In recent years, many studies have shown an increase in coral associated microbial biodiversity in coral diseases. However, the variation of coral diazotroph, which plays an important role in the nitrogen cycle, is still unclear. To explore the changes of nitrogen fixing microbial communities, we analyzed the diversity of nitrogen-fixing bacteria associated with corals in healthy vs. diseased conditions, and corals from two different locations (Xisha and Sanya). The diversity of nitrogen-fixing bacteria associated with two coral species, Porites lutea and Pocillopora damicornis were assessed using cloning and sequencing of the nitrogenase gene nifH. Phylogenetic analysis indicated coral associated diazotrophs community structure varied in different coral species: Chlorobi dominated in Porites lutea, while Alphaproteobacteria and Gammaproteobacteria were the most abundant nitrogen-fixing bacteria in Pocillopora damicornis. The dominant group of nitrogen-fixing bacteria was Chlorobi in healthy Porites lutea, but Cyanobacteria predominated in diseased ones. Moreover, the dominant nitrogen-fixing bacteria Gammaproteobacteria was replaced by Alphaproteobacteria in coral Pocillopora damicornis, from open sea to the coast. The diseased and coastal corals contained higher diversity of nitrogen-fixing bacteria than the healthy and open ocean corals. In conclusion, the nitrogen-fixing bacteria community structure shifted in response to the variation of coral species, coral health status and surrounding environments.
Reef-building coral, a major productivity source of the coral reef ecosystem, have important ecological and esthetic value (Hatcher, 1988, 1990). In recent years, as a result of climate change and ocean acidification, there has been serious degeneration of coral reefs and an increase in the presence of diseased scleractinian corals (Guinotte et al., 2003; Hoegh-Guldberg et al., 2007; Munday et al., 2008). Many reports have revealed the disappearance of coral occuring all over the world (Scavia et al., 2002). With a significant contribution to CO2 fixation, coral reefs play an important role in carbon cycling and global climate change. Therefore, understanding the mechanism of coral disease and maintaining the surviving coral have been critical tasks for marine biologists (Rohwer et al., 2002).
Comprising dynamic and highly diverse microorganisms, coral can enrich nutrients from oligotrophic seawater and recycle nutrients to take full advantage of the symbiosis system, which enables coral reefs to be ecosystems with high primary productivity (NguyenTac-An et al., 2013). The types of symbiotic organisms vary, as do their functions. For instance, the nitrogen-fixing bacteria supply the coral holobiont with nitrogen in the organic nitrogen-lacking environment (Ceh et al., 2013; Vitousek et al., 2002). Based on 16S ribosomal RNA sequencing, the microorganism communities varied in different coral species, sampling sites, and health status. Inferred from the previous results, the microbial community composition adjusts to form new functional structures to adapt to different surroundings (Koren and Rosenberg, 2006; Vega Thurber et al., 2009). However, compared to the functional genes, the 16S ribosomal RNA gene provides limited phylogenetic resolution and less accuracy in studying specific functional microbial community (Yang et al., 2013). To further explore the roles of nitrogen-fixing bacteria in coral, more evidence about the relationship between nitrogen-fixing bacteria in coral and coral diseases is required. To understand how the environment affecting the structure of nitrogen-fixing bacterial community, we chose nifH gene as the molecular marker to investigate the composition of nitrogen-fixing bacteria associated with corals collected from different sites and in different health status.
In this study, to analyze the variation of coral associated diazotrophs communities in different environment, we proposed a comparative study on two distinctive sites, Xisha Shidao Island and Sanya Luhuitou Peninsula. The former is far away from the mainland and therefore with little disturbance of human activity, while the later is near the mainland and suffers more interference from human activity. The results provided insight into the potential relationship among coral nitrogen fixing bacteria, coral disease, and the environmental stress.
Materials and methods
Study sites and sample collection
Samples of Pocillopora damicornis were collected in the Hainan Sanya Luhuitou peninsula (SYZ) in November 2010 and near the Xisha Shidao Island (XSZ) in July 2011. Samples of Porites lutea, both healthy (PJ) and diseased (PJD), were collected at Hainan Sanya Luhuitou Peninsula in July 2013. The samples were directly transferred into sterilized sample-collecting bags surrounded by ice, and returned to the laboratory for further processing. After rinsing with sterilized seawater, the samples were chiseled off for further analysis.
