The Novel Genetic Background of Infectious Bursal Disease Virus Strains Emerging from the Action of Positive Selection

Abstract

The circulation in Europe of novel reassortant strains of infectious bursal disease virus (IBDV), containing a unique genetic background composition, represents a serious problem for animal health. Since the emergence of this novel IBDV mosaic was first described in Poland, this scenario has become particularly attractive to uncover the evolutionary forces driving the genetic diversity of IBDV populations. This study additionally addressed the phenotypic characterization of these emergent strains, as well as the main features affecting the viral fitness during the competition process of IBDV lineages in the field. Our results showed how different evolutionary mechanisms modulate the genetic diversity of co-existent IBDV lineages, leading to the error catastrophe effect, Muller ratchet effect, or prevalence, depending on their genetic compositions. We also determined that the action of the positive selection pressure, depending on the genomic segment on which it is acting, can drive two main phenotypes for IBDV: immune-escaping strains from the selection on segment A or strains with functional advantages from the selection on segment B. This last group seems to possess an increased fitness landscape in the viral quasispecies composition, presenting better adaptability to dissimilar environmental conditions and likely becoming the dominant population. The reassortant strains also exhibited a lower mortality rate compared with the well-known vvIBDV strains, which can facilitate their spreading.

Keywords: genetic evolution; in vivo pathogenicity; reassortment; very virulent infectious bursal disease virus.

Figure 1

Comparison of amino acid substitutions at different positions on VP2 and VP1 proteins between Polish IBDV and reference strains. Dots show an identity with D6948. A—alanine, R—arginine, N—asparagine, D—aspartic acid, Q—glutamine, E—glutamic acid, G—glycine, H—histidine, I—isoleucine, L—leucine, K—lysine, M—methionine, F—phenylalanine, P—proline, T—threonine, Y—tyrosine, V—valine.

Figure 2

Evolutionary history and population dynamics reconstruction of Polish IBDV strains. (A) Maximum clade credibility (MCC) tree based on 175 HVR sequences from the G3a genogroup of IBDV (left) clades A1–A3. The Polish VP2 strains were analyzed for the estimation of tMRCA and phylodynamic reconstruction (right) by a Bayesian skyline plot (BSP) using an exponential, uncorrelated clock model. The x-axis is in units of years, and the y-axis represents the logarithmic scale of Neτ (where Ne is the effective population size and τ is the generation time). (B) MCC tree based on 175 VP1 sequences (left), clades identified within the vvIBDV lineage defined by Polish IBDV strains, and the intermediate B-linage recently reported by Pikuła et al. [32] strains were analyzed for estimation of tMRCA and phylodynamic reconstruction (right) by BSP using an exponential, uncorrelated clock model. The x-axis is in units of years, and the y-axis represents the logarithmic scale of Neτ (where Ne is the effective population size and τ is the generation time). In all cases, Polish IBDV sequences are denoted in red, and the most probable year for the MRCA within each clade or lineage is denoted. The new intermediate B lineage is denoted in green. For simplification, those clades or lineages from sequences that were not directly related with the Polish sequences were collapsed. The rate of substitution per site per year (Mr) is denoted for each clade or lineage. The support of the clades of interest denoted in the figure were supported by a posterior probability value of pp = 1.0.

Figure 3

Representation of the topology and identification of positive selected branches and codon sites of Polish IBDV strains. (A) Maximum likelihood (ML) phylogenetic tree based on 175 HVR sequences from the G3a genogroup of IBDV clades A1–A3, grouping the Polish VP2 strains, and (B) ML phylogenetic tree based on 175 VP1 sequences. In all cases, the clades and lineages that grouped the Polish IBDV strains were analyzed for the branch site model to estimate positive selection using PAML, and the branches that yielded a dN/dS (ω) > 1 are denoted. Polish IBDV sequences are denoted in red. Codon sites found in specific lineages under positive selection (see Table S1) were mapped on the structure of the HVR of the crystal representation of the VP2 protein (PDB code: 2df7) and VP1 (PDB code: 2pgg) using Pymol. Clades A1–A3 for the Polish strains based on VP2 analysis and all B-lineages are denoted. Arrows represent positively selected lineages that also contain codon sites positively selected, which represent an evolutionary advantage.

