Evolutionary dynamics of human rotaviruses: balancing reassortment with preferred genome constellations

Abstract

Group A human rotaviruses (RVs) are a major cause of severe gastroenteritis in infants and young children. Yet, aside from the genes encoding serotype antigens (VP7; G-type and VP4; P-type), little is known about the genetic make-up of emerging and endemic human RV strains. To gain insight into the diversity and evolution of RVs circulating at a single location over a period of time, we sequenced the eleven-segmented, double-stranded RNA genomes of fifty-one G3P[8] strains collected from 1974 to 1991 at Children's Hospital National Medical Center, Washington, D. C. During this period, G1P[8] strains typically dominated, comprising on average 56% of RV infections each year in hospitalized children. A notable exception was in the 1976 and 1991 winter seasons when the incidence of G1P[8] infections decreased dramatically, a trend that correlated with a significant increase in G3P[8] infections. Our sequence analysis indicates that the 1976 season was characterized by the presence of several genetically distinct, co-circulating clades of G3P[8] viruses, which contained minor but significant differences in their encoded proteins. These 1976 lineages did not readily exchange gene segments with each other, but instead remained stable over the course of the season. In contrast, the 1991 season contained a single major clade, whose genome constellation was similar to one of the 1976 clades. The 1991 clade may have gained a fitness advantage after reassorting with as of yet unidentified RV strain(s). This study reveals for the first time that genetically distinct RV clades of the same G/P-type can co-circulate and cause disease. The findings from this study also suggest that, although gene segment exchange occurs, most reassortant strains are replaced over time by lineages with preferred genome constellations. Elucidation of the selective pressures that favor maintenance of RVs with certain sets of genes may be necessary to anticipate future vaccine needs.

Figure 1. Phylogenetic Relationships Among the Genomes of the G3P[8] RVs Sampled in Washington, DC during 1976–1991.

The maximum likelihood tree was constructed using the concatenated ORF nucleotide sequences for each isolate and is mid-point rooted for purposes of clarity. Horizontal branch lengths are drawn to scale, and bootstrap values are shown as percentages for key nodes. Isolates assigned into clades (1976 clades A, B, and C; 1991 major clade) are indicated by boxes.

Figure 2. Phylogenetic Relationships Among the Individual Genes of the G3P[8] RVs: VP7, VP4, VP1, and VP2.

The maximum likelihood trees were constructed using the ORF nucleotide sequences for each gene of each isolate and are mid-point rooted for purposes of clarity. All horizontal branch lengths are drawn to scale, and bootstrap values are shown per 1000 replicates for key nodes. Sequences that cluster together in all algorithms tested were assigned the same color (orange, green, red, cyan, or navy blue), and separate colors indicated divergent gene alleles.

Figure 3. Phylogenetic Relationships Among the Individual Genes of the G3P[8] RVs: VP3, VP6, NSP1, and NSP4.

Details are given in the legend for Figure 2. Horizontal branch lengths are drawn to scale, expect for those interrupted with hash bars, which have been truncated.

Figure 4. Phylogenetic Relationships Among the Individual Genes of the G3P[8] RVs: NSP2, NSP3, and NSP5.

Details are given in the legends for Figures 2 and 3.

Figure 5. Allele-based Genome Constellations of the G3P[8] RVs.

The schematic illustrates the color-coding of each gene allele for the G3P[8] isolates based on the phylogenies shown in Figures 2– 4. The sample name and year of isolation is listed to the left of the corresponding genome constellation. The protein encoded by each gene is listed at the top. Isolates assigned into clades (1976 clades A, B, and C; 1991 major clade) are indicated.

Figure 6. Surface-Exposed Allele-Specific Amino Acid Differences for VP7.

(A) Table of allele-specific differences located on the surface of RRV VP7. The VP7 type corresponds to the alleles defined in Figure 2 (orange, green, red, and cyan) or to reference G3 RV strains (RRV, Wi68, RV3 and P). The amino acid at each position is listed to the right of the VP7 type. Numbering is based on RRV VP7 (GenBank# AF295303). (B) Three-dimensional location of surface-exposed allele-specific differences. The left image shows the architecture of a RV virion (modified with permission from B.V.V. Prasad) and the positions of VP7 and VP4. The right image shows a surface representation of the VP7 trimer crystal structure (PDB# 3FMG). Residues comprising the putative neutralization domains of VP7 are listed in Table S3 and have been colored as follows: pink (7-1A), salmon (7-1B), and purple (7-2). Allele-specific differences are shown in yellow and are labeled for a single monomer of the trimer.

Figure 7. Surface-Exposed Allele-Specific Amino Acid Differences for VP8*.

(A) Table of allele-specific differences located on the surface of RRV VP8*. The VP8* type corresponds to the alleles defined in Figure 2 for VP4 (orange, green, red, and cyan), to the P[3] VP4 RRV, or to the P[8] RV strains Wi61, D and P. The amino acid at each position is listed to the right of the VP8* type. Numbering is based on RRV VP4 (GenBank# AY033150). (B) Three-dimensional location of surface-exposed allele-specific differences. The left image shows a surface representation of the VP4 crystal structure (PDB# 1KQR). A white box defines the position of VP8*. The right images show the VP8* core from two different viewpoints (front or back). The front viewpoint is rotated 90° to the right compared with the image in the white box. The back viewpoint is rotated 180° to the left compared with the front. Residues comprising the putative neutralization domains of VP8* are listed in Table S4 and have been colored as follows: pink (8-1), salmon (8-2), purple (8-3), and olive (8-4). Allele-specific differences are shown in yellow.

Figure 8. Surface-Exposed, Allele-Specific Amino Acid Differences for VP5*.

(A) Table of allele-specific differences located on the surface of RRV VP5*. The VP5* type corresponds to the alleles defined in Figure 2 for VP4 (orange, green, red, and cyan), to the P[3] VP7 RRV, or to the P[8] RV strains Wi61, D and P. The amino acid at each position is listed to the right of the VP5* type. Numbering is based on RRV VP4 (GenBank# AY033150). (B) Three-dimensional location of surface-exposed allele-specific differences. The right image shows a surface representation of the VP4 crystal structure (PDB# 1KQR). A white box defines the position of VP5*. The left images show the VP5* dimer from two different viewpoints (front or side) (PDB# 2B4H). The front viewpoint is rotated 90° to the left compared with the image in the white box. The side viewpoint is rotated 90° to the right compared with the front. Residues comprising the putative neutralization domains of VP5* are listed in Table S5 and have been colored as follows: purple (5-1), blue (5-2, 5-3, 5-4, and 5-5), and salmon (predicted based on alignments; align). Allele-specific differences are shown in yellow.