DNA extraction and polymerase chain reaction
The coral tissue, covered by several layers of Cling film, was smashed into powder with a hammer. Total genomic DNA in the powder was extracted with a soil DNA isolation kit (Omega Biotek). DNA quantity and quality were analyzed on a NanoDrop 2000C spectrophotometer (Thermo Fisher Scientific, USA). To amplify the nitrogenase gene, one pair of nifH-specific primers (Olson et al., 1998) (forward nifHF 5′-ATG TCG GYT GYG AYC CSA ARG C-3′ and reverse nifHR 5′-ATG GTG TTG GCG GCR TAV AKS GCC ATC AT-3′) was used. The PCR reaction mixture consisted of 2.5 μl of 10×PCR buffer (TaKaRa), 200 μmol l−1 dNTP mixture (TaKaRa), each primer at 1 μmol l−1, 10 ng of template DNA, 1.0 U Taq DNA polymerase (TaKaRa), 4% BSA (Sigma), and H2O up to 25 μl. Amplification conditions for the PCR included an initial denaturing step of 95°C for 5 min, followed by 29 cycles of 94°C for 45 s, 55°C for 45 s, and 72°C for 60 s, and a final extension step of 72 °C for 10 min.
Library construction and sequencing
PCR products were checked for size and purity on 2% agarose gel. The target fragments were recovered from the gels with the agarose gel DNA recovery kit (Tiangen), ligated into the pMD18-T vector according to the manufacturer's instructions (TaKaRa), and transformed into competent E. coli DH5α. The positive clones obtained were identified by colony PCR with RV and M13 primers, and digested with restriction endonucleases TaqI and HaeIII. The reaction products were ran on a 3% agarose gel to obtain the distribution of restriction fragments. One unique distribution of restriction fragment length polymorphism (RFLP) was defined as an OTU (Operational Taxonomic Unit) (Tai et al., 2013). A representative clone of each OTU was sent for Sanger sequencing, which was carried out in BGI, Guangzhou. The sequencing data were submitted to the GenBank database under the following accession numbers: KC748139–KC748164, KC748188–KC748202 and KF854478–KF854577.
Phylogenetic trees were constructed using MEGA 6.0 software. Sequences closely related to the ones obtained in this study were retrieved from GenBank BLAST and included in the phylogeny. All library sequences and reference sequences were aligned using Clustal W. Phylogenetic trees were constructed using the neighbor-joining algorithms implemented in MEGA 6.0 (Tamura et al., 2013). We evaluated the robustness of the inferred tree topologies by 1,000 bootstrap replicates of the neighbor-joining data.
Bacterial source analysis
The sequences with the highest BLAST score were identified as the most similar sequences. The source information of the sequence was obtained from the item “source” and “host” in the record from GeneBank.
Ecological index analysis
The biodiversity of nitrogen-fixing bacteria was evaluated by the Shannon–Wiener index, the Simpson index, and evenness (Wang et al., 2013). The biodiversity indexes were calculated using these equations:
Diseased sample analysis
Healthy Porites lutea (PJ) had a relatively smooth surface without apophysis (Figure 1a). In contrast, the diseased sample (PJD) with an uneven surface was covered by many apophyses, with inhomogeneous light brown color (Figure 1b). The detailed picture showed the peak of the apophyses was bright white (Figure 1c). The diseased coral was identified as growth anomaly accompanied with white plague disease (Prakash and Taylor, 2012; Sheridan et al., 2013).
