusual antigenicity and may explain why the virus was able to spread in a well-vaccinated community. Sequence comparisons imply that the virus is not derived from the currently used live attenuated vaccine.

We have determined the complete nucleotide sequence of coxsackievirus A21 (CAV-21), the first member of this enterovirus subgroup to be analysed in molecular detail. The sequence, which is 7401 nucleotides long, encodes an open reading frame of 2206 codons, preceded by a 5' non-coding region of 711 nucleotides and followed by a 3' non-coding region of 72 nucleotides plus a poly(A) tract. The most striking feature is the remarkable homology to the poliovirus (greater than 90% at the amino acid level) in the 3' part of the genome. The rest of the genome is much less homologous, suggesting that CAV-21 is a recombinant virus. Rhinovirus-like characteristics, including the length of the 5' non-coding region and a slight --U/--A imbalance in codon usage, may be related to the fact that CAV-21, like rhinoviruses, infects the upper respiratory tract. However, the sequence sheds little light on the molecular basis of the shared receptor specificity.

We have determined the complete nucleotide sequence of human rhinovirus 1B and made comparisons with other rhinoviruses. Extensive homology was found with serotypes 2 and 89 but the similarity to serotype 14 was considerably less. Rhinovirus-specific characteristics have been noted, in particular the length of the 5' non-coding region and the pattern of codon usage, and these may be sufficient to define the rhinoviruses as a distinct genus rather than being considered as members of the enteroviruses as has been suggested previously.