Introduction

Bacillus members are ubiquitous and are found in a variety of environments ranging from desert sands and hot springs to Arctic soils and from freshwater to marine sediments (You et al. 2013). The predominant characteristics of members of the genus Bacillus are Gram-stain-positive, spore forming, rod shaped, containing menaquinone 7 (MK-7) as the major menaquinone, diphosphatidylglycerol, phosphatidylglycerol and phosphatidylethanolamine as the major polar lipids, and meso-diaminopimelic acid as the diagnostic cell wall diamino acid, with a few exceptions. Most of the Bacillus members have iso-C14:0, iso-C15:0, anteiso-C15:0, iso-C16:0 or anteiso-C17:0 as major fatty acids. The DNA G+C content range of Bacillus species is 36–52  mol% (Claus and Berkeley 1986; Holt et al. 1994; Rheims et al. 1999; Yumoto et al. 2004; Lim et al. 2007; Zhang et al. 2012; You et al. 2013; Sonalkar et al. 2015; Feng et al. 2016). Bacillus strains have a wide biotechnological potential of industrial interest, some of these include production of antibiotics, enzymes and other metabolites (Banat et al. 2000; Balcàzar and Rojas-Luna 2007; Sorokulova et al. 2008). During the study of cultivable bacterial diversity from the Lonar lake, India, strains 18CT and 6C were isolated. This study focuses on the taxonomic position of the strains 18CT and 6C based upon the polyphasic approach.

Materials and methods

Isolation, maintenance of cultures and reference strains

Strains 18CT and 6C were isolated from Lonar lake, located at Buldhana, Maharashtra, India (Latitude 19°58′, Longitude 76°36′), which is a unique basaltic rock meteorite impact crater, situated in the formerly volcanic Deccan trap geological region. The samples (at the time of sample collection, the sample had a pH of 9.5, salinity of 5.6% and temperature of 30 °C) serially diluted in sterile distilled water were spread plated on a alkaline nutrient agar medium consisting of (g l−) peptone (5), NaCl (5), beef extract (1.5), yeast extract (1.5), agar (15) in 1 litre of NaHCO3/Na2CO3 buffer (100 mM in deionized water; pH 10). Pure cultures of the strains 18CT and 6C were obtained by repeated streaking of the isolates on alkaline nutrient agar plates. Pure cultures were then preserved in 15% glycerol stocks at 4 °C for further use.

The type strains B. cohnii KCTC 3572T, B. zhanjiangensis KCTC 13713T and B. halmapalus LMG 17950T were obtained from Korean collection for type culture (KCTC) and Belgian Coordinated Collection of Microorganisms/laboratory of microbiology at the University of Ghent (BCCM-LMG), respectively, and used as reference strains.

Morphological and biochemical characterization

The phenotypic characters of strains 18CT and 6C were characterized following the minimum standards for describing new taxa of aerobic, endospore-forming bacteria recommended by Logan et al. (2009). Morphological properties, such as cell shape, cell size and motility (hanging drop method) were observed by phase contrast light microscopy (Magnus MLX). Further, spore shape, spore position, sporangial swelling and the presence of parasporal bodies were tested by phase contrast microscopy. Flagellum staining was performed as described by Kodaka et al. (1982). The pH range 6–12, with an interval of 0.5 were tested (K2HPO4–KH2PO4 buffer for pH 6.0–8, NaHCO3–NaOH buffer for pH 8.5–11, and Na2CO3–NaOH buffer for pH 11.5–12). The pH tests were conducted in triplicates and the results reported were an average value of the two highest values found during the tests. The temperature (0, 4, 10, 16, 20, 28, 35, 37, 40, 45 and 50 °C) and salt concentration (0–25% w/v, with an interval of 0.5% w/v) ranges for growth were examined in LB broth medium and the results were recorded after 48 h of incubation. Growth under anaerobic conditions was determined on modified NA supplemented with 0.5% (w/v) glucose and with or without 0.1% (w/v) nitrate using the Anaerobic Systems (Himedia). Various biochemical tests, such as hydrolysis of starch, tyrosine, xanthine, hypoxanthine, casein and gelatin, as well as, urease, nitrate reduction, Voges–Proskauer test, methyl red test, H2S production, indole production, oxidase and catalase activities were carried out as mentioned by Smibert and Krieg (1981), Oren et al. (1997) in the nutrient medium or the specified medium. Utilization of various substrates as sole carbon and energy sources or carbon, nitrogen and energy sources was determined using a basal medium with the following composition (g l−1): yeast extract, 0.01; KNO3, 1.0; KH2PO4, 1.0; MgSO4 7H2O, 0.2; (NH4)2HPO4, 1.0; NaCl, 80; Na2CO3, 20. To this liquid medium, a 0.1% (w/v) filter-sterilized substrate was added. Carbohydrates were used at a final concentration of 0.2% (w/v). Antibiotic sensitivity tests were performed by growing the strain as a lawn on the nutrient agar plate and inserting discs containing various antibiotics. The zone of inhibition was measured to identify the antibiotic effect on the strain.

