Molecular Evolution of Human Adenovirus (HAdV) Species C

Abstract

Currently, 88 different Human Adenovirus (HAdV) types are grouped into seven HAdV species A to G. Most types (57) belong to species HAdV-D. Recombination between capsid genes (hexon, penton and fiber) is the main factor contributing to the diversity in species HAdV-D. Noteworthy, species HAdV-C contains so far only five types, although species HAdV-C is highly prevalent and clinically significant in immunosuppressed patients. Therefore, the evolution of species HAdV-C was studied by generating 51 complete genome sequences from circulating strains. Clustering of the whole genome HAdV-C sequences confirmed classical typing results (fifteen HAdV-C1, thirty HAdV-C2, four HAdV-C5, two HAdV-C6). However, two HAdV-C2 strains had a novel penton base sequence and thus were re-labeled as the novel type HAdV-C89. Fiber and early gene region 3 (E3) sequences clustered always with the corresponding prototype sequence but clustering of the E4 region indicated recombination events in 26 out of the 51 sequenced specimens. Recombination of the E1 gene region was detected in 16 circulating strains. As early gene region sequences are not considered in the type definition of HAdVs, evolution of HAdV-C remains on the subtype level. Nonetheless, recombination of the E1 and E4 gene regions may influence the virulence of HAdV-C strains.

Figure 1

Phylogenetic analysis of the nucleic acid sequences of the neutralization ε determinant loop 1 (A) and loop 2 (B) of the hexon. Genbank sequences of the five HAdV-C prototypes are highlighted by a black dot (labelling indicates accession number-species and type). The neighbor-joining tree was generated based on the Kimura two-parameter model with MEGA7. Bootstrap values < 80% are not robust and therefore not depicted. *Strains 29C2 and 47C2 were renamed as the novel type HAdV-C89.

Figure 2

Phylogenetic analysis of whole-genome sequences. Clustering of nucleic acid sequences of circulating strains with Genbank sequences of the five species HAdV-C prototypes (highlighted by a black dot, labelling indicates accession number-species and type). The neighbor-joining tree was generated based on the Kimura two-parameter model with MEGA7. Bootstrap values < 80% are not robust and therefore not depicted. *Strains 29C2 and 47C2 were renamed as the novel type HAdV-C89.

Figure 3

Phylogenetic analysis of the major capsid genes hexon (A) and fiber (B). Clustering of circulating strains and Genbank sequences of HAdV-C prototypes (highlighted by a black dot, labelling indicates accession number-species and type). The neighbor-joining tree was generated based on the Kimura two-parameter model with MEGA7. Bootstrap values < 80% are not robust and therefore not depicted. *Strains 29C2 and 47C2 were renamed as the novel type HAdV-C89.

Figure 4

Phylogenetic analysis of the major capsid gene penton base. Clustering of circulating strains and prototype sequences (highlighted by a black dot, labelling indicates accession number-species and type). Strains 29C2 and 47C2 were highlighted with a blue box and renamed as novel type HAdV-C89. The neighbor-joining tree was generated based on the Kimura two-parameter model with MEGA7. Bootstrap values < 80% are not robust and therefore not depicted.

Figure 5

Multiple amino acid alignments of the hypervariable loops of the penton base. (A) RGD loop sequences and (B) hypervariable region 1 sequences of all HAdV-C prototypes including the novel HAdV-C89.

Figure 6

Phylogenetic analysis of the early gene regions E1 (A) and E4 (B). Clustering of circulating strains and prototype sequences (highlighted by a black dot, labelling indicates accession number-species and type). The neighbor-joining tree was generated based on the Kimura two-parameter model with MEGA7. Bootstrap values < 80% are not robust and therefore not depicted. *Strains 29C2 and 47C2 were renamed as the novel type HAdV-C89.