Analysis of the serotype and genotype correlation of VP1 and the 5' noncoding region in an epidemiological survey of the human enterovirus B species

Abstract

The sequence identity of the enterovirus VP1 gene has been shown to correlate with the serotype concept. Enterovirus molecular typing methods are therefore often based on sequencing of the VP1 genomic region and monophyletic clustering of VP1 sequences of a homologous serotype. For epidemiological surveillance, 342 enterovirus samples obtained from patients with aseptic meningitis in Belgium from 1999 to 2002 were first diagnosed as being enterovirus positive by amplification of the 5' noncoding region (5'NCR) by reverse transcription (RT)-PCR. Subsequently, samples were molecularly typed by RT-nested PCR amplification and sequencing of a portion of the VP1 gene. Phylogenetic analyses were performed to investigate enteroviral evolution and to examine the serotype and genotype correlation of the two genomic regions. Our typing results demonstrated echovirus 30, echovirus 13, echovirus 18, and echovirus 6 to be the most predominant types. Echoviruses 13 and 18 were considered to be emerging human serotypes since 2000 and 2001, respectively, as they had been rarely reported before. Several serotypes existed as multiple genotypes (subtypes) from 1999 to 2002, but genomic differences mainly resided at synonymous sites; these results strongly suggest that the subtypes exhibit similar antigenic properties. Phylogenetic analyses confirmed that VP1 is an adequate region for molecular typing. Serotype-specific clusters are not observed commonly in phylogenetic trees based on the 5'NCR, and the phylogenetic signal in the 5'NCR was found to be particularly low. However, some substructure in the 5'NCR tree made a tentative prediction of the enterovirus type possible and was therefore helpful in PCR strategies for VP1 (e.g., primer choice), provided some background knowledge on the local spectrum of enteroviruses already exists.

FIG. 1.

Diagram showing locations of oligonucleotide primers and PCR products in the enteroviral genome (Table 1). IRES, internal ribosome entry site.

FIG. 2.

VP1 phylogram (left) and 5′NCR phylogram (right) based on 342 partial VP1 sequences (346 to 358 bp) and 342 partial 5′NCR sequences (185 bp), respectively. Both neighbor-joining trees were evaluated with 100 bootstrap pseudoreplicates; only bootstrap values of over 75% are shown. The scale bar represents the genetic distance (nucleotide substitutions per site). Distinct clusters representing strains of the same (sub)type are shown as triangles (E, echovirus; CV, coxsackievirus). An illustration of the detailed trees demonstrating the positions of the individual samples can be found at the following website: http://www.kuleuven.ac.be/rega/mvr/Figure_trees_detailed_JCM.pdf. Symbols used in the 5′NCR tree: *, E13 taxon; •, CVB3 taxon; ▪, CVB4-b taxon; ○, E30-c taxon; □, E9-a taxon.

FIG. 3.

Scatter diagram of the mean percent VP1 sequence divergence versus the mean percent 5′NCR sequence divergence within each genotype, excluding the genotypes with only one or two representatives. R² represents the coefficient of determination of the trend line.

FIG. 4.

Likelihood mapping results for VP1 (a), 5′NCR (b), and a concatenated alignment (c).