16S rRNA gene sequencing, phylogenetic analysis, DNA–DNA relatedness and G+C composition determination

Genomic DNA was extracted and purified according to the method of Marmur (1961). The 16S rRNA gene sequences of strains 18CT and 6C were obtained by PCR as described earlier (Vishnuvardhan Reddy et al. 2013). Identification of phylogenetic neighbours and calculation of pairwise 16S rRNA gene sequence similarity were achieved using the Ezbiocloud server (http://www.ezbiocloud.net//; Yoon et al. 2016). The CLUSTAL_W algorithm of MEGA 5 (Tamura et al. 2011) was used for sequence alignments and the phylogenetic analysis of the near complete (~1450 bp) sequence of the 16S rRNA gene. Distances were calculated using the Kimura correction in a pairwise deletion manner (Kimura 1980). Neighbour-joining (NJ), minimum evolution (ME), maximum likelihood (ML) and maximum parsimony (MP) methods in the MEGA 5 software (Tamura et al. 2011) were used to construct phylogenetic trees. Percentage support values were obtained using a bootstrap procedure. The taxonomic relationship between strains 18CT, 6C and reference strains was examined using DNA–DNA hybridization which was determined using a membrane filter technique (Tourova and Antonov 1987), using Nick translation kit (code no. LCK-1) supplied by BRIT, Jonaki, CCMB campus, Hyderabad. Hybridization was performed with three replications for each sample (control: reversal of strains was used for binding and labelling). α-P32 dCTP was used for labelling the probe. The DNA immobilized on the blots (nylon membranes) were probed with labelled DNA and then exposed to phosphor-imaging screen (Amersham Biosciences). The phosphor-imaging screen was scanned and quantified using a Typhoon (3480) variable mode imager. The percent hybridization was calculated according to the formula: % hybridization = (counts obtained from heterologous hybridization/counts obtained from homologous hybridization) × 100. The mol% G+C of the DNA of strains 18CT and 6C were determined by HPLC (Mesbah et al. 1989).

Chemotaxonomic characterization

The chemotaxonomic characterization of strains 18CT, 6C and reference strains were analysed from cells grown in Luria–Bertani (LB) medium at 37 °C with 7.5 pH and 2% (w/v) NaCl. Cells were harvested by centrifugation (10,000g for 15 min at 4 °C) on reaching a cell density of 70% of the maximum optical density (100% = 0.8 OD540) and the lyophilized pellet was used for analysis. Cellular fatty acids of strains 18CT, 6C and reference strains were methylated, separated and identified according to the instructions for the Microbial Identification System (Microbial ID; MIDI 6.0 version; peak identification was done based on RTSBA6 data base) [Sasser (1990); revised http://www.midi-inc.com]. FAME analysis was outsourced to Royal Research Labs, Secunderabad, India. Polar lipids were extracted from 1 g of freeze-dried cells with methanol:chloroform:saline (2:1:0.8 v/v) as described by Kates (1986) and were separated using silica gel TLC (Kieselgel 60 F254; Merck) by two-dimensional chromatography using chloroform:methanol:water (75:32:4 v/v) in the first dimension and chloroform:methanol:acetic acid:water (86:16:15:4 v/v) in the second dimension (modified after Tindall 1990a, b; Oren et al. 1996). Total polar lipids profiles were detected by spraying with 5% ethanolic molybdophosphoric acid and further characterized by spraying with ninhydrin (specific for amino groups), molybdenum blue (specific for phosphates), Dragendorff’s reagent (quaternary nitrogen) or α-naphthol (specific for sugars) (Kates 1972; Oren et al. 1996). Quinones of strain, 18CT and reference strains were determined by extraction with chloroform:methanol (2:1 v/v) mixture purified by TLC and analysed by HPLC (Tamaoka et al. 1983).

The peptidoglycan of strain 18CT and reference strains were isolated after disruption of the cells by shaking with glass beads and subsequent trypsin digestion, according to the method of Schleifer (1985). The cell wall was hydrolysed for amino acid analysis and analysed as described by Schleifer and Kandler (1972) and Hasegawa et al. (1983).