Biodiversity of coral-associated nitrogen-fixing bacteria
The biodiversity was evaluated by the Shannon–Wiener index, the Simpson index, and evenness criteria. The amount of OTUs varies greatly in the sample PJ, PJD, SYZ and XSZ. When comparing PJ to SYZ, two different species of healthy coral samples collected from the same site, the results showed that the nitrogen-fixing bacteria in different host species of coral varied greatly with regard to biodiversity. According to our results, all biodiversity indexes of the diseased Porites lutea coral samples were higher than that of the healthy ones (Table 1). This showed that as the coral became diseased, the biodiversity of nitrogen-fixing bacteria in the coral raised. Moreover, the biodiversity indexes of coral-associated nitrogen-fixing bacteria in the Pocillopora damicornis collected in Sanya Luhuitou were higher than those in Xisha Shi Dao.
We constructed two separate phylogenetic trees of nifH gene sequences from Porites lutea and Pocillopora damicornis. In Porites lutea, the nifH gene sequences were clustered into Chlorobi, Firmicutes, Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, Deltaproteobacteria, and Cyanobacteria in the phylogenetic tree (Figure 2). However in the phylogenetic tree of the nifH gene sequence from the Pocillopora damicornis, relevant bacteria were identified as Alphaproteobacteria, Gammaproteobacteria, Deltaproteobacteria, and Chlorobi (Figure 3). Judging by the number of OTUs, we determined that the dominant groups of nitrogen-fixing bacteria were Chlorobi in Porites lutea, but Alphaproteobacteria and Gammaproteobacteria in Pocillopora damicornis. This analysis revealed that different coral species hosted different population of nitrogen-fixing bacteria.
Compositions of nitrogen-fixing bacteria in coral samples
The dominant group of nitrogen-fixing bacteria was Chlorobi (89%) in the healthy Porites lutea, and Cyanobacteria (28%) in the diseased sample (Figure 4a). In the diseased coral, the proportion of the dominant group of nitrogen-fixing bacteria decreased, while the proportion of other non-dominant groups scaled up. However, in the healthy Porites lutea, with the existence of an absolutely dominant group, the non-dominant groups accounted for only about 10%. That means the taxonomic composition of nitrogen-fixing bacteria in the healthy Porites lutea is simpler, while that in the diseased one is more complex. Nitrogen-fixing bacteria from the coral was found 89% of the time in healthy coral samples, but only 28% of the time in diseased coral samples. In the diseased Porites lutea coral samples, the ratio of nitrogen-fixing bacteria from seawater, soil, mangroves, and beach sand increased (Figure 4b).
For the analysis of compositions of nitrogen-fixing bacteria from different sites, we conducted the same calculation used in the disease-associated analysis (Figures 4c and d). The taxonomic composition of nitrogen-fixing bacteria in coral Pocillopora damicornis from Sanya Luhuitou contained Alphaproteobacteria (46%), Gammaproteobacteria (38%), Chlorobi (7%), and Deltaproteobacteria (8%), while the nitrogen-fixing bacteria in coral from Xisha Shidao were mainly Gammaproteobacteria (70%), Alphaproteobacteria (27%), and Chlorobi (2%). The communities of nitrogen-fixing bacteria in Pocillopora damicornis collected in Xisha Shidao clustered into three phyla, with one dominant (Gammaproteobacteria), but in the sample collected in Sanya Luhuitou there were four phyla with two dominant (Alphaproteobacteria and Gammaproteobacteria). The taxonomic composition of nitrogen-fixing bacteria in Pocillopora damicornis collected in Sanya Luhuitou was more diverse than that in the sample from Xisha Shidao. The source composition of nitrogen-fixing bacteria in corals from Sanya Luhuitou was more complex than those from Xisha Shidao.
Coral disease with the rise of diversity of nitrogen-fixing bacteria
Coral associated microbes vary greatly with host species, local environmental factors and health status (Croquer et al., 2013; Roder et al., 2014a,b). Although it has been reported that the biodiversity of coral associated bacteria varies greatly among coral species, little is known on the change of nitrogen-fixing bacterial diversity. Previous studies revealed there are more diverse microbes in diseased coral (Croquer et al., 2013; Koren and Rosenberg, 2008; Meron et al., 2012; Roder et al., 2014a,b). It appears that the diversity of microbes is an indicator to monitor the coral's health status. Our results revealed that the diversity of coral associated nitrogen-fixing bacteria was higher in the diseased coral. The higher biodiversity of nitrogen-fixing bacteria in coral might be a survival strategy for the holobiont facing disease (Krediet et al., 2013). More details and further research are required to determine the role that nitrogen-fixing bacteria play in coral holobiont.
