Rhabdonatronobacter sediminivivens gen. nov., sp. nov. isolated from the sediment of Hutong Qagan Soda Lake

A novel Gram-stain-negative bacterium, designated as IM2376T, was isolated from the sediment of Hutong Qagan Lake in the Ordos, Inner Mongolia Autonomous Region of China. Phylogenetic analysis based on 16S rRNA gene sequence revealed that the strain IM2376T had the highest similarity with Roseinatronobacter thiooxidans DSM 13087T (96.2%) and Rhodobaca bogoriensis LBB1T (96.2%) of the family Rhodobacteraceae. Genomic relatedness analyses showed that strain IM2376T was clearly distinguished from other species in the family Rhodobacteraceae, with average nucleotide identities, average amino acid identities, and in silico DNA–DNA hybridization values not more than 74.1, 68.5, and 20.2%, respectively. The fatty acids were mainly composed of C18:1ω7c (64.9%), iso-C16:0 (16.3%), and C16: 1ω7c/C16:1ω6c (6.0%). The major polar lipids were diphosphatidylglycerol, phosphatidylglycerol, and phosphatidylcholine. The predominant ubiquinone was Q-10 (94.9%). The genomic DNA G + C content was 66 mol%. Based on all these results, strain IM2376T was considered a novel species of a new genus in the family Rhodobacteraceae, for which the name Rhabdonatronobacter sediminivivens gen. nov., sp. nov. is proposed. The type strain of Rhabdonatronobacter sediminivivens is IM2376T (= CGMCC 1.17852T = KCTC 92134T).


Introduction
The family Rhodobacteraceae was first established by Garrity et al. (2005). Till date, approximately 200 genera are validly published and correctly named under this family (https:// lpsn. dsmz. de/ family/ Rhodo bacte raceae). Rhodobacteraceae are mainly of aquatic origin and are commonly found in marine environments (Pujalte et al. 2014). Some species of this family can synthesize bioplastic materials polyhydroxyalkanoates (PHAs) or more commonly polyhydroxybutyrate (PHB) as carbon reserve material (Boldareva et al. 2007;Hwang and Cho 2008;Li et al. 2017;Pujalte et al. 2014). Soda lakes could be a favorable environment for the isolation of strains of this family. Soda lake has high salinity and pH characteristics caused by the accumulation of sodium (bi)carbonate owing to evaporation (Jones and Grant 2000). Since soda lake is rich in hydrochloride and bicarbonate but poor in phosphorus, this environment is considered conducive for the accumulation of PHA or PHB (Chen et al. 2019). Besides, bacteria in this family often contain sulfide-quinone oxidoreductase (Ssqr) cytochrome subunit of sulfide dehydrogenase (fccA), or sulfide dehydrogenase [flavocytochrome c] flavoprotein chain (fccB) genes. These genes' products are helpful in sulfur-containing wastewater or waste gas treatment (Yu et al. 2011). Isolation and identification of strains of this family from soda lakes would help in enriching the microbial resources and provide newfound assets for industrial development.
The sediment sample was diluted and spread on the surface of the LN agar plates (with 1.5% agar) and then incubated at 37 ℃ for about 2 weeks until the colonies were observed. Subsequently, the colonies were sub-cultured thrice on LN agar plates with the dilution separation method to obtain the pure culture. The isolated strain IM2376 T was stored in 20% (v/v) glycerol with 10% NaCl at -80 ℃.