Results and discussion

Morphological and biochemical characterization

Cells of the strains 18CT and 6C were facultatively anaerobic, Gram-stain-positive, terminal endospore forming, motile rods with peritrichous flagella. The sporangium was unswollen without the presence of any parasporal bodies. The strains 18CT and 6C differed from B. zhanjiangensis KCTC 13713T and B. halmapalus LMG 17950T which were both aerobic sub-terminal endospore-forming rods. Moreover, the cells of strains 18CT and 6C formed chain-like structures in alkaline nutrient agar (Supplementary Fig S1). Both strains formed colourless colonies on nutrient agar in contrast to the cream white colonies formed by all the three reference strains (Spanka and Fritze 1993; Chen et al. 2011; Nielsen et al. 1995, respectively). The substrates which supported growth and the other biochemical characterization of the strains were mentioned in the species description. The strain 18CT was sensitive to ciprofloxacin (5 µg), tetracycline (30 µg), chloramphenicol (20 µg), streptomycin (10 µg), penicillin (10 µg), erythromycin (15 µg) and resistant to ampicillin (30 µg), amikacin (30 µg), kanamycin (30 µg) and nalidixic acid (30 µg). However, strain 6C showed resistance to gentamycin (120 µg), vancomycin (30 µg), to which the strain 18CT was sensitive. The differentiating phenotypic properties of strain 18CT from the related species of the genus Bacillus are summarized in Table 1.

Table 1 Characteristics used to distinguish strain 18CT from the type strains of phylogenetically related Bacillus species
Full size table

16S rRNA gene sequencing and phylogenetic analysis

The almost complete 16S rRNA gene sequences (1467 and 1450 bp) of strains 18CT and 6C were obtained (GenBank/EMBL/DDBJ accession numbers LT617055 and LT837689, respectively). The results of phylogenetic analysis of the 16S rRNA gene sequences suggested that strains 18CT and 6C formed a cluster with its nearest Bacillus neighbour B. cohnii KCTC 3572T (a composite tree is shown as Fig. 1). Ezbiocloud server search analysis revealed that strains 18CT and 6C were most closely related to B. cohnii KCTC 3572T (99.3 and 99.9%, respectively), B. zhanjiangensis KCTC 13713T (97.4 and 98.0%, respectively), B. halmapalus LMG 17950T (97.0 and 97.6%, respectively) and other members in the genus Bacillus (<97.0%). However, the DNA–DNA relatedness between the novel strains 18CT and 6C and B. cohnii KCTC 3572T (49.6 ± 0.9 and 51.6 ± 0.7, respectively), B. zhanjiangensis KCTC 13713T (42.9 ± 0.8 and 47.1 ± 0.3, respectively) and B. halmapalus LMG 17950T (39.9 ± 0.8 and 40.8 ± 0.3, respectively) indicated that the novel strains are remotely related to the reference strains and the hybridization values are within the recommended standards to delineate a bacterial species (Stackebrandt and Goebel 1994; Stackebrandt and Ebers 2006; Meier-Kolthoff Jan et al. 2013). Further, the high 16S rRNA gene sequence similarity (100%) and DNA–DNA relatedness (90 ± 5%) indicated that strains 18CT and 6C were closely related and represent a single bacterial species. The mol% G+C content of the DNA of strain 18CT and 6C was 39.3 and 39.2%, respectively, which was similar to that of the nearest phylogenetic neighbours.

Fig. 1
figure 1

Phylogenetic analysis of strains 18CT and 6C with other closely related members based on 16S rRNA gene sequences available from the EMBL database (accession numbers are given in parentheses). Multiple alignments, distance calculations (distance options according to the Kimura 2-parameter model) and clustering with the neighbour-joining method were performed using the software package MEGA version 5 (Tamura et al. 2011). Bootstrap values based on 1000 replications are listed as percentages at the branching points. Bar 0.01 nucleotide substitutions per nucleotide position

