Molecular Epidemiology of Human Enterovirus 71 Strains and Recent Outbreaks in the Asia-Pacific Region: Comparative Analysis of the VP1 and VP4 Genes

Mary Jane Cardosa, David Perera, Betty A. Brown, Doosung Cheon, Hung Ming Chan, Kwai Peng Chan, Haewol Cho, Peter McMinn

Emerging Infectious Diseases. 2003;9(4) 

In This Article


Figure 1 presents an overview of the VP4-based phylogenetic tree, generated by including representative members of each of the genogroups A, B, and C. While there are a few outliers, this comprehensive VP4-based dendrogram accurately reproduces the genogroup clusters B1, B2, B3, B4, C1, and C2[21,22] supported by the bootstrap values indicated. However, the Korean isolates form a distinct cluster in genogroup C with high bootstrap support, which we have designated genogroup C3. Selected HEV71 strains from the collection of the Centers for Disease Control and Prevention (CDC) were used to anchor the tree to maintain consistency of nomenclature between our VP4-based tree and the VP1-based tree published earlier by Brown et al..[21] The viruses in genogroups B and C share at least 78.3% nucleotide sequence identity. Within the B genogroup, the strains share at least 87.9% identity; the strains within the C genogroup share at least 84.5% identity. Overall, the C genogroup has greater diversity than the B genogroup. Within all subgenogroups, virus strains have >90% nucleotide sequence identity ( Table 2 ). The divergence between genogroups A and B as well as A and C is 20% to 21%; the divergence between genogroups B and C is 16% to 25%. The strains in our collection from two of the more recently described genogroups B3 and C3 showed very high similarity to each other within the genogroup (>98% and 99% identity, respectively), while those within the B1 and C1 genogroups showed the widest divergence. The phylogenetic relationships between the consensus sequences of the different genogroups based on VP4 analysis are shown as an unrooted tree (Figure 2). This cladogram clearly illustrates that the C genogroup viruses are more divergent than the B genogroup viruses; however, because we might not have strains that are evenly distributed temporally and geographically, we were unable to determine if this difference in divergence means that genogroup B viruses have evolved more recently than the genogroup C viruses.

The four different genogroup B clusters described in our earlier VP1-based study[22] were accurately reproduced with VP4 gene sequences (Figure 3). This approach allows the inclusion of HEV71 strains from Taiwan and Japan published by Shimizu et al.[15] and Chu et al..[19] We show that, in the 1998 Taiwan epidemic, HEV71 strains belonging to genogroup B4 cocirculated in small numbers (e.g., 7008/98, 5929/98, and J1263P4). These strains are genetically similar to B genogroup viruses circulating in Japan during 1997 (V-14433, V-14429, C7/Osaka) but are distinct from the B3 genogroup of viruses circulating at the same time in Sarawak (e.g., MY104-9/SAR/97 and SK-EV006). Genogroup B4 strains have been circulating in the Asia-Pacific region at least since 1997, as these strains were isolated sporadically in Japan and Singapore in 1997 and in Taiwan during 1998, before the large outbreaks in 2000 in Singapore, Sarawak, and Taiwan associated with viruses from this genogroup. HEV71 strains of the B4 genogroup continued to be isolated in Singapore during 2001 and 2002.

A Taiwanese group reported high death rates during an HEV71 outbreak in 2000.[11] Using phylogenetic analysis based on 840 nt of the VP1 gene (full length 891 nt), these investigators described a shift in genogroup from C in 1998 to B in 2000. To place the most recent Taiwanese strains in a regional context, we have included four representative genogroup B strains from the 2000 outbreak in Taiwan as well as two Taiwanese genogroup B strains from 1998 and 1999 in a limited VP1 gene-based dendrogram, together with representative strains from genogroups B1, B2, B3, and B4 (Figure 4). This analysis clearly identifies the Taiwan 2000 strains as belonging to genogroup B4. The temporal shift of B4 genogroup strains that occurred from one lineage in 1997 through 1999 (in Taiwan and Singapore) to a second lineage in 2000 through 2001 (in Taiwan, Singapore, and Sarawak) gave rise to large epidemics in these countries in 2000.

Another notable point that arises from VP4 gene-based phylogenetic analysis of genogroup B is that some older U.S. strains (2228-NY72 and 6910-OK87), which clearly belong within genogroup B1 in VP1-based analysis,[21] do not lie within B1 in the VP4-based dendrogram (Figure 3). Moreover, in the United Kingdom a similar B genogroup "outlier" was isolated as recently as 1999. These outliers in the VP4-based analysis are very closely related to the B1 cluster, and this discrepancy probably results from the lower discriminating power obtained when using the VP4 gene for this analysis, as evidenced by lower bootstrap values at the major nodes.

Phylogenetic analysis that uses VP4 gene sequences also confirms previous observations that the major strains circulating in the Taiwan outbreak of 1998 were from genogroup C, more specifically C2, following the nomenclature of Brown et al.[21] (Figure 5). The Taiwanese isolates of 1998 clustered closely together in a lineage related to C2 strains isolated in Japan in the previous year and to those causing severe neurologic disease in Perth in 1999. This finding is supported by VP1-based analysis of Taiwanese strains from 1998 (Figure 6). We also located in GenBank a number of VP4 gene sequences of strains isolated in the United Kingdom between 1997 and 1999. These strains formed a cluster within genogroup C2, together with strains from Japan and Australia during 1997 and 1999, respectively. In our previous VP1-based study, the Australian C2 genogroup strains isolated during 1999 formed two distinct lineages,[22] which are reproduced in the VP4 analysis. Similarly, Japanese isolates from 1997 form distinct lineages within genogroup C2.

An outbreak of HFMD with cases of aseptic meningitis and paralysis occurred in Korea during 2000; 11 strains of HEV71 from this outbreak are included in this study. All Korean strains formed a distinct cluster in genogroup C and showed 90% to 92% nt sequence identity with C1 and C2 subgroups and 79% to 82% identity with the different genogroup B strains. A dendrogram comparing the VP1 gene sequences of a limited number of HEV71 genogroup C strains with the 11 Korean isolates from 2000 and the single 2002 isolate from Sarawak is shown in Figure 6. This VP1-based dendrogram supports the VP4-based analysis, suggesting that the Korean HEV71 isolates form a distinct cluster, which we have designated C3. The Sarawak strain isolated in 2002 is from the C1 genogroup. Chu et al.[19] described the major genogroup circulating in Taiwan in 1998 as C3. However, because of the extent of variation in clusters, our previous analysis based on the VP1 gene designates the Taiwan outbreak strain as belonging to genogroup C2;[22] this current study establishes the designation C3 for the new Korean strains isolated during 2000.

The Japanese 1978 isolate "Yamanashi" is an outlier of the C genogroup. Twenty years later (1998), another C genogroup outlier was isolated in China (SHZH98). Both of these strains appear to be ancestral to C1, C2, and C3, raising the possibility that China may still have ancestral strains of HEV71 currently circulating.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: