Taxonomic characterization of Sphaerotilus microaerophilus sp. nov., a sheath-forming microaerophilic bacterium of activated sludge origin

A microaerophilic Gram-stain-negative bacilliform bacterial strain, FB-5 T, was isolated from activated sludge in Yokohama, Japan, that exhibited filamentous growth and formed a microtube (sheath). Cells were motile using a single polar flagellum. The optimum growth temperature and pH were 30 °C and 7.5, respectively. Strain FB-5 T was catalase-negative. Peptides and amino acids were utilized as energy and carbon sources. Sugars and organic acids were not utilized. Vitamin B12 enhanced the growth of strain FB-5 T. Sulfur-dependent lithotrophic growth was possible. Major respiratory quinone was UQ-8. Major fatty acids were C16:1ω7 and C16:0. The genomic DNA G + C content was 69.16%. Phylogenetic analysis of the 16S rRNA gene suggested that strain FB-5 T belongs to the genus Sphaerotilus. The close relatives were S. natans subsup. sulfidivorans and S. natans subsup. natans with 98.0% and 97.8% similarity based on the 16S rRNA gene analysis, respectively. The genome size (6.06 Mbp) was larger than that (4.39–5.07 Mbp) of the Sphaerotilus strains. The AAI values against the related strains ranged from 71.0 to 72.5%. The range of ANI values was 81.7 − 82.5%. In addition to these distinguishable features of the genome, the core genome and dDDH analyses suggested that this strain is a novel member of the genus Sphaerotilus. Based on its physiological properties and genomic features, strain FB-5 T is considered as a novel species of the genus Sphaerotilus, for which the name S. microaerophilus sp. nov. is proposed. The type strain is FB-5 T (= JCM 35424 T = KACC 23146 T). Supplementary Information The online version contains supplementary material available at 10.1007/s00203-024-03991-9.


Introduction
Bacterial strains of the genera Sphaerotilus and Leptothrix are sheath-forming bacteria within the class Betaproteobacteria, which are collectively called the Sphaerotilus-Leptothrix group due to their morphological, physiological, and phylogenetic relationships (van Veen et al. 1978;Siering and Ghiorse 1996).The Sphaerotilus-Leptothrix group belongs to the family Comamonadaceae of the order Burkholderiales (Willems et al. 1991).Recently, the family Sphaerotilaceae, which is closely related to the family Comamonadaceae and consists of members of the Sphaerotilus-Leptothrix group and related genera, has been approved by the International Committee on Systematics of Prokaryotes (ICSP) (Liu et al. 2022;Oren and Göker 2023).Members of the Sphaerotilus-Leptothrix group are widely distributed in aquatic environments including streams, springs, and activated sludge (van Veen et al. 1978;Emerson and Ghiorse 1992;Gaval et al. 2003;Baskar et al. 2012;Schmidt et al. 2014).The genus Sphaerotilus comprises four species, Sphaerotilus mobilis (formerly Leptothrix mobilis), Sphaerotilus hippei, Sphaerotilus montanus, and Sphaerotilus natans (further classified as S. natans subsp.natans and S. natans subsp.sulfidivorans) for which type strains are available (Gridneva et al. 2011;Liu et al. 2022).Reclassification of S. natans subsp.sulfidivorans as Sphaerotilus sulfidivorans has been proposed in 2021 (Grabovich et al. 2021).Currently, S. sulfidivorans is a synonym of S. natans subsp.sulfidivorans according to the List of Prokaryotic names with Standing in Nomenclature (LPSN).The genus Leptothrix is characterized by its ability to oxidize manganese which is not observed in the genus Sphaerotilus (van Veen et al. 1978), excluding S. mobilis (Spring et al. 1996).L. mobilis was reclassified to the genus Sphaerotilus as S. mobilis in 2022 based on the genus boundary values of  and AAI (67.12-71.55%)proposed for the Sphaerotilus-Leptothrix group and related genera (Liu et al. 2022).The strains belonging to the non-validly described species "Leptothrix discophora" were classified as Leptothrix cholodnii and L. discophora in 1996 (Spring et al. 1996).After classification, the genus Leptothrix comprises four species: Leptothrix ochracea, L. cholodnii, L. discophora, and Leptothrix lopholea.The type strains of L. ochracea, L. lopholea, and L. cholodnii are not available from culture collections (Spring et al. 1996;Yarza et al. 2013), although L. ochracea is the type species of the genus Leptothrix (Skerman et al. 1980).Instead of the type strain, a reference strain of L. cholodnii was proposed and is available (Spring et al. 1996).No genomic data is available for the type strains of L. ochracea, L. lopholea, and L. cholodnii in the database.
In this study, we report the isolation and taxonomic characterization of a bacterial strain originating from the activated sludge of the sewage treatment plant in Yokohama exhibiting a sheathed morphology typical of the genus Sphaerotilus.Here, we describe the isolation of strain FB-5 T , a member of the genus Sphaerotilus, and propose that the isolate belongs to a novel species, S. microaerophilus sp.nov.

