Abstract

Herpesviruses are a notable evolutionary success, infecting a wide variety of vertebrate hosts including mammals, birds, reptiles, amphibians, and fish, as well as a single invertebrate host (oyster). Placement of viruses in the family Herpesviridae is based upon the presence of a linear, double-stranded DNA genome packaged within an icosahedral capsid that is surrounded by a proteinaceous tegument layer and finally a host-derived envelope.

The family Herpesviridae is comprised of three subfamilies: Alphaherpesvirinae (also containing bird viruses), Betaherpesvirinae, and Gammaherpesvirinae. Reptilian herpesviruses fall phylogenetically into the Alphaherpesvirinae. The taxonomical structure of these subfamilies is well supported by biological and molecular data, but the situation is less clear for herpesviruses from fish, amphibians, and oyster, which are classified as members of the family outside the subfamilies. The available data indicate that fish and amphibian viruses share genetic similarities, but are related only tenuously to the mammalian, avian and reptilian viruses. Similarly, oyster herpesvirus is related marginally to the vertebrate viruses. These observations promote a debate on the taxonomic position of the less well characterized viruses. For example, understanding whether agents such as koi herpesvirus (KHV, CyHV-3) are indeed herpesviruses, and thus generate stereotypical latent infections, is crucial to their containment as well as the development of therapies. To improve our understanding of the relationships of the fish and amphibian herpesviruses, we generated a molecular dataset for the two known amphibian herpesviruses that infect ranid frogs and herpesviruses from acipenserid, cyprinid, anguillid, percid, salmonid, and ictalurid fishes.

Using two degenerate primer sets, we amplified, by polymerase chain reaction (PCR), a 500 bp region of the DNA polymerase gene and a 340 bp region of the terminase gene. Amplicons were visualized by agarose gel electrophoresis, cloned and sequenced. Phylogenetic relationships were reconstructed with PAUP* 4.0 using both discrete and distance methodologies.

Analysis of both genes indicated a close relationship between the fish and amphibian herpesviruses. The terminase dataset supported the current three subfamily scheme, as well as a second lineage composed of the fish and amphibian herpesviruses, and a third monotypic clade containing the invertebrate oyster herpesvirus. This analysis along with the remarkable similarity in capsid architecture exhibited by viruses in this group supports their taxonomic cohesion. However, to incorporate this phylogenetic information, it would be appropriate to segregate the current family Herpesviridae into three families: mammalian, avian and reptilian viruses into the first, fish and amphibian viruses into the second, and the oyster virus into the third. These families could be grouped by the erection of a new order (Herpesvirales).