Emergence of human G2P[4] rotaviruses containing animal derived gene segments in the post-vaccine era

Abstract

The introduction of Rotarix into the Belgian immunization program in June 2006 coincided with an increase of the relative prevalence of G2P[4] strains. However, the genetic composition of these persistent G2P[4] strains has not been investigated. Therefore, we have investigated the NSP4 gene of 89 Belgian G2P[4] strains detected between 1999 and 2013, covering both pre- and post-vaccination periods. The NSP4 genes were divided over seven separate clusters of which six were more closely related to animal than to human strains. The NSP4 genes that clustered more closely to animal DS-1-like strains were isolated after 2004-2005 and were found throughout multiple seasons. Complete genome sequencing of 28 strains identified several other gene segments that clustered more closely to animal than to human DS-1-like strains. These findings suggest that frequent interspecies reassortments may have played a role in the spread of G2P[4] rotaviruses in the post-vaccination period in Belgium.

Figure 1. Maximum likelihood tree of the NSP4 gene.

Belgian G2P[4] rotaviruses were found in seven distinct clusters (one human cluster (red) and six animal clusters (shades of blue)). Strains analyzed in this study are indicated with a filled circle. The RotaTeq vaccine NSP4 is indicated by the black triangle. Selected strains for complete genome sequencing are highlighted in bold face. Only bootstrap values above 70% are shown.

Figure 2. Temporal distribution of 89 Belgian NSP4 genes which are most closely related to typical human NSP4 genes (red) or to bovine and/or bovine-like human NSP4 genes (blue).

Figure 3. Genotype constellations of 28 Belgian G2P[4] rotaviruses.

Human DS-1-like gene segments are indicated in red, whereas DS-1-like gene segments more closely related to bovine strains are indicated in different shades of blue. AU-1-like gene segments are indicated in orange.

Figure 4. Maximum likelihood tree of VP7 and VP4.

Belgian G2P[4] rotaviruses are indicated with a filled red circle. The RotaTeq vaccine strain is indicated by a black triangle. Only bootstrap values above 70% are shown.

Figure 5. Maximum likelihood tree of NSP1 and NSP3.

Belgian G2P[4] rotaviruses are indicated with a filled red circle. Only bootstrap values above 70% are shown.

Figure 6. Maximum likelihood tree of NSP5 and VP6.

Belgian G2P[4] rotaviruses are indicated with a filled circle and colored red when clustering most closely to human DS-1-like strains and colored in different shades of blue when clustering most closely to bovine or bovine-like human rotaviruses. Orange circles indicate AU1-like genotypes. Bovine or bovine-like human reference strains are indicated with an open circle in different shades of blue. The RotaTeq vaccine strain is indicated by a black triangle. Only bootstrap values above 70% are shown.

Figure 7. Maximum likelihood tree of VP1 and VP2.

Belgian G2P[4] rotaviruses are indicated with a filled circle and colored red when clustering most closely to human DS-1-like strains and colored in different shades of blue when clustering most closely to bovine or bovine-like human rotaviruses. Bovine or bovine-like human reference strains are indicated with an open circle in different shades of blue. The RotaTeq vaccine strain is indicated by a black triangle. Only bootstrap values above 70% are shown.

Figure 8. Maximum likelihood tree of NSP2 and VP3.

Belgian G2P[4] rotaviruses are indicated with a filled circle and colored red when clustering most closely to human DS-1-like strains and colored in different shades of blue when clustering most closely to bovine or bovine-like human rotaviruses. Bovine or bovine-like human reference strains are indicated with an open circle in different shades of blue. The RotaTeq vaccine strain is indicated by a black triangle. Only bootstrap values above 70% are shown.