The cis-acting replication elements define human enterovirus and rhinovirus species

Abstract

Replication of picornaviruses is dependent on VPg uridylylation, which is linked to the presence of the internal cis-acting replication element (cre). Cre are located within the sequence encoding polyprotein, yet at distinct positions as demonstrated for poliovirus and coxsackievirus-B3, cardiovirus, and human rhinovirus (HRV-A and HRV-B), overlapping proteins 2C, VP2, 2A, and VP1, respectively. Here we report a novel distinct cre element located in the VP2 region of the recently reported HRV-A2 species and provide evolutionary evidence of its functionality. We also experimentally interrogated functionality of recently identified HRV-B cre in the 2C region that is orthologous to the human enterovirus (HEV) cre and show that it is dispensable for replication and appears to be a nonfunctional evolutionary relic. In addition, our mutational analysis highlights two amino acids in the 2C protein that are crucial for replication. Remarkably, we conclude that each genetic clade of HRV and HEV is characterized by a unique functional cre element, where evolutionary success of a new genetic lineage seems to be associated with an invention of a novel cre motif and decay of the ancestral one. Therefore, we propose that cre element could be considered as an additional criterion for human rhinovirus and enterovirus classification.

FIGURE 1.

Schematic diagrams of human rhinovirus (HRV) and enterovirus (HEV) genomic organization showing the location of known cre elements in HEV, HRV-A, and HRV-B as well as newly predicted HRV-A2 cre and putative HRV-B cre (underlined) that are characterized in the text.

FIGURE 2.

Effects of nucleotide changes within the putative HRV14 2C cre motif on viral growth. (A) Quantification of mutant virus growth in HeLa cells. Each HRV14 2C cre mutant is named according to the substituted nucleotide. The amino acid changes that follow the nucleotide mutations are indicated in parentheses after the mutation and those conserving the amino acid properties are marked by an asterisk. The frequency plot of amino acid residue conservation in 25 HRV-B serotypes is shown at the foot of Panel A (http://weblogo.berkeley.edu/). Quantification of virus growth was measured by immunofluorescence 12 h post-infection and expressed as the mean of positive cells per total cells and expressed relative to that of HRV14-WT. (B) Schematic representation of the putative HRV14 2C cre structure as predicted by MFOLD from nucleotides 4256 to 4303. Nucleotide substitutions at position 4277, 4278, 4279, and 4286 (corresponding, respectively, to A₁, A₂, R², and R³ positions within the consensus R¹NNNA₁A₂R²NNNNNNR³ sequence) are represented in bold. (C) Quantification of virus growth for HRV14-DM versus HRV14-WT. Measurements were conducted by immunofluoresence as described in A. (D) Schematic representation of 2C cre motif for HRV14-DM (nucleotides 4256–4303) following the introduction of 12 silent mutations (surrounded nucleotides) as predicted by MFOLD. None of the crucial positions within the consensus cre loop sequence were mutated, except one at position 4279 (A → G) known to be permissive for efficient HRV14 replication (A). (E) HeLa cells infected by HRV14-WT, HRV14-DM, or Mock. Infected positive cells were detected by immunofluorescence (IF).

FIGURE 3.

WPGMA structure-based cluster tree for 86 HEV and 25 HRV-B 2C cre elements. A consensus 2C cre structure is shown for each cluster except for HRV42 and HRV93, which do not belong to any specific cluster (marked *). The crucial positions within the consensus cre structures are marked by bold letters for each cluster.

FIGURE 4.

Conservation of a predicted VP2 cre secondary structure for HRV-A2. Multiple sequence alignment across all available full-length HRV-A2 (NAT001, NAT045, EF186077, and EF582385-7) and VP2 partial sequences (Nat059, NAT069, Nat083, EU131891-2, and EF512649-65) showing a consensus secondary cre structure in VP2. The structure is shown in the dot-bracket format above alignment. Each corresponding bracket represents a consensus base pair of the alignment columns beneath. Sequences are color-coded according to consistent and compensatory mutations in the aligned sequences regarding the conserved structure (see figure text box). The sequence conservation profile is shown in gray bars below the alignment. The secondary structure of the conserved predicted cre VP2, color-coded according to the different types of base pairs in the corresponding alignment columns, is shown on the left side. The conserved cre motif nucleotides are marked in bold. Note that the amino acid sequence corresponding to the loop region is almost 100% conserved in all species (C-G-F-S-D-R-L-K-Q-I-T-I-G/N-S-T). Mutations supporting the structure (consistent mutations) occur almost exclusively at the third codon position, which leads to synonymous codons and the amino acid conservation.

FIGURE 5.

Boxplots of structural distances within the respective WPGMA trees of HRV-A 2A cre, putative HRV-B 2C cre, HRV-B VP1 cre, predicted HRV-A2 VP2 cre, and HEV 2C cre. The mean distances are 400, 1229, 654, 282, and 475, respectively. Note that the nonfunctional putative HRV-B 2C cre shows the largest distances, thus showing that the structures of the individual cre sequences of these species do not form a tight cluster of mostly similar structures as is the case for the other cre motifs. The distance distributions of all the other elements are below that of the nonfunctional cre element. The distances are directly related to the scores for the clusters from the LocARNA package.