Isolation and cultivation
Strain FB-5 T was isolated from the activated sludge of the sewage treatment plant in Yokohama, Japan.The activated sludge was suspended with water and the suspension was streaked on an agar medium (named Screening medium) followed by incubation at 25 °C for 3 days.Colony isolation was performed three times at three-day intervals using the same medium at 25 °C.Screening medium was composed of 15 g/L agar, 0.15 g/L soluble starch, 50 mg/L (NH 4 ) 2 SO 4 , 50 mg/L K 2 HPO 4 , 50 mg/L MgSO 4 •7H 2 O, 50 mg/L KCl, 100 mg/L CaCO 3 , 14 mg/L Ca(NO 0.02 mg/L NiCl 2 , 0.2 mg/L CuCl 2 , 0.4 mg/L thiamin HCl, and 0.01 mg/L vitamin B 12 .Colonies with a rough (hairy) appearance, possibly attributable to filamentous growth, were selected using this medium.For maintains, the isolate (strain FB-5 T ) was subcultured on an agar plate (named APP medium) composed of 15 g/L agar, 2 g/L Proteose-Peptone No. 3 (Difco), and 0.01 mg/L vitamin B 12 at 15-25 °C in the range of one week.For taxonomic characterization, the strain was statically cultured at 25 °C for 5 days in a medium (named GPPY medium) composed of 4 g/L glucose, 2 g/L Proteose-Peptone No. 3 (Difco), 0.2 g/L yeast extract (Difco), 0.2 g/L MgSO 4 ‧7H 2 O, and 0.01 mg/L vitamin B 12 .For short-term storage in the range of one month, subcultures were performed at 15 °C using a semi-solidified GPPY medium containing 5 g/L agar.For long-term storage, the cells statically grown in GPPY medium at 25 °C for 5 days was stored at −80 °C in the presence of 20% glycerol.Lyophilization was not performed because of a loss of viability.