Phylogenetic and phylogenomic analyses
The genomic DNA of strain IM2376 T was extracted according to the method described by Marmur (1961) and Xu et al. (2007). The genome was sequenced using the Illumina Novaseq 6000 platform by the BioMarker technologies, PR China. The genome sequences were assembled using SPAdes 3.13.0 with default parameters (Nurk et al. 2013). The draft genome was annotated by the NCBI Prokaryotic Genome Annotation Pipeline (PGAP) version 4.12 (https:// www. ncbi. nlm. nih. gov/ genome/ annot ation_ prok/). The 16S rRNA gene sequence similarity of the strain IM2376 T was carried out through the BLAST program (http:// blast. ncbi. nlm. nih. gov/ Blast. cgi) (Johnson et al. 2008) and the EzBio-Cloud webserver (Yoon et al. 2017a). The 120 conserved single-copy genes in the IM2376 T genome were obtained by employing the Genome Taxonomy Database (GTDB) (Chaumeil et al. 2019;Parks et al. 2020Parks et al. , 2018. Multiple sequence alignments were performed using the CLUSTAL W program (Larkin et al. 2007). Furthermore, the phylogenetic tree was constructed based on the 16S rRNA gene and the 120 conserved single-copy proteins in the MEGA X software (Sudhir et al. 2018) with the Maximum-likelihood (ML) (Felsenstein 1981), Neighbor-Joining (NJ) (Saitou 1987), and the Minimum Evolution (ME) (Felsenstein 1981) methods. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) were shown next to the branches (Felsenstein 1985). The G + C content was calculated from the whole-genome sequence. The OrthoANI algorithm was used to calculate the Average Nucleotide Identity (ANI) value (Yoon et al. 2017b). The genome sequence data were uploaded to the Type (Strain) Genome Server (TYGS), a free bioinformatics platform (https:// tygs. dsmz. de) for the genome-to-genome distance analysis in accordance with in silico DNA-DNA hybridization (isDDH) (Meier-Kolthoff and Göker 2019). The average amino acid identity (AAI) was also calculated using the AAI calculator (http:// enve-omics. ce. gatech. edu/ aai/ index) (Rodriguez-R and Konstantinidis 2014).
The following substances were used to test the utilization as sole carbon and energy source (2 g/L for sugars, alcohols, and organic acids while 1 g/L for amino acids) in LN medium with 0.1 g/L yeast extract (without the fish peptone, sodium formate, sodium acetate, and sodium pyruvate): galactose, starch, trehalose, mannitol, D-xylose, D-maltose, sucrose, D-glucose, D-mannose, L-rhamnose, lactose, arabinose, cellobiose, fructose, sorbose, glycerol, sorbitol, acetate, malate, pyruvate, DL-lactate, succinate, fumarate, citrate, ornithine, arginine, glutamate, glycine, histidine, cysteine, isoleucine, valine, lysine, and aspartate. Biochemical tests were performed in LN medium (containing 0.01 g/L yeast extract) according to the methods described by Xu et al. (2007) and Mata et al. (2002), and encompassed the activity of oxidase, catalase and urease, the reduction of nitrate and nitrite, the production of H 2 S and indole, as well as the O-nitrophenyl-β-d-galactopyranoside (OPNG) and Voges Proskauer (VP) test. The antibiotic sensitivity test was conducted on LN agar plate with the following antibiotic discs (µg per disc unless otherwise noted): Nitrofurantoin (300), Erythromycin (15)

Chemotaxonomy characteristics
The strain was cultivated in the LN medium for 2 days, and then the fatty acid extraction and quantification by Gas Chromatography-Mass Spectrometry (GC-MS) were performed by following the previously described method (Kuykendall et al. 1988). The frozen dry cells (about 200 mg) were used to extract isoprenoid quinones with chloroform/methanol (2:1, by vol.). The extracted isoprenoid quinones were measured by reversed-phase High-Performance Liquid Chromatography (HPLC) (Wu et al. 2009). The polar lipids analysis was done using one-and two-dimensional Thin-Layer Chromatography (TLC) by following the method described by Kamekura and Kates (1988).

Morphology, physiology, and biochemical analysis
The cells of strain IM2376 T were non-motile and had no flagellum (Fig. S1, Table 1). The strain IM2376 T was sensitive to Nitrofurantoin, Erythromycin, Bacitracin B, Rifampicin, Ciprofloxacin, Novobiocin, Neomycin, Norfloxacin, Cefoxitin, Tetracycline, Tobramycin, Amoxycillin, Cefotaxime, and Vancomycin, while resistant to Chloramphenicol, Penicillin G, and Streptomycin. Other features of strain IM2376 T are indicated in the species description. The detailed features comparison between strain IM2376 T and its close species are listed in Table 1.

Phylogenetic and phylogenomic analyses
The 16S rRNA gene (1503 bp) and whole-genome sequence of strain IM2376 T were obtained. The genome size was 4.063 Mb. The genomic DNA G + C content was 66 mol% which was higher than that of the reference species (59-62 mol%, Table 1). The strain IM2376 T had 1 16S rRNA gene, 1 23S rRNA gene, 3 5S rRNA genes, 41 tRNAs, and 3 ncRNAs. Three clustered regularly interspaced short palindromic repeat sequences (CRISPRs) arrays were predicted. A total of 3671 coding sequences (CDSs) were predicted in the whole genome.
The genomic DNA G + C content of the type strain is 66 mol% (calculated from the genome sequence).
The type strain IM2376 T (= CGMCC 1.17852 T = KCTC 92134 T ) was isolated from a soda lake.
The GenBank accession number for the 16S rRNA gene sequence is MW750412. The whole-genome has been deposited in GenBank under the accession number JACBXS000000000.