Characterization of Symbiodinium-associated viruses and implications for coral health
Coral reefs are in decline worldwide. Much of this decline is attributable to mass coral bleaching events and disease outbreaks, both of which are linked to anthropogenic climate change. Despite increased research effort, much remains unknown about these phenomena, especially the causative agents of many coral diseases. In particular, coral-associated viruses have received little attention, and their potential roles in coral diseases are largely unknown. This study aimed to address this lack of information by characterising the viruses associated with reef-building corals and Symbiodinium (dinoflagellates that can form symbioses with corals). Symbiodinium virus screening experiments revealed the presence of numerous and varied virus-like particles (VLPs) inside cells. Of the 49 Symbiodinium cultures screened, approximately one third contained putative latent viral infections that could be induced to enter their lytic cycle by UV irradiation. Electron microscope examination revealed VLPs closely resembling viruses previously found in dinoflagellates and other microalgae. Three cultures that showed evidence of latent viral infections were chosen for whole transcriptome sequencing, which revealed the presence of viral genes that were expressed in several different types of Symbiodinium. The relationship between the detected genes and known viral gene sequences suggested that the cells were infected with double-stranded DNA (dsDNA) viruses. In order to determine how the host cell responds to stress-induced viral infection, the expression levels of genes associated with stress response and viral infection were measured. The expression levels of many genes were unchanged following UV stress, and expression of genes that were predicted to be upregulated following stress, such as those encoding antioxidant enzymes, in fact showed lower expression levels. Despite this, several groups of genes involved in viral infection and host cell response were upregulated following stress, providing further evidence for stress-induced latent or chronic viral infections. In addition to the research carried out on Symbiodinium cell cultures, viruses associated with three coral diseases were studied using electron microscopy. Virus-like particles were present in coral and Symbiodinium cells from all three diseases, but viral abundance was correlated with disease state in only one: white patch syndrome (WPS) of Porites australiensis. The locations and morphologies of the VLPs associated with WPS suggested the presence of dsDNA and single-stranded RNA (ssRNA) viruses infecting both the coral animal and Symbiodinium cells. DNA sequences obtained from WPS-affected corals matched closely with sequences obtained from VLP-containing Symbiodinium cells. Based on the evidence gathered from Symbiodinium cell cultures and coral tissues, I propose a theoretical model of viral infection in WPS. In this model, the coral animal cells are routinely subject to chronic viral infections, and Symbiodinium cells harbour two types of chronic or latent infections – a dsDNA and an ssRNA virus – that can be induced via stress, resulting in cell lysis or loss of the cells from the coral host. In addition to detection and rudimentary identification of viruses infecting Symbiodinium cells, this study generated the largest dinoflagellate transcriptomic dataset to date. These data will prove valuable for future research into Symbiodinium, both in terms of viral infections and more generally.