Aquisphaera Insulae Sp. Nov., a New Member in Isosphaeraceae Isolated From the Floating Island of Loktak Lake and Emended Description of the Genus Aquisphaera


 Strain JC669T was isolated from a floating island of Loktak lake, Manipur, India and shared highest 16S rRNA gene sequence identity with Aquisphaera giovannonii OJF2T. The strain is an aerobe, Gram-stain-negative, yellowish orange coloured, non-motile, NaCl intolerant, spherical to oval shaped, grows in single or aggregates and produce structures which appear like fruiting bodies. Strain JC669T grows well up to pH 9.0, has MK6 as respiratory quinone, C18:1ω9c, C16:0 and C18:0 as major fatty acids and phosphatidylcholine, an unidentified amino lipid, an unidentified choline lipid (UCL) and six unidentified lipids (UL1,2,3,4,5,6) as polar lipids. The genome size of strain JC669T is 10.04 Mbp and genomic G+C content of 68.5 mol%. Based on phylogenetic, polyphasic including genomic analyses support strain JC669T as a novel species of the genus Aquisphaera, for which we propose the name Aquisphaera insulae sp. nov. Type strain is JC669T ( =KCTC 72672T = NBRC 114306T).


Introduction
The phylum Planctomycetes is one among the PVC superphylums ( There are six genera (Paludisphaera, Aquisphaera, Singulisphaera, Tundrisphaera, Isosphaera, and Tautonia) under the family Isosphaeraceae and many of these are monospeci c except for Singulisphaera and Tautonia (Kaushik et al. 2020). Isosphaeraceae members are mesophilic, non-motile and their cells divide by polar budding. The Genus Aquisphaera belongs to the family Isosphaeraceae and it was rst described by Bondoso et al. (2011). Genus Aquisphaera contains only one species with valid name; Aquisphaera giovannonii which was isolated from a fresh water aquarium's sediments. The genus Aquisphaera is characterized by diphosphatidylglycerol and phosphatidylcholine as major polar lipids with C 16:0 and C 18:1 ω9c as major fatty acids. A. giovannonii is an obligate aerobe, slow-growing, chemoheterotroph, spherical and yellowish-orange/pink pigmented bacterium reproduce through budding.
Loktak lake is the largest freshwater lake of northeast India, having a spanning area of about 289 km 2 .
Loktak lake is famous for its oating islands (Phumdis; Fig. 1), which are heterogeneous biomass of vegetation, soil and organic matter at various decomposition stages (Reddy et al. 2005). Phumdis constitute a dense rhizosphere extending up to the lake sediment and hence, serve as an ecological habitat for several groups of bacteria like Actinobacteria, Acidobacteria, Proteobacteria, Verrucomicrobia and Planctomycetes (Puranik et al. 2016). This lake is of ecological hotspot with remarkable ora and fauna diversity and declared as Ramsar site in 1990. Insight of the previous metagenome study (Puranik et al. 2016) and during our survey for planctomycetes from India (Kumar et al., 2020a,b,c;Kaushik et al., 2020; Kumar et al. 2021a,b), strain JC669 T was isolated from the rhizosphere soil of a plant from the oating island of the Loktak lake, whose 16S rRNA gene sequence identity was highest (96.8%) with A. giovannonii OJF2 T . We characterized strain JC669 T using a polyphasic approach and genomic information and compared the characters with the type species of the genus Aquisphaera, A. giovannonii .

Materials And Methods
Isolation, cultivation and preservation Rhizospheric soil of Zizania latifolia (commonly known as Manchurian wild rice) of the phumdis ( oating island) was collected from Loktak lake situated in the northeast state of Manipur (GPS: 24°30'21" N 93°47'43"), India. The lake at the time of sample collection had a pH of 7.0 and temperature of 18°C. Soil samples were subjected to enrichment and cultivation in a medium as described by (Kaushik et al. 2020).
The rhizospheric soil sample (100 mg) was mixed in a 20 ml transparent serum vials with 10 ml of media and sealed with butylated stopper and further incubated at 25°C for ve months. After ve months of incubation light yellowish orange colour colonies appeared at the bottom of the serum vial, further puri ed on the same solid media after subsequent sub-culturing. Light yellowish orange cultures were maintained on the agar plates by repeated streaking / preserved at 4ºC and this culture is designated as strain JC669 T .
Isolation of DNA and 16S rRNA gene sequence analysis The axenic culture was used to isolate genomic DNA using commercial DNA isolation it (Nucleo-pore gDNA Fungal Bacterial Mini Kit, from M/s. Genetix Biotech Asia Pvt. Ltd, India). Genomic DNA was used for genome sequencing and16S rRNA gene ampli cation as well. Planctomycetes speci c primers F40 (Köhler et al., 2008) combined with universal primer R1388 (Stackebrandt et al. 1993) are used to perform PCR for 16S rRNA gene ampli cation. Ampli ed PCR products were further puri ed and sequenced (16S rRNA gene ) by Agri Genome Pvt. Ltd. (Kochi, India).
Genomic and in-silico analysis Genome sequencing was outsourced to AgriGenome Pvt. Ltd, Kochi, India. Illumina HiseqX10 platform was used to generate whole genome sequencing (WGS) and sequence coverage of 100x was generated using the paired-end library. Unicycler assembler was used for De novo assembly (Wick et al. 2017) with default k-mer sizes and for further downstream analysis unicycler assembly was used. Contamination in the genome sequence was checked using ContEst service of EZBiocloud