The invasion of exogenous microbes and coral diseasing
The rise of biodiversity in coral-associated nitrogen-fixing bacteria in diseased Porites lutea indicates an invasion of exogenous microbes and a change of endogenous microbial functional structure (Meron et al., 2012; Wegley et al., 2007). The roles of nitrogen-fixing bacteria in the host might be disturbed by the invasion of exogenous microbes and variations in the composition and functional structure of the nitrogen-fixing bacterial community (Kimes et al., 2010; Prakash and Taylor, 2012).
To gain more details on nitrogen-fixing bacterial composition, we constructed phylogenetic trees to analyze the taxonomic composition and its potential source. In Porites lutea, compared to healthy coral, one more phylum of nitrogen-fixing bacteria was detected in the diseased individual, and the dominant phylum changed from Chlorobi to Cyanobacteria. The complex taxonomic composition and the source composition of nitrogen-fixing bacteria were higher in the diseased coral. Based on the source composition, the higher diversity of nitrogen-fixing bacteria in the diseased coral is probably due to contamination with seawater, mangroves, and sandy beaches. This coincides with the invasion of exogenous nitrogen-fixing bacteria.
High nutrients: A potential cause of microbial composition in coral diseases
The results showed that the increase of biodiversity nitrogen-fixing bacteria in coral, the change of composition of nitrogen-fixing bacteria, and the invasion of exogenous nitrogen-fixing bacteria are closely related to the coral's health and habitats. Previous research showed that environmental stress drives the invasion of exogenous nitrogen-fixing bacteria and the increase in bacterial diversity, which might promote induction of coral diseases (Fabricius, 2005; Vega Thurber et al., 2009; Voss and Richardson, 2006). The pH and nutrient concentrations in seawater data from sample collecting sites was provided by the annual investigation of the Tropical Marine Biological Research Station in Hainan and the National Science and Technology Supporting Program (Table 2). This data clearly showed a much higher concentration of nutrients in seawater at Sanya Luhuitou than Xisha Shidao, which might be due to pollutants from human activity; for instance sewage emission and seawater pollution provide high concentrations of nutrients to seawater (Doney, 2010; Liu, P. et al., 2012; Liu, S. et al., 2005; Misic et al., 2011), and may cause changes in nitrogen-fixing bacteria, contributing to coral diseases.
In this study, we focused on the biodiversity and compositions of nitrogen-fixing bacteria in corals that differ in both health status and habitat. The biodiversity of nitrogen-fixing bacteria in coral from Sanya Luhuitou was higher than that from Xisha Shidao. Nitrogen-fixing bacteria in diseased coral also showed greater biodiversity. Complexities of both the taxonomic composition and the source composition of nitrogen-fixing bacteria were increased in the diseased coral. There is an adaption of the composition of nitrogen-fixing bacteria to different coral species, sites and coral health status. High amount of nutrients in the seawater might contribute to the invasion of exogenous nitrogen-fixing bacteria in coral and increase the coral disease occurrences. However, the detailed mechanisms remain unclear. Further research on the roles of coral associated nitrogen-fixing bacteria in coral diseases is needed.
The research was supported by the National High Technology Research and Development Program of China (863 Program, 2012AA092104, 2013AA092901 and 2013AA092902), Strategic Priority Research Program of the Chinese Academy of Sciences (XDA11020202), the National Natural Science Foundation of China (41276113; 41276114; 31270528; 41206082; 41006069), Science and technology cooperation projects of Sanya (2013YD74), and the Sanya Station Database and the Information System of CERN, the Knowledge Innovation Program of the Chinese Academy of Sciences (SQ201218), the Open Fund of Key Laboratory for Ecological Environment in Coastal Areas, State Oceanic Administration (201304), the Key Laboratory of Marine Ecology and Environmental Science and Engineering, SOA (MESE-2013-02) and the Key Laboratory for Ecological Environment in Coastal Areas, State Oceanic Administration (No. 201211).