Full size image

Chemotaxonomic characterization

Whole cell fatty acid analysis of strains 18CT and 6C revealed that iso-C15:0 (32.5 and 32.9%, respectively) was the predominant fatty acid. However, significant proportions of anteiso-C15:0 (19.5 and 19.8%, respectively), C16:0 (11.5 and 10.9%, respectively), iso-C17:0 (9.5 and 9.2%, respectively) and anteiso-C17:0 (6.3 and 6.9%, respectively) were also detected in strain 18CT (Table 2). Polar lipids of strain 18CT include diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylethanolamine (PE), unknown phospholipid (PL) and three unknown lipids (L1-3) (Supplementary Fig. S2). These profiles were somewhat similar to the polar lipid profile of Bacillus subtilis subsp. subtilis DSM 10T (Kämpfer et al. 2006). Major quinone of strain 18CT was MK-7 (98.5%) with traces of MK-6 (1.5%), a similar MK-7 predominant quinone system was found in the type species of the genus Bacillus (Bacillus subtilis subsp. subtilis) (Collins and Jones 1980). The peptidoglycan cell wall amino acids of strain 18CT contains meso-DAP as the diagnostic diamino acid (Supplementary Fig S3) with peptidoglycan type A1γ (Schleifer and Kandler 1972) or A31 (Schumann 2011). The peptidoglycan is in contrast to its nearest neighbour B. cohnii KCTC 3572T which has peptidoglycan type A4β with ornithine as the diamino acid and aspartic acid as the interpeptide bridge instead of meso-DAP (Spanka and Fritze 1993).

Table 2 Fatty acid compositions of strain 18CT and phylogenetically closely related Bacillus species
Full size table

Conclusions

The absence of phenotypic characteristics such as activities of oxidase, catalase and the presence of phenotypic characteristics such as urea hydrolysis, H2S production, indole production and Voges–Proskauer test showed the distinctiveness of strains 18CT and 6C from all their nearest neighbours (Table 1). Further, the presence of C16:0 as a major fatty acid, C14:0, summed feature 3 and summed feature 8 as the minor fatty acids clearly differentiate the strains 18CT and 6C from their closest neighbours. Additionally, meso-DAP as the diagnostic diamino acid (Supplementary Fig S3; Table 1) distinguishes the strains 18CT and 6C from their closest neighbour B. cohnii KCTC 3572T, which was supported by phylogenetic analysis as well as DNA–DNA relatedness studies. Hence, the novel isolates 18CT and 6C are proposed as a new member of the genus Bacillus for which, the name Bacillus catenulatus sp. nov., is proposed.

Description of Bacillus catenulatus sp. nov. (ca.te′nu.la.tus. N.L. masc. adj. catenulatus chain like, referring to the chain-like cell feature of the bacterium, observed when grown at pH 9.5).

Colonies grown on nutrient agar were colourless, circular (1.0–2.5 mm in diameter), convex, opaque with entire margin. Cells were Gram-stain-positive with 0.2–0.4 µm wide and 1–4 µm long (Supplementary Fig. S1). Cells were rod shaped and motile with the help of peritrichous flagella. The cells formed chain-like structures when grown at alkaline conditions. Facultative anaerobic, terminal endospore forming inside the unswollen sporangium, parasporal bodies were absent. Growth occurred at a pH range of 7.0–10.5 with an optimum at 7.5. NaCl is not essential for growth and can be tolerated up to 15% (w/v) with an optimum growth at 2–3% (w/v). Optimum growth occurred at 37 °C with a range of 10–45 °C. Casein, hippurate, urea and starch were hydrolysed, whereas esculin, DNA, cellulose, tyrosine, xanthine, hypoxanthine and Tween 20 were not hydrolysed by the strain. Gelatin was liquefied. Oxidase and catalase activities were negative. The indole production from tryptophan was positive. The strain produced H2S, showed positive result for nitrate reduction and VP test but citrate utilization, methyl red test, activities of arginine hydrolase, phenylalanine deaminase, nitrite reduction and ornithine decarboxylase activities were all negative. Acids were not produced from most of the sole carbon sources tested. Growth of the strain was supported by lactose, d-maltose, inositol, d-mannitol, cellobiose, d-glucose, sucrose and d-fructose. Ammonium chloride and urea were the most suitable nitrogen sources, but growth was also observed with glutamate and aspartate. Major fatty acids (>5%) were iso-C15:0, anteiso-C15:0, C16:0, iso-C17:0 and anteiso-C17:0. Polar lipids include diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylethanolamine (PE), unknown phospholipid (PL) and three unknown lipids (L1-3). The predominant isoprenoid quinone was MK-7. The DNA G+C content of the type strain was 39.3 mol%.

The type strain is 18CT (=KCTC 33781T = CGMCC 1.15475T), isolated from an alkaline sediment sample of Lonar lake, India. An additional strain 6C with almost similar features and DNA G+C content of 39.2 mol% was also isolated from the same sediment sample. The additional strain 6C differed with the type strain urea hydrolysis.