Phylogenetic analysis
ContEst16S service of EZBiocloud was used to extract the 16S rRNA gene sequence (1514 nt ) of the strain JC669 T and NCBI-BLAST (Johnson et al. 2008) was used for the analysis of its sequence identity. The full length 16S rRNA gene sequences of all type species of the family Isosphaeraceae, including strain JC669 T were obtained from EZBioCloud. The sequences were aligned using MUSCLE implemented in MEGA7.0 (Kumar et al. 2016). In a pairwise deletion procedure, the distances were calculated by using the Kimura 2-parameter (Kimura 1980).To nally construct the phylogenetic tree based on 16S rRNA gene sequences, MEGA7 software was used using minimum evolution (ME), Neighbor-joining (NJ), and maximum likelihood (ML) methods having Bootstraps of 1000 replication (Felsenstein 1985). Ninety two core genes were used for the RAxML based phylogenomic tree (Na et al. 2018;Kumar et al. 2021) for which the genome sequences were obtained from publicly available genomes.

Physiological analysis
Ten ml of basal medium (Bondoso et al. 2011) in test tubes (25 x 250 mm) was used to test organic substrates and nitrogen sources utilization of strain JC669 T and A. giovannonii DSM 22561 T (= OJF2 T ) with slight modi cations. Yeast extract (0.05% w/v) was supplemented to the basal medium and (NH 4 ) 2 SO 4 (0.1% w/v) as nitrogen source. For the organic carbon substrate utilization, various organic carbon substrates (0.5% w/v) were tested for the growth of both strains. Similarly, for nitrogen source utilization, glucose was used as a carbon source and nitrogen sources (0.1% w/v) were tested for their utilization as described earlier by Kaushik et al. (2020). Vitamin B 12 requirement and nitrate reduction (Smibert et al. 1981) of strain JC669 T and A. giovannonii DSM 22561 T were tested in liquid media as described by Kaushik et al. (2020). Basal media containing glucose (0.5% w/v) and ammonium sulphate (0.1% w/v) as carbon and nitrogen sources, respectively was used to test NaCl and temperature tolerance of strain JC669 T and A. giovannonii DSM 22561 T . API ZYM kit (Biomerieux, France) was used to assay enzymatic activities of strain JC669 T and A. giovannonii DSM 22561 T using the manufacturer's protocol.

Chemotaxonomic characterization
Analysis of fatty acids of strain JC669 T and A. giovannonii DSM 22561 T was done from cells which were harvested by centrifugation (8,000g for 15 min at 4 ºC) when cell density reached 70% of the maximum optical density (100% = 0.9 OD 660 ). The methylated cellular fatty acids were analysed from the data generated by M/s Royal Research Labs, Secunderabad, India, who performed the analysis according to the Microbial Identi cation System's instructions (Microbial ID; MIDI 6.0 version; method, RTSBA6; Sasser, 1990). Polar lipids of strain JC669 T and A. giovannonii DSM 22561 T were extracted, separated and characterized as described previously by Kates (1972) and Oren et al. (1996). Quinones were extracted and analysed through HPLC as described earlier by Imhoff (1984). Polyamines for strain JC669 T and A.
giovannonii DSM 22561 T were identi ed using HPLC after a process of extraction which was previously described (Kumar et al. 2020b).

Microscopy
Carl Zeiss LSM880 microscope was used for observing cell shape, morphology, size and cell division, Philips XL3O was used for scanning electron microscopic studies and H-7500 Hitachi was used for transmission electron microscopic studies.