Phenotypic analyses
Unless otherwise described, the cells statically grown in GPPY medium at 25 °C for 5 days was used for phenotypic characterization.Gram-staining was performed using Favor G Nissui (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) according to the supplier's instructions.The utilization of organic compounds as sole carbon and energy sources was determined by monitoring the increase in turbidity (absorbance at 660 nm) after triplicate subcultures in a mineral medium (named NCM medium), which was used for the characterization of the Sphaerotilus strains (Gridneva et al.2011), mainly composed of (NH 4 ) 2 SO 4 , CaCl 2 , MgSO 4 ‧7H 2 O, Na 2 S 2 O 3 , and organic compound to be tested.Sulfur-dependent lithotrophic growth was examined using batch cultures under the conditions reported by Gridneva and collaborators (Gridneva et al.2011).Sulfurdependent lithotrophic growth was also examined using fed-batch cultures feeding with Na 2 S solution (1 g/L) in a 500 mL-flask (four-necked round-bottom flask containing 300 mL of medium) at 25 °C using a modified DSMZ573 medium (named m-DSMZ573 medium) under the following conditions: medium, DSMZ573 medium (without sodium acetate) supplemented with 1 mL/L Wolfe's vitamin solution (ATCC 2094 AN 1 medium); aeration, 0.1 L/min without agitation; pH, maintained at 7.2-7.8using a pH controller (NPH-690D, Nisshin Rika, Tokyo, Japan); Na 2 S feeding, H 2 S concentration in the exhaust was maintained at 0.2-1 ppm using a H 2 S sensor (FECS50-100, Figaro Engineering, Osaka, Japan) in combination with a control device composed of E5CD and G3NA-205B-UTU (OMRON, Kyoto, Japan).In the fed-batch culture, bacterial H 2 S consumption (removal) was calculated based on the amount of Na 2 S solution added to the flask to maintain the H 2 S concentration in the exhaust.Growth was detected as relative light units (RLU) by the ATP (including ADP and AMP) bioluminescence assay using a Lumitester Smart (Kikkoman, Tokyo, Japan).Manganese oxidation was examined in a stab culture (25 °C, one month) using MSVP agar (Emerson and Ghiorse 1992;Siering and Ghiorse 1996;Takeda et al. 2002) containing 1 mM Mn 2+ .L. cholodnii ATCC 51168 (= SP-6) and S. natans JCM 20382 (= ATCC 15291) were used as the positive and negative controls, respectively.Note that the subspecies affiliation of S. natans JCM 20382 is not determined (Gridneva et al. 2011).Although these strains are not the type strains, they are commonly used as references to compare their phenotypic properties with those of related isolates (Sawayama et al. 2011;Nott et al. 2020;Kashiwabara et al. 2021;Kunoh et al. 2021).Growth under aerobic conditions was examined on GPPY agar (GPPY medium solidified with 1.5% agar) and APP medium at 25 °C for 10 days.Growth under microaerophilic conditions was examined by semi-solid-state cultivation in a test tube with a diameter of 1.8 cm at 25 °C for 10 days using GPPY medium supplemented with 0.5% agar.Microaerophilic growth was also examined on GPPY agar (GPPY medium solidified with 1.5% agar) at 25 °C for 10 days using AnaeroPouch-MicroAero (Mitsubishi Gas Chemical, Tokyo, Japan).Likewise, anaerobic growth was examined using AnaeroPouch-Anaero (Mitsubishi Gas Chemical), respectively.Catalase and oxidase activities were evaluated using 3% (v/v) H 2 O 2 and 1% (w/v) tetramethyl-p-phenylenediamine, respectively, as previously described (Smibert 1994).The following characteristics were determined using the ID test NF-18 (Nissui Pharmaceutical): nitrate reduction to nitrite; nitrite reduction to nitrogen; gelatin liquefaction; indole production; hydrolysis of esculin, urea, and arginine; decarboxylation of lysine and ornithine; and β-galactosidase activity.The effect of temperature on growth was investigated using GPPY medium at 5-50 °C (5 °C intervals).The effect of pH was investigated at 25 °C using GPPY medium at pH 3-10 (1 pH unit intervals) adjusted with HCl or NaOH.The effect of NaCl concentration was investigated at 25 °C using GPPY medium supplemented with 0, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 1, 2, or 3% (w/v) NaCl.The growth in static cultures was checked by monitoring turbidity (absorbance at 660 nm) for 5 days.

Chemotaxonomic analysis
The cellular fatty acid composition was determined using the Sherlock Microbial Identification System (MIDI) according to the manufacturer's instructions.The major respiratory quinones were determined using the method described by Tamaoka and collaborators (Tamaoka et al. 1983).PHB accumulation was determined using gas chromatography, as described previously (Takeda et al. 1995).For these experiments, cells grown in GPPY medium were used because strain FB-5 T did not grow in the medium used for chemotaxonomic characterization of the closely related Sphaerotilus strains (Gridneva et al. 2011;Grabovich et al. 2021).

