Microbial community composition of the samples, including external and internal tissues, sponge cell- and prokaryotic cell-enriched samples. The red block represented the Gsub bacterium in the present study. The sequencing coverages of the contigs where the 16S rRNA genes located were taken as the coverages of the 16S rRNA genes and were considered their relative abundance. The lowest assigned taxonomic levels are shown. The phylum names and the class names for Proteobacteria are shown in parentheses.
Maximum likelihood (ML) unrooted tree based on the full-length 16S rRNA genes (A) and 31 conserved single-copy genes (B) showing the phylogenetic position of the Gsub bacterium and its close relatives. GenBank accession numbers of the reference sequences are shown in parentheses, and the Gsub bacterium is found to be closely related to the cluster of endosymbionts of the deep-sea clam, with the highest 16S rRNA gene identity of 94.5% to “Candidatus Ruthia magnifica” strain Cm. The black circles indicate the reference genomes involved in the subsequent genomic comparison. Bootstrap values of >70% are shown. The length of the branch in the horizontal dimension indicates the amount of change in the evolutionary lineage. The bars at the bottom of panels A and B represent 0.02 and 0.05 nucleotide substitutions per position. aff., affiliated; str., strain.
Comparison of functions between the Gsub bacterium and the reference genomes. (A) Venn diagram showing the distribution of Kyoto Encyclopedia of Genes and Genomes (KEGG) genes in the Gsub, Rmag, Voku, Tcru, and Tsul bacteria. (B) Venn diagram showing the distribution of KEGG genes in the Gsub, Glop, Ghal, Tcru, and Tsul bacteria. (C) Heatmap and clustering of the functional profile (first level of SEED subsystem) in the genomes of Gsub and reference bacteria. The genomes of Gsub, Rmag, and Voku contain lower numbers of genes for carbohydrate metabolism and respiration, cofactor synthesis, cell signaling and regulation, virulence factors, motility and chemotaxis, stress response, and membrane transport.
↵a The reference genomes include two endosymbionts from vent clams, “Candidatus Ruthia magnifica” strain Cm (Rmag) and “Candidatus Vesicomyosocius okutanii” strain (Voku), two extracellular SOB in sponges, the SOB in the deep-sea glass sponge Lophophysema (Glop) and in the shallow water sponge Haliclona (Ghal), and two free-living relative SOB, Thioalkalivibrio sulfidophilus HL-EbGr7 (Tsul) from shallow water and Thiomicrospira crunogena XCL-2 (Tcru) from the deep sea.
Comparison of detailed genomic features on the resistance to toxic compounds and heavy metals, oxidative stress, osmotic stress, and DNA repair in the bacterium Gsub genome and reference bacteria
Information on the unassembled and assembled metagenomes. The statistics of Illumina HiSeq and MiSeq paired-end reads of the sponge samples, the HiSeq reads of the water sample after quality control using the NGS QC toolkit, and the statistics of assembled contigs (>500 bp) are shown. Download TABLE S1, PDF file, 0.1 MB.
Microbial community composition of the surrounding water sample collected at the same time as sponge collection. The sequencing coverage of the contigs where the 16S rRNA genes located were taken as the coverage of the 16S rRNA genes and were considered their relative abundance. The lowest assigned taxonomic level according to Greengene is shown. Download FIG S1, PDF file, 0.1 MB.
Binning of the Gsub bacterium. (A to D) Binning by comparing the sequencing coverage of the contigs in the external and internal tissue samples. (A) Coverage of all contigs (the area indicates contig length, and the color indicates taxonomy at the phylum level; the axes are in log scale). (B) Contig group selected for the Gsub bacterium. Some additional contigs were included to test the clustering of tetranucleotide frequency (TNF) in panel D. (C) Correspondence analysis (CA) from CA1 to CA5. CA decomposes the chi-square statistic associated with the two-way table into orthogonal factors that maximize the separation between row and column scores. CA1 to CA5 are the 5 most important components of the CA analysis. (D) The CA clustering of CA1 and CA2, which further purifies the bacterium Gsub genome. (E to H) Binning of Gsub by comparing the sequencing coverage of the contigs in the sponge cell- and prokaryote cell-enriched samples. Download FIG S2, PDF file, 0.3 MB.
ML tree based on the full-length SoxB proteins. The SoxB protein sequence in Gsub showed the highest identity (87.5%) to the thioautotrophic gill symbiont of Bathymodiolus azoricus. Bootstrap values of >70% are shown. Download FIG S3, PDF file, 0.1 MB.
Fluorescence in situ hybridization (FISH) experiment on the sponge tissue using a Gsub-specific probe for the target sulfur-oxidizing bacterium (SOB). (A and B) Sponge cells (indicated by the arrow) and prokaryotic cells stained with DAPI which exhibit blue fluorescence (note that there were many prokaryotic cells with weak fluorescence). (C) Bacteria stained with the Cy3-labeled universal probe EUB338 (red fluorescence) corresponding to the field shown in panel A. (D) Target Gsub stained with Cy3-labeled Gsub-specific probe corresponding to the field shown in panel B. (E and F) Bright-field microscopy views of panels A and B, respectively. Download FIG S4, PDF file, 0.2 MB.
Metabolic pathway reconstruction for the Gsub bacterium. (A) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation showed that the Gsub bacterium contained a complete set of genes involved in reverse dissimilatory sulfate reduction (dsrAB, aprAB, and sat) and genes involved in the sulfur oxidation system (soxABXYZ, but not soxCD). (B) Calvin-Benson cycle. (C and D) Gsub lacks flagellum genes (C) and chemotaxis genes (D), which is consistent with the two thioautotrophic endosymbionts (Rmag and Voku) of vent clams. The two free-living SOB (Tsul and Tcru) have nearly complete sets of flagellum and chemotaxis genes, and the other genomes lack them. The seven units of the rectangles indicated Gsub, Rmag, Voku, Glop, Ghal, Tsul, and Tcru, respectively. Download FIG S5, PDF file, 0.1 MB.
The 39 unique KEGGs for Gsub compared with Rmag, Voku, Tcru, and Tsul, as revealed by the Venn diagram in Fig. 3A, and the 34 unique KEGGs for Gsub compared with Glop, Ghal, Tcru, and Tsul, as revealed by the Venn diagram in Fig. 3B. Download TABLE S3, PDF file, 0.1 MB.
Phylogenetic analysis and metabolic pathway for the Nsub bacterium. (A) ML unrooted tree based on the 16S rRNA genes showing the phylogenetic positions of Nsub and its close relatives. GenBank accession numbers of the reference sequences are shown. The 16S rRNA gene of Nsub displays the highest identity of 99% to an uncultured marine archaeal group 1 crenarchaeote from the brine-seawater interface of the Shaban Deep of the Red Sea. Bootstrap values of >70 are shown. (B) KEGG pathway annotation showed that the Nsub bacterium contained a nearly complete set of genes involved in carbon fixation in the hydroxypropionate-hydroxybutyrate cycle. Download FIG S6, PDF file, 0.2 MB.