Results And Discussion
Accession numbers of nucleotide sequences and 16S rRNA gene BLAST search analysis The 16S rRNA gene sequence and whole-genome sequence of strain JC669 T with the accession number LR782133 and JAALJH000000000 are deposited with EMBL and NCBI, respectively. BLAST analysis of both amplicon-based 16S rRNA gene sequence and that extracted from the genome of strain JC669 T showed the 96.8 % identity with the only species of genus Aquisphaera; Aquisphaera giovannonii ( Fig. 2).

Phylogenetic inference
The high POCP and AAI values between strain JC669 T and A. giovannonii OJF2 T are 67.3% and 76.6%, respectively. This values ( for genera delineation and thus support strain JC669 T belong to the genus Aquisphaera. OrthoANI and dDDH value between JC669 T and A. giovannonii OJF2 T are 81.4% and 24.3 %, respectively which are below the recommended cut off values of 95-96% for ANI and 70% for dDDH for species delineation ; Meier et al. 2014) (Fig. 2). Thus, the strain JC669 T represents a novel species of the genus Aquisphaera.
Further, phylogenetic tree based on 16S rRNA gene sequences (Fig. 3) and RAxML tree based on constructed ninety two core genes (Fig. 4)  whereas genome of A. giovannonii OJF2 T contains a total of 7,692 genes of which 7,415 are protein coding, 73 tRNA genes, 9 rRNA operons and 171 genes are pseudogenes. G + C content of genome of strain JC669 T is 68.5 mol% which is about 3.2% lower than G + C content (70.8%) of A. giovannonii OJF2 T . The difference in G + C between the strains indicate that these two are distinct species since a 3% difference in the G + C content was recommended for species delineation.
PATRIC software was used to align the genome sequences to identify the multiple maximal matches and local collinear block (LCBs; www. patricbrc.org) of strain JC669 T and A. giovannonii OJF2 T . The results showed signi cant difference in the alignments of the LCBs in both strains (Fig. S1) which could be due to the shu ing or inversion of genes due to DNA rearrangement or recombination of homologous regions. Majority of proteins of the strain JC669 T share 60-80% similarity with those of A. giovannonii OJF2 T (Fig. S2). Both strain JC669 T and A. giovannonii OJF2 T share 20-30% similarity with other Isosphaeraceae family members (Fig. S2).

