Host-Symbiont Biomass Regulation in the Cnidarian-Dinoflagellate Symbiosis

In the cnidarian-dinoflagellate symbiosis, the regulation of host-symbiont biomass is essential for maintaining symbiosis stability and preventing cellular stress. Symbiont biomass is primarily thought to be regulated by pre-mitotic (cell-cycle arrest of the symbiont population) and post-mitotic (autophagy, apoptosis and expulsion) controls. However, there are still large knowledge gaps about the molecular events associated with this regulation, and how these events interact to generate observed patterns of host-symbiont specificity. I therefore aimed to: (1) characterise the molecular mechanisms underpinning the cell-cycle arrest of the symbiont population, and those co-ordinating host and symbiont growth in hospite; and (2) determine how the main regulatory mechanisms (symbiont cell-cycle arrest, host apoptosis, host autophagy and expulsion) interact during the onset, establishment and maintenance of symbioses between homologous (i.e. native) and heterologous (i.e. non-native) symbiont species.

In Chapter 2, I focused on the mechanism we know least about, cell-cycle arrest of the symbiont population. In particular, using bioinformatics, I identified which evolutionarily-conserved cell-cycle progression proteins (cyclins and cyclin-dependent kinases (CDKs)) are present in symbiotic dinoflagellates (family: Symbiodiniaceae), whether these proteins differ between species, and how the expression of Symbiodiniaceae cell-cycle genes is influenced by symbiotic state (i.e., when the dinoflagellates are in culture versus in the host). Cyclins and CDKs, that are related to eumetazoan cell-cycle and transcriptional cyclins and CDKs, were identified in Symbiodiniaceae, alongside several alveolate-specific cyclins and CDKs, and those related to protist and apicomplexan taxa. Alveolate-specific CDKB was proposed as a homolog to the main cell-cycle CDKs in Saccharomyces cerevisiae, Cdc28/Pho85, due to its phylogenetic position, conservation across Symbiodiniaceae species, and the presence of the canonical CDK motif. Symbiont species was found to influence the presence of CDK and cyclins with Cladocopium species and D. trenchii containing CDK and cyclins related to parasitic taxa, whilst a Symbiodinium species contained CDKs and cyclins that were all most closely related to the free-living dinoflagellate Amphidinium. Several alveolate-specific CDKs and two protist P/U-type cyclins exhibited altered expression when in symbiosis, suggesting that the symbiotic state influences the expression of symbiont cell-cycle genes. These findings help us to understand the molecular mechanisms that may underpin cell-cycle arrest of the symbiont population in hospite.

In Chapter 3, I focused on the co-ordination between symbiont and host biomass during symbiosis and how the presence of symbionts alters the expression of host cell-cycle genes in the symbiotic gastrodermis and the asymbiotic epidermis, using pre-existing transcriptomics data for the model sea anemone Exaiptasia pallida (‘Aiptasia’) in stable symbiosis with the dinoflagellate Breviolum minutum. The presence of symbionts in the gastrodermis elicited host cell-cycle arrest in the G1 phase and the inhibition of DNA synthesis and mitosis, compared with the aposymbiotic (i.e. temporarily symbiont-free) gastrodermis. As well as reducing cell-cycle progression, the presence of symbionts negatively impacted host apoptosis, with the symbiotic gastrodermis having elevated levels of host apoptotic inhibitors and depressed levels of host apoptotic sensitisers when compared with the aposymbiotic gastrodermis and the epidermis of symbiotic hosts, respectively. Also, I observed increased expression of genes associated with the persistence of non-pathogenic ‘non-self’ cells in symbiotic gastrodermal tissues, while genes associated with sensitivity to reactive oxygen species (ROS) were down-regulated. These events may contribute to the persistence of symbionts in the host gastrodermis. In epidermal cells, a single gene required for mitotic completion was up-regulated in symbiosis compared with aposymbiotic anemones, suggesting that the presence of symbionts in the gastrodermis stimulates mitotic completion in the epidermis, possibly through the nutritional benefits provided by the symbiosis. Microscopical analysis using the S phase indicator, EdU, confirmed that there were significantly more proliferating host cells in both the gastrodermis and epidermis in the symbiotic state compared with the aposymbiotic state, agreeing with the tissue-specific transcriptomic analysis. These findings help us to understand both how symbionts persist in a host and how symbionts stimulate the growth of the host during symbiosis on a molecular level.

In Chapter 4, I inoculated aposymbiotic Aiptasia with one of four different species of Symbiodiniaceae: homologous Breviolum minutum (ITS2 type B1), and heterologous Symbiodinium microadriaticum (A1), Cladocopium goreaui (C1) and Durusdinium trenchii (D1a). I then measured host apoptosis, expulsion and symbiont cell-cycle phase during the onset, establishment and maintenance of the symbiosis, and compared this with an unmanipulated symbiosis (i.e. permanently symbiotic Aiptasia). The relative importance of these mechanisms shifted over time, even though they all continued to play a role. In particular, after an early peak, host apoptosis declined, but symbiont expulsion compensated for this by becoming more dominant. However, as symbiosis reached a steady state, the number of symbionts arrested in the G1 phase of their cell cycle increased, while the number of cells cycling through their cell cycle decreased, emphasising that symbiont cell-cycle control is an important regulator of host-symbiont biomass in the stable symbiosis. Similar regulatory patterns were seen in permanently symbiotic anemones to those seen in the fully re-established symbiosis, except that permanently symbiotic hosts expelled a smaller proportion of their symbionts. Species-specific differences were apparent, however, especially with respect to rates of host apoptosis and expulsion. For instance, D. trenchii-colonised anemones showed the earliest decline in host apoptosis, and anemones inoculated with heterologous symbiont species consistently expelled a higher proportion of their symbiont population than those colonised by the homologous symbiont. Moreover, in the fully-established symbiosis D. trenchii had the highest proportion of its population arrested in the G1 phase.

Overall, my results provide a detailed overview of the cell-cycle machinery in the Symbiodiniaceae, and highlight that symbiotic state alters the expression of both host and symbiont cell-cycle genes, so allowing growth to be co-ordinated and the symbiosis to persist. I also show that a range of regulatory mechanisms influence symbiont population density, and shift in their relative importance between the onset and full establishment of the symbiosis. Symbiont identity influences the extent to which these mechanisms are used, though not the general patterns seen over time. These findings help us to better understand the cellular events that underlie a successful symbiosis and patterns of host-symbiont specificity, with implications for the formation and persistence of novel, potentially more thermally tolerant, host-symbiont pairings in the face of climate change.