Figure 4

Mapping of positively selected sites and functional divergence sites on the three-dimensional structure of VP2 and VP1 of IBDV. (A) Folding of the monomer of the VP2 protein (PBD:2DF7); the shell domain is denoted in aqua and the HVR is denoted in light pink, positively selected sites identified by the site models M2a and M8 vs. M1 and M7, respectively (see Table S2), are denoted with red spheres, and the functional role for each position under positive selection is also denoted. (B) Folding of the monomer of the VP1 protein (PDB:2PGG); the N terminus domain is presented in yellow, the fingers domain is presented in blue, the palm is denoted in light red, the C terminal domain is presented in hot pink, the functional divergent selected sites are denoted by red spheres, and the functional role for each Type II functionally selected position is also denoted.

Figure 5

Molecular and biological comparison between the novel IBDV reassortant (vvIBDV-A/transIBDV-B) strains and vvIBDV (vvIBDV-A/vvIBDV-B) strains. (A) Schematic representation of the genome characteristic of the novel reassortant strains, the vvIBDV strain used in the experimental infection study, and the attenuated IBDV strains (ϕ only used for comparison purposes in the molecular features) are shown. The representations of both genome segments A and B are presented, all of the molecular signatures found in the ORF encoding regions are denoted in the alignment, and the positions are denoted. All of the amino acid replacements found in the mature protein were denoted in the 3D structure for each protein. All of the amino acid replacements in the 3D structure were denoted using spheres and using the same color code presented in the amino acid alignments. For the VP5 protein, the model generated by [45] and kindly provided by Dr. Ganguly was used, and for the remaining proteins VP2: 2DF7, VP3: 2Z7J, and VP4: 4IZJ PDB, cartons were used. All of the functional domains were amino acid replacements, shown by the novel reassortant strains, are denoted. (B) Survival curves of the challenged SPF chickens; different curves indicated different groups, and each group was denoted. (C) Estimated bursa damage index determined to 3–5 dpi in the left panel and 10 dpi in the right panel. (D) Histological overall scoring. In all cases, significant differences by Brown–Forsythe and Welch ANOVA for multiple comparisons with a Dunnett’s T3 multiple comparisons were denoted (**** p < 0.0001). In all cases, the groups are denoted (Group A: animals inoculated with the reassortant strain Bug/03/Poland/2003, Group B: animals inoculated with the reassortant strain 117/14/Poland/2014, Group C: animals inoculated with the very virulent reference strain 75/11/Poland/2011, and group D: control group).

Figure 6

Pathological lesions observed at necropsy in SPF chickens inoculated with the re-assortant strains (Bug/03/Poland/2003 and 117/14/Poland/2014). (A) Kidneys from a bird inoculated with Bug/03/Poland/2003 that died at 3 dpi. (B) Proventriculus from a bird that died at 5 dpi inoculated with 117/14/Poland/2014. (C–E) Bursae and spleens from birds euthanized at 10 dpi inoculated with Bug/03/Poland/2003 (C), 117/14/Poland/2014 (D), and from the normal control (E).

Figure 7

Histopathology of the bursa of Fabricius of SPF chickens. (A) The control group at 10 dpi (group D). (B) Group C inoculated with reference very virulent strain 75/11/Poland/2011 at 4 dpi. (C,D) Group A inoculated with Bug/03/Poland/2003 at 3 and 10 dpi, respectively. (E,F) Group B inoculated with 117/14/Poland/2014 at 3 and 10 dpi, respectively. The magnification is represented by the black band (short = ×50, long = ×100).

Figure 8

Shift in the dominant genotype strain of IBDV in the field. The evolutionary process of IBDV in the field, year of emergence, and genetic composition for each main lineage are shown. The illustration of quasispecies composition and the bottleneck of different severities induced by the vaccination are also denoted. Effective vaccination is shown in red, and unsuccessful vaccination is denoted in black. The effect of the bottleneck on mutant spectra is shown by denoting relevant adaptive mutations, which are hypothetically located and highlighted with colored symbols.