Introduction

The genus Anoxybacillus, belonging to the phylum Firmicutes, was proposed by Pikuta et al. (2000) with Anoxybacillus pushchinoensis as the type species and its description was considerably emended by Pikuta et al. (2003). At the time of writing, this genus comprised 24 species with validly published names and three species with not validly published names (http://www.bacterio.net). Anoxybacillus species are widely distributed and isolated from geothermally heated environments. The taxonomy of Anoxybacillus members was predominantly based on 16S rRNA gene sequence analysis and DNA–DNA hybridization (DDH). However, it is widely known that the resolving power of 16S rRNA gene analysis often shows limited variation for discrimination of closely related species, such as the Anoxybacillus species. DDH is time-consuming and labor-intensive method and it is impossible to establish a central database. Phylogeny using whole-genome sequences-based metrics such as average nucleotide identity (ANI), digital DDH, and average amino acid identity (AAI) have become important tools for the delineation of prokaryotic taxa (Orata et al. 2018), and is being used for the reclassifcation of several bacterial taxa (Liu et al. 2019; Rao et al. 2022).

The type strain G2T of Anoxybacillus gonensis was isolated from hot spring in Turkey by Belduz et al. in 2003 and described as validly named species based on a polyphasic taxonomic approach. Anoxybacillus salavatliensis DSM 22626T was isolated from a high temperature well-pipeline sediment sample in Turkey by Cihan et al. (2011) and was validated in IJSEM (Validation List No. 138; Euzéby 2011). In the original article, Cihan et al. (2011) proposed A. salavatliensis DSM 22626T as a new species in the genus Anoxybacillus based mainly on DNA–DNA hybridization values between A. salavatliensis DSM 22626T and A. kamchatkensis DSM 14988T, A. amylolyticus DSM 15939T. Phylogenetic tree based on 16S rRNA gene sequences in the original article showed that A. salavatliensis DSM 22626T, A. gonensis G2T, A. kamchatkensis DSM 14988T, A. ayderensis AB04T and A. thermarum DSM 17141T clustered together. In the orginal article, Cihan et al. (2011) stated that A. salavatliensis DSM 22626T could be clearly differentiated from all of the closely related Anoxybacillus species (A. kamchatkensis DSM 14988T, A. thermarum DSM 17141T, A. ayderensis AB04T, A. gonensis G2T, A. flavithermus DSM 2641T and A. amylolyticus DSM 15939T) based on the results of Rep-PCR and ITS fingerprinting. Also, in the original article Cihan et al. (2011) determined DDH values between A. salavatliensis DSM 22626T and A. kamchatkensis DSM 14988T, A. amylolyticus DSM 15939T, they did not determine DDH values between A. salavatliensis DSM 22626T and the other closely related Anoxybacillus species (A. thermarum DSM 17141T, A. ayderensis AB04T, A. gonensis G2T, A. flavithermus DSM 2641T). During our genome-based analysis, we observed that A. salavatliensis DSM 22626T and A. gonensis G2T shared similar features; as a result, we attempted to clarify the relationship between that A. salavatliensis DSM 22626T and A. gonensis G2T through genomics-based methods, biochemical and chemotaxonomic characteristics. The data presented in this study provides evidence that A. salavatliensis DSM 22626T is later heterotypic synonym of A. gonensis G2T.

Materials and methods

A. salavatliensis DSM 22626T was purchased from the German Collection of Microorganisms and Cell Cultures GmbH (DSMZ). A. gonensis G2T was isolated by us and validly published, so A. gonensis G2T was obtained from our own laboratory collection. Two type strains were grown on trypticase soy agar (TSA) incubated at 50 °C for 24 h.

The genome sequencing of A. salavatliensis DSM 22626T was performed in this study, while genome sequences of A. gonensis G2T (JRZG00000000) was downloaded from NCBI Database (https://www.ncbi.nlm.nih.gov/genome/). For whole genome sequencing, genomic DNA was isolated from culture of A. salavatliensis DSM 22626T by using the QIAamp DNA Mini Kit according to the manufacturer's instructions (Qiagen, Hilden-Germany). Whole-genome sequencing of A. salavatliensis DSM 22626T was performed on an Illumina HiSeq 2500 next-generation platform with a 250-bp paired-end sequencing protocol by MicrobesNG (http://www.microbesng.uk, Birmingham, United Kingdom). Assemblies of raw sequence data were achieved using the full SPAdes assembly strategy on the patric web server (https://patricbrc.org/) (Wattam et al. 2017). The draft genome sequences were annotated by using the Rapid Annotations Using Subsystems Technology (RAST) server (Aziz et al. 2008). The obtained draft genome sequences were deposited in the National Centre for Biotechnology Information (NCBI) database under accession number JANGZY000000000.