In-silico metabolic characterization
In-silico metabolic characterization shows that both strain JC669 T and A. giovannonii OJF2 T are able to synthesize isopentenyl pyrophosphate (IPP) through 2-C-methyl-D-erythritol 4-phosphate/1-deoxy-Dxylulose 5-phosphate (MEP) pathway. These ve carbon isoprene units is the precursor for quinones' and carotenoids biosynthesis (Eisenreich et al. 2004;Zhao et al. 2013). In the genomes of strain JC669 T and A. giovannonii OJF2 T , major genes encoding for carotenoid biosynthetic enzymes are present like lycopene β-cyclase (5.5.1.19), zeaxanthin epoxidase (1.14.15.21), β-carotene hydroxylase (1.14.13.129) and neoxanthin synthase (5.3.99.9) were present. Thus, both strains are putatively capable of synthesizing neoxanthin from lycopene via intermediates like γ-carotene, β-carotene, zeaxanthin and violaxanthin (Fig. S3). The annotated genome of the strain JC669 T showed that it has capability for fermentation and aerobic respiration and is well supported by the presence of fermentative enzymes like pyruvate dehydrogenase (EC 1. Annotated genome shows that both strain JC669 T and A. giovannonii OJF2 T contains complete Embden-Meyerhof-Parnas pathway, TCA cycle and both oxidative and non-oxidative pentose phosphate pathways. Sugars are probably not the preferred carbon source for energy and biomass formation since genes encoding for enzymes in gluconeogenesis and glyoxylate shunt were not detected (Table S1). The genome mining also showed genes nifH, nifD, and nifK that encode structural components of molybdenum-dependent (Mo-dependent) nitrogenase that help in dinitrogen xation by converting dinitrogen to ammonia were observed in strain JC669 T and absent in A. giovannonii OJF2 T . Strain JC669 T and A. giovannonii OJF2 T also showed incomplete pathway of nitri cation involving conversion of ammonia to nitrate. The gene encoding for methane/ammonia monooxygenase which helps in conversion of ammonia to hydroxylamine an intermediary step in nitri cation is absent in both strains. The genome also revealed the presence of alcohol dehydrogenase (NADP+) and chlorobenzene. Strain JC669 T showed beta-Lactam resistance, Bla system pathway and the same is not observed in A. giovannonii OJF2 T . Strain JC669 T showed genes encoding for chitin degradation and showed no genes encoding for N-acetylglucosamine-6-sulfatase and N-sulfoglucosamine sulfohydrolase that helps in degradation of sulphated glycopolymers. This is in line with the experimental analysis showing that N-acetyl glucosamine is not required for growth. The dbCAN metaserver is used for CAZy (Carbohydrate active enzymes) annotation and it predicted that both the strains have a similar genes responsible for encoding carbohydrate-active enzymes (CAZymes). Strains JC669 T and A. giovannonii OJF2 T have large number of carbohydrate binding molecules (CBMs) followed by glycosyl transferases (GTs) and glycoside hydrolases (GHs). The major families in both strains are GH0, GT4, CBM48, GT2, CBM32 and CBM35 and are likely to arranged in clusters (Fig. S4) and also most of these enzymes in both strains are predicted to involve in the extracellular hydrolysis of substrates as they contain Nterminal signal peptides.
Twenty two functional categories of proteins were identi ed using Clusters of Orthologys (COG) and proteins belonging to categories general functional prediction only (R) nuclear structure (Y) were not observed in both strains. Both strains showed similar pattern and number of COG categories with a few minor differences where majority of proteins belongs to category S (unknown function) followed by energy production and conversion (C) and transport and metabolism (Fig. S5). Both strains are predicted to show resistance against antibiotics like uoroquinolone and tetracycline as they have genes encoding for the resistance-nodulation-cell division (RND) antibiotic e ux pump. Both strain JC669 T and A. giovannonii OJF2 T have the genes to putatively produce staurosporine (anti-cancerous compound) from L-tryptophan. Planctomycetes, believe to produce bioactive secondary metabolites since they have complex life style (Wiegand et al., 2020). Strains JC669 T and A. giovannonii OJF2 T have putative biosynthetic genetic clusters (BGCs) for producing secondary metabolites like bacteriocin, Type I Polyketide Synthase (T1PKS),Non-Ribosomal Peptide Synthetase clusters (NRPS), heterocyst-glycolipid synthase-like PKS (hgIE-KS), terpenes and indole. Putative genetic cluster of lassopeptide and lanthipeptide is exclusive for strain JC669 T and A. giovannonii OJF2 T respectively (Fig. S6).
Morphological and physiological analysis TEM image of the strain JC669 T shows cytoplasm, nucleoid region, invagination of enlarge periplasm, cytoplasm, nucleoid region with mbriae in daughter cell only and cell divide by budding (Fig. 5a). In SEM images, cells of strains JC669 T (Fig. 5b,c) and A. giovannonii DSM 22561 T (Fig. 5d,e) are spherical to oval-shaped with crateriform structures distributed throughout their surfaces (Fig. 5b,c,d,e). Both strain JC669 T (Fig. 5b,c) and A. giovannonii DSM 22561 T (Fig. 5d,e) produce fruiting bodies like structures which are distinct from each other. Free spore-like bodies were also observed (Fig. 5c) which were also stained (Fig. 6a,b) with the spore speci c uorescent stain "sporotan" (Senthilnathan et al., 2020; Ali et al., 2020). Out of curiosity we have done the genome mining of strain JC669 T and A. giovannonii for genes encoding for putative proteins responsible for forming fruiting bodies/spores. Amino acids sequences of strain JC669 T and A. giovannonii extracted from the NCBI and P-Blast was performed with the amino acids sequences of various important proteins involved in the formation of endospore as in Firmicutes represented by Bacillus subtilis and exospore as in Actinobacteria represented by Stretomyces coelicolor.
Blast analysis in the form of heat maps showed important sporulation proteins in strain JC669 T and A. giovannonii. However, their percentage of identity is low and mainly in the range of 20-40% (Fig. 7a,b). These ndings are similar to the peptidoglycan synthesis and cell division proteins discovered among Planctomycetes members, where the percentage of identity was also low (Jeske et al. 2015;). We do not conclude from this study with regard to the fruiting bodies and spores among the two taxa since it requires a more in-depth systematic analysis, however leave with a question of such possibility.
NaCl tolerance (0-7% w/v) for strain JC669 T and A. giovannonii DSM 22561 T was tested at 25°C and pH 7. Strain JC669 T and A. giovannonii DSM 22561 T didn't grow in the presence of NaCl even at 0.5%. The optimum temperature (5°C-45°C, at 5°C interval) required for cell growth was tested at pH 7. The range of temperature tolerance of the strain JC669 T is between 10°C and 30°C with maximum growth was observed between 22-25°C (optimum temperature). The optimum temperature for the cell growth of A.
giovannonii DSM 22561 T is 28-30°C and growth was observed between 15°C and 40°C. pH tolerance (4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0) for growth of both the strains JC669 T and A. giovannonii DSM 22561 T was tested at 25°C in buffered broth, as earlier described by (Bondoso et al. 2015). The optimum pH for cell growth of both strain was pH 7.0 and up to pH 9.0 cells growth was observed in both the strains.
Strain JC669 T and A. giovannonii DSM 22561 T were unable to reduce nitrate. Vitamin B 12 was not obligate for their cell growth. Cell growth observed with various organic substrates/nitrogen sources of strain JC669 T and A. giovannonii DSM 22561 T are given in Table. 1 or in the species description.
Enzymatic activity for strain JC669 T and A. giovannonii DSM 22561 T was performed using API ZYM kit.