16S rRNA gene phylogeny
The 16S rRNA gene was amplified using colony PCR with the primers 27F (Jensen et al. 2009) and 1500R (Weisburg et al. 1991) and followed by Sanger sequencing (Applied Biosystems 3730xl DNA Analyzer; Applied Biosystems, Waltham, MA, USA) in two directions.The sequence was confirmed to be the same as that extracted from the genome assembly and deposited in DDBJ/ENA/GenBank under the accession number LC775240.Phylogenetic analysis using neighbor joining (NJ), unweighted pair group method with arithmetic mean (UPGMA), maximum likelihood (ML), and maximum parsimonious (MP) methods was performed based on the 16S rRNA gene sequences.Genes for analysis were extracted from the genome sequences (Table S1) of the reference strains using Barrnap 0.9 (https:// github.com/ tseem ann/ barrn ap), followed by multiple alignment (Q-INS-i algorithm) using MAFFT version 7 (Katoh et al. 2019).NJ and UPGMA trees were constructed using PHYLIP 3.698 (https:// evolu tion.genet ics.washi ngton.edu/ phylip.html) based on pairwise sequence similarities (Table S2).Similarities were calculated using blastn implemented in BLAST 2.14.0 + .MP tree was constructed using MEGA 11.0.13(Tamura et al. 2021).For a reliable estimation, only sequences within the Sphaerotilus-Leptothrix group were used, and the root was selected using the midpoint rooting method.The topological robustness of the NJ, UPGMA, and MP trees was evaluated using a bootstrap analysis with 1000 replicates.The maximum likelihood tree was constructed using IQ-TREE 2.3.3 (Minh et al. 2020) under the TN + F + I + G4 as the best-fit nucleotide substitution model, which was selected by the ModelFinder (Kalyaanamoorthy et al. 2017) according to the Bayesian information criterion.
Branching robustness was estimated using the SH-like approximate likelihood ratio test (SH-aLRT) with 1000 replicates.

Genome sequencing and analysis
The genome sequence of strain FB-5 T was determined using the following procedure: after freezing in liquid nitrogen, the cell pellet was ground, and DNA was extracted using the Wizard Genomic DNA Purification Kit (Promega, Madison, CA, USA).After purification using AMPure XP (Beckman Coulter, Brea, CA, USA) and a DNeasy Power Clean Pro Cleanup Kit (Qiagen, Venlo, Netherlands), a DNA library was constructed using the SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences, Menlo Park, CA, USA).The library was sequenced using Sequel IIe (Pacific Biosciences) in combination with the Binding Kit 2.2 (Pacific Biosciences).Based on these sequences, consensus sequences were prepared and HiFi reads were obtained by omitting sequences of low reliability (< QC 20) using SMRT Link v11 software (Pacific Biosciences).After removing short reads (< 1000 bp) using Filtlong (https:// github.com/ rrwick/ Filtl ong), HiFi reads were assembled using Flye (https:// github.com/ fende rglass/ Flye/ blob/ flye/ docs/ USAGE.md), and accuracy was confirmed using Bandage (Wick et al. 2015) and CheckM (Parks et al. 2015).Complete genomic data, including the plasmid, were deposited in DDBJ/ENA/ GenBank under accession numbers AP025730 (genome) and AP025731 (plasmid).The assembled sequences were annotated using DFAST (Tanizawa et al. 2016), and 5311 genes were identified.The basic features of the genomic data used in this study, including those of strain FB-5 T , are listed in Table S1.The contamination and completeness were estimated using CheckM2 (Chklovski et al. 2023).The dDDH value was calculated using GGDC 3.0 (Meier-Kolthoff et al. 2022) with formular 2. A heat-map style matrix was produced using TBtools 2.016 (Chen et al. 2020).Multilocus sequence analysis based on the core gene set (concatenated sequences of 92 core genes from 32 genomes) was performed using UBCG 3.0 (Na et al. 2018), and a phylogenetic tree was constructed using FastTree (Price et al. 2010) implemented in the UBCG pipeline.Branching robustness was estimated using SH-aLRT with 1000 replicates.The genome-to-genome distance was calculated by AAI using EzAAI 1.2.2 (Kim et al. 2021), and the AAI matrix was produced using TBtools 2.016.The corresponding matrix plot was generated using PHYLIP 3.698.The genome-to-genome distance was further estimated by ANI using FastANI 1.33 (Jain et al. 2018).The ANI matrix was produced using TBtools 2.016, and the corresponding matrix plot was produced using PHYLIP 3.698.

