The surrounding sea area of West Island is located in the nearshore of Sanya City, Hainan Province, China.

The surface water samples (under water 8 m) were collected using 3 l water sampler at five localities in the sea area surrounding West Island, Hainan Province, China in June 2021 (Fig. 1). The distance between the two localities is not less than 1 km. Three replicated 2 l water samples were collected at each locality (Fig. 2). The eDNA collection was performed using the filtration method with 0.45 µm MCE membranes (Pall Whatman, U.K.). Each filtration membrane was separately preserved in a 2 ml centrifuge tube and stored at -10°C until taken back to the laboratory. The negative control was made through filtering 300 ml of pure water. Sterile masks and latex gloves were always worn during sampling and changed between different samples to prevent cross-contamination. Both the sampling devices and the filtration tools were cleaned with 10% bleach containing sodium hypochlorite to prevent external DNA contamination.

Figure 1.  

Location of eDNA sampling sites.

Figure 2.  

Photos of the habitat of sea area surrounding West Island and fieldwork.

The eDNA was extracted using the DNeasy Blood and Tissue Kit (QIAGEN, Germany) and the following modifications were made compared to the spin column protocol: a) add 720 µl ATL Buffer and 80 µl Proteinase K into centrifuge tubes containing filtration membranes and incubated 3-4 hours at 56°C; b) add 800 µl AL Buffer and 800 µl 96% ethanol after incubation; c) elution was done in 2 × 50 μl TE Buffer for a final volume of 100 μl. For each round of extraction, an extraction blank sample was made for monitoring cross-contamination, with smaller volumes being added: 180 μl ATL Buffer, 20 μl Proteinase K, 200 μl AL Buffer and 200 μl 96% ethanol. The laboratory tables, aseptic operation tables and experimental equipment were regularly cleaned with 5% bleach and then 75% ethanol before DNA extraction. The extracted DNA samples were stored at -20°C.

PCR amplification was performed using “Tele02” primers developed by Taberlet et al. (2018)(Forward: AAACTCGTGCCAGCCACC; Reverse: GGGTATCTAATCCCAGTTTG). Both forward and reverse primers were tagged with oligos designed using OligoTag (Coissac 2012) consisting of three random nucleotides and each sample was given a specific tag (Table 1). A PCR reaction total was 25 µl volume, containing 12.5 µl of 2 × Pfu PCR MasterMix (TIANGEN, China), 0.5 µl of each primer (10 µM), 0.5 µl BSA (10 mg/ml), 1µl DMSO, 0.5 µl DNaseI (5 U/μl), 4 µl of extracted DNA (10 ng/µl) and 6 µl ddH₂O. The samples were incubated with the DNaseI for 15 min at 37°C and the DNaseI was then inactivated by incubation at 60°C for 15 min before the template DNA and the primers were added. The PCR reaction conditions were the following: pre-denaturation for 10 min at 95℃, followed by 49 cycles of denaturation (20 sec at 95°C), annealing (15 sec at 60°C) and elongation (15 sec at 72°C) and final elongation for 5 min at 72°C. PCR reactions of each sample were replicated four times. Additionally, two negative controls (PCR blank) and two positive controls (template DNA extracted from tissue of fish species Sinocyclocheilus anophthalmus) were in each PCR reaction, so as to monitor contamination and detect the accuracy of the target fragments, respectively. The PCR products were verified on 2% agarose gels stained with Super GelRed^TM. In addition, the PCR products were pooled in equal volumes and each PCR pool contained one replicate of every sample. PCR blank products were also pooled together as negative controls. The PCR pools were purified using a PCR Purification Kit (QIAGEN, Germany). The libraries were built and then sequenced on an Illumina Novaseq platform (Beijing Genomics Institution, China) using 150 bp paired-end sequencing.

The raw sequences obtained from the Illumina Novaseq platform were initially processed using QIIME 2 software (Bolyen et al. 2019). The paired-end sequences were demultiplexed, based on the tags and primers were removed from the ends. Then the reads were trimmed to 160-180 bp used CUTADAPT software (Martin 2011). Next, the data obtained by sequencing was analysed by splitting, splicing and filtering to obtain high-quality sequences that were clustered into molecular operational taxonomic units (MOTUs) using VSEARCH software with ≥ 97% similarity (Zhang et al. 2019). Finally, the sequences of MOTUs were aligned with NCBI-BLAST (https://www.ncbi.nlm.nih.gov/BLAST) under the default parameters settings for taxonomic annotation in October 2021 (Djurhuus et al. 2017). We used the following criteria for taxonomic assignments: a) if the query sequence matched one single locally-occurring species in the database with ≥ 99% identity, this species was assigned; b) if the query sequence matched more than one locally-occurring species in the database with ≥ 99% identity, the lowest taxonomic level (i.e. genus or family) that included all of these species was assigned; c) if the query sequence matched one non-local species in the database with ≥ 99% identity and this species belongs to the same genus with the known local species, this genus was assigned. The sequences with an assigned taxon of ‘NA’ or assigned to human, bird, mammal or amphibian were removed. The geographic distribution of each species was evaluated by searching the FishBase website (http://www.fishbase.org/search.php).

We surveyed five localities in the surrounding sea area of West Island (Fig. 1). The investigation involved nearly 1777 m² measured using ArcGIS 10.1 software.

18.230 and 18.239 Latitude; 109.347 and 109.369 Longitude.

In total, three orders, 16 families, 28 genera and 41 coral reef fish species were detected by using eDNA in the surrounding sea area of West Island.