Study Site and Sampling
The sponge individuals were collected at the low tide sites S1 (109°29′33″ E 18°15′37″ N), S2 (109°29′11″ E 18°15′37″ N), and S3 (109°26′18″ E 18°15′37″ N) from the intertidal zone of Hainan Island in the South China Sea at 8:00 – 10:00 am in July 6th, 2019. The seawater temperature was 27.2 – 27.4°C. The sponge is solitary individual of globular size in appearance, typically up to 4 – 6 cm diameter. In each site, triplicate samples of sponge tissue slice, ambient seawater (two liters each), and ambient sediment (0 – 2 cm, ~ 5 g) were separately collected (Fig. S1 A and B). Collections and pretreatments of sponge, seawater, and sediment samples referred to the reported strategy [19, 20, 43] before being transferred into the RNAprotect Bacteria Reagent (Qiagen, Hilden, Germany). Identification of the sponge species based on gross morphology and PCR confirmation of the mitochondrial coxl (cytochrome c oxidase subunit I) gene using the primer pair CinaF2/dgHCO2198 . The amplified coxl clone gene showed 99% nucleotide sequence identity to the reported C. australiensis voucher LB_815 mitochondrial cox1 gene (JX177880) and was submitted to GenBank with the accession number MT913441. The time between sample acquisition and fixation was no longer than 20 min to minimize RNA degradation . All the RNA protector-fixed samples were stored at -80°C before total RNA and DNA extraction within two weeks.
RNA, DNA Extraction and cDNA Synthesis
RNA protector-fixed C. australiensis, seawater, and sediment samples were ground in liquid nitrogen with a sterilized mortar and pestle. Both RNA and DNA were extracted from ground powders using the PrepRNA/DNA Mini Kit (Qiagen, Hilden, Germany) following the manufacturer’s instruction. RNA and DNA were separately extracted from each sample of C. australiensis, seawater, and sediment. RNase-free DNase I (Fermentas, Hanover, USA) was used to digest the residual genomic DNA at 37°C for 60 min. RNA quality and integrity were checked by gel electrophoresis and by examining the A260/A280 ratio (ranging from 1.97 to 2.02) using a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, USA). The final RNA concentration and purity were quantified using the Qubit system (Invitrogen, Darmstadt, Germany). First-strand cDNA synthesis was performed using the SuperScript FirstStrand Synthesis System (Invitrogen, Carlsbad, USA). Each reaction volume was10 μl containing 100 ng RNA, 0.5 μl random hexamer primer (50 ng μl-1), 5 μl cDNA Synthesis Mix, and proper RNase-free water. This reaction system was incubated at 25°C for 10 min and then 50°C for 50 min and terminated at 85°C for 5 min. All cDNA aliquots were stored at -80°C before PCR amplification.
Amplification and Sequencing
The AOA amoA, AAOB amoA, CoAOB amoA, and AnAOB hzsB gene fragments were amplified, respectively, using the cDNA and DNA templates with the primers listed in Table 1. PCR amplifications were performed in a total volume of 40 μl containing 2 μl cDNA or 2 ng DNA, 0.1 μM of each primer, and 20 μl TaqMasterMix (CoWin Biotech, Beijing, China) on a Thermocycler (Eppendorf, Hamburg, Germany) according to the following procedures: 95°C for 5 min; followed by 30 cycles at 95°C for 40 s, annealing (temperature referring to Table 1) for 30 s and 72°C for 30 s, and finally 72°C for 10 min. For negative control, a similar procedure was carried out using purified RNA to ensure that there was no genomic DNA contamination. PCR products originated from the triplicate samples of C. australiensis, seawater, or sediment from each sampling site were pooled to reduce potential amplification bias and maximize the transcript richness referring to the previous strategies . The presence and sizes of these amplification products were estimated by gel electrophoresis (1.5% agarose gel). Since this study focused on the transcriptional activity of the ammonia-oxidizing community, the performance of DNA-based PCR amplification was to only test the presence of the targeted genes in the investigated biotopes, whose PCR products were not sequenced . cDNA-based PCR products were gel-purified with MinElute Gel Extraction Kit (Qiagen), cloned with pUCm-T Vector Rapid Cloning Kit (Sangon Biotech, Shanghai), and transformed to the DH5α competent cells (Sangon Biotech) according to the standardized instructions. The positive clones were screened by ampicillin resistance and identified by PCR screening with vector-specific M13 primers. A variable number of clones (13 – 69) from each clone library were sequenced (Table S1) on an ABI 3100 capillary sequencer (Sangon Corp., Shanghai, China).