Isolation and maintenance
Initially, we attempted to isolate strains of the genus Haliscomenobacter (a genus in the phylum Bacteroidota comprising sheath-forming filamentous bacteria) capable of degrading various macromolecules, including starch and hyaluronate (Mori et al. 2023), using Screening medium containing starch.After repeated streak purification, a filamentous strain was isolated as a single colony and designated as strain FB-5 T .The colonies of strain FB-5 T on Screening medium exhibited a loss of viability after 7 days of cultivation.Growth was observed on a starch-free Screening medium.No growth was observed in agar-free (liquid) Screening medium.These results suggested that the growth of the isolate was supported by organic impurities in the agar and that strain FB-5 T was not a member of the genus Haliscomenobacter.The colonies formed on APP medium were irregularly (rough) shaped (Fig. S1a).Smooth colonies were rarely observed as well (Fig. S1b).Colonies of either shape produced colonies of both shapes in subculture.A rough colony was used for taxonomic characterization and subculture.

Morphology
Using phase-contrast microscopy, both the filaments (lines of cells enclosed by sheaths; Fig. 1a) and single rod-shaped cells (cells without a sheath; Fig. 1b) were observed in static cultures using GPPY medium.Single rod-shaped cells occasionally exhibited motility.Cells were not connected, and filamentation was caused by sheath formation.Filamentation (sheath formation) was promoted by the addition of 0.1 g/L CaCO 3 to the medium.The enhancement of sheath formation by Ca 2+ and Mg 2+ has been observed in L. cholodnii (Kunoh et al. 2021).However, no effect of Mg 2+ for sheath formation of strain FB-5 T was suggested, as similar filamentation was observed in GPPY medium with or without MgSO 4 •7H 2 O. Sheath formation was confirmed by SPM observation (Fig. S2).Filaments were commonly observed in the rough colonies (Fig. S1a) formed on a solid medium.In contrast, filaments were rare in the smooth colonies (Fig. S1b).Accordingly, a rough colony was selected to maintain the sheath-forming ability.Sheaths were not detected using TEM (Fig. 2a) or SEM (Fig. 2b), suggesting a weak sheath-forming capability of strain FB-5 T .The electron microscopy images (Fig. 2) revealed a cell size of 0.7-1.0µm in width and 2.0-6.5 µm in length.The sthA (AB050638) and lthB (ACB33244) genes are essential glycosyl transferase genes for sheath formation in S. natans and L. cholodnii, respectively (Suzuki et al. 2002;Kunoh et al. 2023).A homology search using BLAST showed a sthA-like gene (WP_251972251/BDI04101) with 78% identity in the genome of strain FB-5 T , whereas a lthB-like gene was not detected, suggesting that strain FB-5 T forms a Sphaerotilustype sheath (Kondo et al. 2011;Kashiwabara et al. 2021).Sphaerotilus-type sheath is mainly composed of glucuronic acid, galactosamine and glucose (Kondo 2011;Kashiwabara et al. 2021).In contrast, Leptothrix-type sheath contains N-acetylgalactose, galactosamine, galacturonic acid, and glucosamine (Takeda et al. 2005).Purification of the sheath and subsequent sugar composition analysis are required to determine the type of sheath of strain FB-5 T , which will be our future study.Traces of a single polar flagellum (the remaining short flagellum) were rarely observed in the SEM images (Fig. 2c, d).We assume that the motility of strain FB-5 T is attributed to a single polar flagellum, as described for other members of the genera Sphaerotilus and Leptothrix (Spring et al. 1996;Grabovich et al. 2021).

