Abstract
Dominant coral-associated Endozoicomonas bacteria species have been long hypothesized to play a role in the global sulfur cycle by metabolizing Dimethylsulfoniopropionate (DMSP) into Dimethylsulfide (DMS), which releases sulfur into the atmosphere; however, no sequenced genome to date harbors genes for this process. We assembled high-quality (>95 % complete) genomes of two new strains (Acr-1 and Acr-5) of a recently added species Endozoicomonas acroporae isolated from the coral Acropora muricata. We identified and functionally characterized the first DMSP lyase–a dddD gene homolog found in all E. acroporae, capable of metabolizing DMSP into DMS via the DddD cleavage pathway–using qPCR and GC-MS. Comparative genomic analysis identified that Endozoicomonas has high genomic divergence and a high percentage oxidative stress responsive genes and that E. acroporae strains have the highest number of Type III secretion system genes (T3SS) in the genus. This study confirms the role of the genus Endozoicomonas in the global sulfur cycle.
Introduction
The genus Acropora contains some of the most abundant reef-building corals in the Indo-Pacific [1], and these corals are also some of the most significant producers of dimethylsulphoniopropionate (DMSP) [2,3]. DMSP is present in coral tissue, mucus and symbiotic algae [4]. It is the central molecule in the marine sulfur cycle and precursor to dimethylsulphide (DMS), a climate-active gas [5,6]. DMSP is hypothesized to be part of the coral holobiont antioxidant system [7] and it act as an osmoprotectant against salinity fluctuations [3]. DMSP also acts as a signal molecule that attracts specific bacterial groups, which can form coral holobionts and underpin coral health [8].
Coral-associated bacteria use DMSP produced by corals and their symbiotic algae as a reduced sulfur and carbon source [8, 9]; they can also metabolize it into DMS [5,6]. DMSP degradation by marine organisms takes place via two pathways, the cleavage pathway and the demethylation pathway [9, 10]. Raina et al. [11] recently reported that the majority of DMSP-degrading bacteria belong to class Gammaproteobacteria, which includes Alteromonas-, Arhodomonas-, Idiomarina-, Pseudomonas- and Spongiobacter-related organisms. Of these, Arhodomonas-, Pseudomonas-, and Roseobacter-related species harbor a DMSP lyase–i.e. the dddD gene, first identified in Marinomonas sp. for degrading DMSP [12]. Endozoicomonas species, which are predominantly associated with keeping their coral host healthy [13], have been hypothesized to play role in the global sulfur cycle by effectively metabolizing DMSP into DMS [14,15]. However, no previous study has confirmed the genus’ role.
Material and Methods
We de-novo assembled high quality (>95 % complete) genomes of two new strains (Acr-1 and Acr-5) of a recently added species Endozoicomonas acroporae isolated from the coral Acropora muricata and identified for the first time a dddD gene homolog capable of metabolizing DMSP into DMS via the DddD cleavage pathway in all the E. acroporae strains. Furthermore, we functionally characterized the expression of the DddD gene and quantified the amount of DMS released using qPCR and Gas chromatography-mass spectrometry (GC-MS). Comparative genomic analysis of genus Endozoicomonas was performed to ascertain its genomic characteristics and features. We also profiled the abundance of E. acroporae species in Penghu, Taiwan and the Red Sea, Saudi Arabia (for details see supplementary data).
Results and Discussion
We assembled high quality genomes (>95 % complete) of the two E. acroporae strains and also used the previously assembled type strain E. acroporae Acr-14T [16,17] (Table S1, Fig S1). E. acroporae species are dominant coral-associated bacteria in the Red Sea, Saudi Arabia (Fig S2A, B) and Penghu, Taiwan (Fig S2C, D). All three strains of E. acroporae have a dddD gene homolog that encodes a DMSP lyase. qPCR analysis of the dddD gene from E. acroporae Acr-14T cultured in 1mM DMSP resulted in 42.77, 56.52, and 91.37 times higher expression than samples cultured without DMSP after 16, 24 and 48hrs, respectively (Fig 1A). The amount of DMS released when the culture (E. acroporae Acr-14T) was incubated in a DMSP-rich environment (Treatment A) was significantly higher (t-test, p-value <0.05) than controls (Fig 1B). The temporal increase in the concentration of released DMS confirms that E. acroporae can metabolize DMSP into DMS. The discovery of the dddD gene in Endozoicomonas provides new insights into the evolution of the DMSP cleavage pathway and further confirms the hypothesis that Endozoicomonas plays a role in the global sulfur cycle.
Comparative genomic analysis identified high genomic divergence using Amino-Acid Identity (AAI), Average Nucleotide Identity (ANI) and DNA-DNA Hybridization (DDH) (Fig 2 A, B, and C) in the genus and also a reduced core genome (308 genes) (Fig S3). Genomes of Endozoicomonas species are large (5.43 ~ 6.69 Mb) (Table S2) and encode genes for all essential amino-acids [18], giving clues about not predominant genome streamlining as identified in symbiotic bacteria [19] and other symbiotic life stages [20]. Moreover, E. acroporae species have the highest numbers of T3SS genes in Endozoicomonas (data not shown), suggesting an intricate relationship with their host. Moreover, E. acroporae strains have different IS elements than E. montiporae, hinting that the two coral isolates have different evolution histories (Fig S4). Moreover, diverse phage insertions in Endozoicomonas species genomes suggest different infection histories (Table S3). In addition, E. montiporae and E. acroporae do not share any branches, according to core-genome based phylogenetic analysis; instead, their strains cluster tightly within their clades (Fig 2D). These results indicate that host and Endozoicomonas species have a complex nature of co-diversification. All species in this genus have a high percentage of oxidative stress responsive genes, which might be attributed to resistance against low oxygen environment in the ocean as well as highlight the marine adaptation of genus Endozoicomonas (Fig S5).
Conclusion
This study identified and functionally characterized DMSP lyase–a dddD gene homolog– in E. acroporae and, in doing so, confirms the role of this coral dominant bacterium in the global sulfur cycle. We also report two high quality genomes for new strains of E. acroporae and performed an up-to-date comparative genomic analysis on this genus. We identified a high genomic divergence and high percentage of oxidative stress response genes in all the species of this genus. We compared two coral host-specific Endozoicomonas and report a diverse array of IS elements in the genomes, giving clues about genome plasticity.
Data Availability
E. acroporae Acr-1 and Acr-5 assembled draft genomes are submitted to GenBank under accession numbers SAUT00000000 and SAUU00000000, respectively.
Author Contributions
K.T and S.L.T conceived the idea of this study. K.T assembled the genomes, performed bioinformatics analysis and wrote the manuscript. P.W.C cultured the strains and performed qPCR analysis. C.Y.L and Y.F.C performed GC-MS experiments and analysis. S.H.Y and N.H.W helped write the manuscript. P.Y.C, H.Y.C, and M.S.C helped in GC-MS experiments and provided the instruments for conducting the experiment. S.L.T supervised the overall study and modified the manuscript.
Acknowledgements
This study was supported by funding from Academia Sinica. KT would like to acknowledge the Taiwan International Graduate Program (TIGP) for its fellowship towards his graduate studies.