All the obtained nucleotide sequences were trimmed manually by using ClustalW implemented in MEGA Ⅹ with default settings . The trimmed sequences were performed BLAST searches against the NCBI Nucleotide database. A phylotype was defined by 3% dissimilarity threshold [47, 48] for amoA, or by 5% for hzsB , using the Mothur Version 1.44.1 package . Phylotype representative sequences were taxonomically classified using BLASTn against NCBI Nucleotide database. Rarefaction curves and Good’s coverage estimators were determined using the Mothur package  to estimate whether the sequencing depth is enough to cover most of the transcribed genes in each clone library.
One representative sequence from each phylotype and its closest sequence retrieved from the NCBI Nucleotide database were aligned using ClustalW implemented in the MEGA Ⅹ . Maximum-likelihood (ML) tree was constructed by using the MEGA Ⅹ with the Kimura-2 parameter model according to a published guideline . Bootstrap analysis was used to estimate the reliability of phylogenetic reconstructions (1000 replicates).
RT-qPCR assays were performed using an ABI 7500 Fast Real-time qPCR platform (Applied Biosystems, Foster, USA), following the reported strategy on sponges . Gene expression was tested using technical triplicates for each sample of C. australiensis, seawater, and sediment. PCR was performed in a total volume of 25 μl containing 12.5 μl of SYBR Premix Ex Taq™ II (Takara, Dalian, China), 1 μl of cDNA template (tenfold serial dilution), and 0.1 μM of each primer (Table 1). PCR thermocycling steps were set as follows: 95°C for 5 min and 40 cycles at 95°C for 45 s, annealing (temperature setting showed in Table 1) for 45 s, and 72°C for 45 s. For quantification, standard curves (log-linear R2 > 0.99, E = 92% - 110%) were generated using purified and quantified plasmids containing AOA amoA (sequence of the clone AOA-spg-1, GenBank ID MT925791), parAOB amoA (sequence of the clone parAOB-spg-1, GenBank ID MT925730), comAOB amoA (sequence of the clone comAOB-sed-1, GenBank ID MT925742), or AnAOB hzsB (sequence of the clone AnAOB-spg-1, GenBank ID MT925769) fragment in a dilution series that spanned from 101 to 107 gene copies per reaction. All standard dilutions were prepared in 10 ng μl−1 aqueous tRNA solution (Sigma-Aldrich, Steinheim, Germany). Plasmid DNA was extracted using the PurePlasmid 96 Kit (CoWin Biotech), and the plasmid concentration was measured using the Qubit system (Invitrogen). Since the sequences of the vector and PCR insert are known, copy numbers of transcribed amoA or hzsB genes were directly calculated according to the reported formula: copy numbers μl−1 = (A × 6.022 × 1023) × (660 × B)−1, where A is the plasmid concentration (g μl−1), B is the recombinant plasmid length (bp) containing the amoA or hzsB fragment, 6.022 × 1023 is the Avogadro’s number, and 660 is the average molecular weight of 1 bp . For negative control, a similar procedure was performed using purified RNA to ensure that there was no genomic DNA contamination. After the qPCR assay, the specificity of amplification was verified by the generation of melting curves (in steps of 0.5°C for 5 s, with temperatures ranging from 60 to 95°C) and the qPCR product size and specificity were checked by 2% agarose gel electrophoresis.
Data acquisition of the qPCR assay was performed using the 7500 System SDS Software Version 1.2 (Applied Biosystems). One-way analysis of variance (ANOVA) was performed to evaluate the abundance variations of the amoA or hzsB transcripts between C. australiensis, seawater, and sediment niches using the commands in SPSS 19.0. The thetaYC matrix distances based on the amoA or hzsB transcript sequences were calculated via Mothur commands and were visualized by the principal co-ordinate analysis (PCoA) in Canoco 5.0. Comparison between the amoA or hzsB-harboring communities from C. australiensis, seawater, and sediment niches were analyzed using the analysis of similarity statistics (ANOSIM) on thetaYC indices through Mothur. Statistical differences were determined at the level of α = 0.05.