Physiology
Strain FB-5 T was stained Gram-negative.Growth was observed in GPPY medium containing 0-0.5% NaCl but not observed in the presence of 1-3% NaCl.Growth was observed only in media containing amino acid or amino acid-based organic compounds such as peptone, Proteose-Peptone No. 3, tryptone, yeast extract, and casamino acids (Table S3).To the best of our knowledge, Proteose-Peptone No. 3 is the most suitable energy and carbon source for growth.No sugars or organic acids served as the sole growth substrate.The preference for peptides and amino acids is a distinguishing feature of strain FB-5 T in comparison to other members of the genus Sphaerotilus.Vitamin B 12 enhanced the growth of the strain because its growth was poor when vitamin B 12 was omitted from GPPY medium.Vitamin B 12 requirements are commonly recognized in the Sphaerotilus and Leptothrix strains (Okrend and Dondero 1964;Emerson and Ghiorse 1992).Sulfur-dependent lithotrophic growth was not observed in batch culture; however, growth (increase in RLU) and consumption of H 2 S were detected in fed-batch culture.Continuous supply of sulfide is probably desirable for lithotrophic cultivation of strain FB-5 T .As shown in Fig. S3, manganese oxidation was not observed in strain FB-5 T .Poor growth was observed in the uppermost part of the stab culture of strain FB-5 T , suggesting the strain is facultatively microaerophilic and prefers microaerophilic conditions rather than aerobic conditions, unlike S. natans and L. cholodnii.S. natans has been reported to oxidize iron, coupled with nitrate reduction catalyzed by nitrate reductase (WP_037485935) (Park et al. 2014).The amino acid sequence of the nitrate reductase of S. natans showed 71% identity with a putative nitrate reductase (WP_251972978) of strain FB-5 T in a BLAST search, suggesting that the strain has iron-oxidizing potential.To confirm the preference of strain FB-5 T for growth under microaerophilic conditions, a semi-solid culture was performed.Poor growth was observed at the air-medium interface, whereas colonies formed 0.3-1 cm below the interface different from S. natans (Fig. S4), indicating that strain FB-5 T is microaerophilic.No growth was observed under microaerophilic and anaerobic conditions produced by AnaeroPouch-MicroAero and AnaeroPouch-Anaero, respectively.Since AnaeroPouch-MicroAero produces atmosphere of 6-12% O 2 , strain FB-5 T is expected to prefer O 2 concentrations above 12% and less than 21%.Additionally, strain FB-5 T was cytochrome oxidase-positive same as the Sphaerotilus strains, but catalasenegative different from the Sphaerotilus strains.Nitrate was reduced to nitrite.Nitrite was not reduced to nitrogen.Gelatin was not liquefied.Indole was not produced.Esculin, urea, and arginine were not hydrolyzed.Lysine and ornithine were not decarboxylated.β-Galactosidase activity was negative.Growth was observed in a temperature range of 15-35 °C, with 30 °C being optimal.The optimum pH was 7.5, whereas growth was observed in the pH range of 7-8.

16S rRNA gene-based phylogeny
The pairwise identities of the 16S rRNA gene sequences of strains related to strain FB-5 T are listed in Table S2.As shown in the phylogenetic tree (NJ) based on the 16S rRNA gene (Fig. 3), strain FB-5 T was closely related to strains of the genus Sphaerotilus, particularly to S. natans subsup.sulfidivorans (98.0%similarity) followed by S. natans subsup.natans (97.8% similarity).Phylogenetic analysis using ML, UPGMA, and MP methods supported this result.The level of identity with S. hippei was 97.8%, with S. montanus 97.2%, and with S. mobilis 96.5%.