Chemotaxonomic characterization
The major fatty acid pro le of the strain JC669 T was found to be in congruence with the type species of the genus mainly contain C 18:1 ω9c, C 16 (Table S2).
The polar lipid pro le of strain JC669 T (Fig. S7a) was not following type species of Aquisphaera, A.
giovannonii DSM 22561 T (both from this study [ Fig. S7a, b] and from the literature (Bondoso et al., 2015) using A. giovannonii OJF2 T . However, the major polar lipids of A. giovannonii DSM 22561 T in this study (Fig. S7b) are similar to what was reported earlier except for the absence of unidenti ed phospholipid, unidenti ed glycolipids, diphosphatidylglycrol and presence of six unidenti ed lipids (Bondoso et al., 2015). Polar lipids of strain JC669 T are phosphatidylcholine (PC), an unidenti ed amino lipid (AL), unidenti ed choline lipid (UCL), and six unidenti ed lipids (UL1-6) (Fig. S7a). For strain A. giovannonii DSM 22561 T we identi ed phosphatidylcholine (PC), a phosphatidylglycerol (PG) and six unidenti ed lipids (Fig. S7b). Strain JC669 T differ from A. giovannonii DSM 22561 T in not having phosphatidylglycerol but additionally having an unidenti ed aminolipid and an unidenti ed choline lipid. MK6 is the predominant quinone of both strains. Both strains have two polyamines; symhomospermidine and putrescine, while A. giovannonii DSM 22561 T in addition has an unidenti ed polyamine (Fig. S8).
Strain JC669 T resembles the type strain of its phylogenetically closest neighbour regarding having light yellowish orange colonies, major polar lipids and major fatty acids. However, strain JC669 T has about 3.2% less in the G + C mol% than A. giovannonii OJF2 T . The phenotypic differences (Table 1) are supported by the molecular differences (16Ss rRNA phylogenomic analysis) and genomic relatedness Table 1 Differential characteristics of (1) strain JC669 T and 2. A. giovannonii DSM 22561 T . 16S rRNA gene sequence GenBank accession number is LR782133. Genome of strain JC669 T has been deposited in GenBank under the accession numbers JAALJH000000000. The version described in this paper is version JAALJH010000000.
Proposal to emend the genus description of " Aquisphaera" The genus "Aquisphaera" is less understood as it contains single species ("A. giovannonii OJF2 T ") which is described so far. The original description of this genus was based mostly on the morphological, chemotaxonomic and genomic features (Bondoso et al., 2011). We propose to broaden the genus characters of "Aquisphaera" by emending the genus description. Gaurav and Ramana thank DBT, New Delhi for the award of SRF and TATA innovation fellowship, respectively. Sasikala, Dhanesh, Shabbir and Jagadeeshwari thank UGC for Mid-career award, RA, SRF and TEQIP awards for JRF respectively. We also thanks Mr. Bernhard Schink for providing species name and etymology for strain JC669 T . Prof. T. Radhakrishnan and Dr. Senthilnathan Department of Chemistry are thanked for providing "Sporotan".
Author contributions KL performed sample collection, KG performed media optimization and polar lipid analysis, KG and KL isolated the strain, performed the initial cultivation, strain deposition and strain characterization; KG and DK performed the electron microscopic analysis; JU performed the genomic and phylogenetic analysis; SA performed and analyzed the data for polyamines; GMK and KG performed genome mining for genes related to sporulation; KG and KL wrote the manuscript; Ramana and Sasikala supervised the study and contributed to text preparation and revised the manuscript. All authors read and approved the nal version of the manuscript.
Con icts of interest: The authors declare that there is no con ict of interest.
Ethical statement: Not applicable