Genomic features
The genome sequence of strain FB-5 T determined in this study exhibited a low level of contamination (0.26%), with 99.99% completeness, as shown in Table S1.The genomic DNA G + C content of strain FB-5 T was 69.16%, which was within the range (68.16 − 70.03%) of the Sphaerotilus strains.In the genome, 5,236 protein-coding and 75 noncoding genes were identified.The genome size of strain FB-5 T (6.06 Mbp) was higher than that of the Sphaerotilus strains (4.39 − 5.07 Mbp), which distinguished the isolate from related strains of the genus Sphaerotilus.Supporting the lithotrophic growth capability of strain FB-5 T in fed-batch culture, the genes required for sulfur metabolism and Calvin-Benson-Bassham cycle were observed in the genome as listed in Tables S5 and S6, respectively.Accordingly, strain FB-5 T was revealed to be lithotrophic, the same as S. natans subsp.sulfidivorans (Gridneva et al. 2011;Grabovich et al. 2021).Comparative genomic analysis was performed to further characterize the strain as a member of the Sphaerotilus-Leptothrix group.
The dDDH values between strain FB-5 T and the strains of the genus Sphaerotilus ranged from 21.8 to 22.6% as shown in the heat map style matrix (Fig. S5).The values were much lower than the species cutoff value of 70% (Goris et al. 2007;Meier-Kolthoff et al. 2013), indicating that strain FB-5 T represents a new species in the genus Sphaerotilus.In the phylogenetic tree constructed based on the core genes set (Fig. 4), strain FB-5 T was located outside the clade formed by other members of the Sphaerotilus-Leptothrix group.
The AAI values against the species in the Sphaerotilus-Leptothrix group ranged from 71.0 to 72.5% (Fig. S6), with the lowest value (71.0%) against S. natans subsp.natans and the highest value (72.5%) against L. cholodnii.In the AAI matrix plot (Fig. S7), strain FB-5 T formed a clade with L. cholodnii, L. discophora, and S. mobilis, suggesting that it was closely related to these species.A neighboring clade was formed by the Sphaerotilus strains other than S. mobilis.AAI values of L. cholodnii and L. discophora against the Sphaerotilus strains, including strain FB-5 T ranged from 70.7 to 72.6% (Fig. S6).Considering the AAI relatedness and Sphaerotilus-Leptothrix group-specific genus boundary AAI value of 67.12-71.55%(Liu et al. 2022), strain FB-5 T should be classified into the genus Sphaerotilus.The ANI values between strain FB-5 T and the strains of the Sphaerotilus-Leptothrix group ranged from 81.7 to 82.5% (Fig. S8).In the ANI matrix plot (Fig. S9), strain FB-5 T was located distantly within the clades of the genera Sphaerotilus and Leptothrix, suggesting its novelty.Based on the Sphaerotilus-Leptothrix group-specific genus boundary ANI value of 78.95-82.14%(Liu et al. 2022), strain FB-5 T should be classified into the genus Sphaerotilus.

Conclusion
Table 1 summarizes the major phenotypic features of strain FB-5 T that distinguish it from known closely related species.Because strain FB-5 T did not oxidize manganese, it could be easily distinguished from S. mobilis and strains of the genus Leptothrix.The cell shape and size were almost identical to those of the known species of the genus Sphaerotilus.Sheath formation by strain FB-5 T was not particularly stable in the absence of calcium, which differs from that observed in other known species of the genus Sphaerotilus.Strain FB-5 T is microaerophilic and catalase-negative, whereas the type strains of the genus Sphaerotilus are strictly aerobic and catalase-positive.Because lithotrophic sulfur oxidation is possible, strain FB-5 T is the second lithotrophic member found after S. natans subsp.sulfidivorans in the genus Sphaerotilus.The inability of strain FB-5 T to utilize sugars and organic acids as the sole energy and carbon sources distinguished it from known species in the genus Sphaerotilus.Based on the differences in phenotypic properties and genomic features, this strain should be classified as the type strain of a new species, Sphaerotilus microaerophilus sp.nov.

Fig. 1 Fig. 2
Fig.1Phase-contrast microscopy images of strain FB-5 T .Strain FB-5 T was statically cultivated at 25 °C for 4 days using GPPY medium.The aggregated cells were enclosed within a sheath (a) whereas non-aggregated cells remained without a sheath (b)

Fig. 3
Fig. 3 Phylogenetic tree (NJ) based on 16S rRNA gene sequences showing the relationship between strain FB-5 T and related strains.The percentages (> 80%) for 1000 bootstraps are shown next to the branches.The scale represents 0.01 nucleotide substitutions per site

Fig. 4
Fig. 4 Phylogenetic tree (ML) based on core genome sequences showing the relationship between strain FB-5 T and related strains.The percentages (> 70%) for 1000 bootstraps are shown next to the branches.The scale represents 0.1 nucleotide substitutions per site