The 16S rRNA gene sequence identity between A. salavatliensis DSM 22626T and A. gonensis G2T was compared using the pairwise alignment feature implemented on the EZBioCloud server (https://www.ezbiocloud.net/tools/pairAlign). The 16S rRNA gene sequences of closely related type strains were dowloaded from EzBioCloud server at https://www.ezbiocloud.net/ (Yoon et al. 2017a) and edited by using the BioEdit software (Hall 1999). Multiple sequence alignment of 16S rRNA gene sequences was performed using the ClustalW (Thompson et al. 1994). Evolutionary distances were calculated with the Kimura’s two-parameter model (Kimura 1980). The phylogenetic trees were generated in Mega-X using neighbor-joining (Saitou and Nei 1987), maximum parsimony (Kluge and Farris 1969) and maximum-likelihood method (Felsenstein 1981), with bootstrap values based on 1000 replications.

The phylogenetic analysis of A. salavatliensis DSM 22626T and A. gonensis G2T was carried out using the type strain genomes server pipeline (TYGS, https://tygs.dsmz.de/) (Meier-Kolthoff and Göker 2019). The digital DNA-DNA hybridization (dDDH) value between the draft genome sequences of A. salavatliensis DSM 22626T and A. gonensis G2T was calculated with the Formula 2 of the online Genome-to-Genome Distance Calculator at http://ggdc.dsmz.de/distcalc2.php (Meier-Kolthoff et al. 2013). Average nucleotide identity (ANI) values were calculated for evaluating the genetic relationship between between A. salavatliensis DSM 22626T and A. gonensis G2T by using the orthoANIu algorithm and an online ANI calculator (www.ezbiocloud.net/tools/ani) (Lee et al. 2016; Yoon et al. 2017b). A phylogenetic tree based on whole-genome sequences was constructed using the TYGS web server (https://tygs.dsmz.de/) (Meier-Kolthof and Göker 2019). The amino acid identity (AAI) value was calculated with CompareM (https://github.com/dparks1134/CompareM). For the phylogenetic and pangenome analyses, the genomes of A. salavatliensis DSM 22626T, A. gonensis G2T and all other Anoxybacillus species registered in RefSeq were re-annotated using Prokka 1.14.5 with default settings to avoid bias resulted from diferent annotations (Seemann 2014). The phylogenetic trees were then constructed using the ‘insert genome into species tree app’ (version 2.2.0), utilizing the FastTree 2 algorithm (Price et al. 2010). The pangenome was constructed using the genome as mentioned earlier set by ‘build pangenome with OrthoMCL app’ (v2.0) available at KBase platform (https://www.kbase.us/) (Arkin et al. 2018). The pangenome-based phylogenomic analysis was performed by the ‘phylogenetic pangenome accumulation (v1.4.0) app’ (Li et al. 2003; Arkin et al. 2018).

The API 20E, API 50CH strips and Vitek2 Bacilli Identification Card (BCL) microtest systems (bioMérieux) were used to evaluate the biochemical properties of A. salavatliensis DSM 22626T and A. gonensis G2T according to the manufacturer’s instructions. The polar lipids of strain A. salavatliensis DSM 22626T and A. gonensis G2T were extracted from 100 mg freeze-dried cells by using two-dimensional thin-layer chromatography (TLC) according to the method of Tindall (1990a, 1990b). Polar lipids were separated by two-dimensional TLC on silica gel. The first direction was developed with chloroform:methanol:water (65:25:4, v/v) and the second direction with chloroform:methanol:acetic acid:water (80:12:15:4, v/v). Detection was performed using 5% ethanolic molybdophosphoric acid for the total lipids, molybdenum blue for phospholipids, ninhydrin for aminolipids, and α-naphthol for glycolipids (Tindall et al. 2007). Standard lipids (dihosphatidylglycerol [DPG], phosphatidic acid [PA], phosphatidylglycerol [PG], phosphatidylethanolamine [PE], phosphatidylcholine [PC], phosphatidylserine [PS], and phosphatidylinositol [PI]) were used as reference. Isoprenoid quinones were extracted and purified from freeze-dried cells by following the procedure of Collins (1985) and analysed by high performance liquid chromatography (HPLC).

Results and discussion

The phylogenetic analysis based on whole genome sequences has clarified the taxonomic inconsistence of prokaryotic taxa; as a result, several bacterial species have been reclassified (Orata et al. 2018). In the present study, the taxonomic relationship of A. salavatliensis DSM 22626T and A. gonensis G2T was re-evaluated by using whole-genome phylogenetic analysis, biochemical and chemotaxonomic features. A. salavatliensis DSM 22626T was isolated from a high temperature well-pipeline sediment sample in Turkey; A. gonensis G2T was isolated from hot spring in Turkey.

In the original article, Cihan et al. (2011) stated that the Rep-PCR and ITS fingerprinting profiles differentiated A. salavatliensis DSM 22626T from all of the closely related Anoxybacillus species: A. kamchatkensis DSM 14988T, A. thermarum DSM 17141T, A. ayderensis AB04T, A. gonensis G2T, A. flavithermus DSM 2641T and A. amylolyticus DSM 15939T. In the original article, A. salavatliensis DSM 22626 T formed a cluster with A. gonensis G2T, A. kamchatkensis DSM 14988T, A. ayderensis AB04T and A. thermarum DSM 17141T in the phylogenetic tree based on 16S rRNA gene sequences. However, in the original article Cihan et al. (2011) determined DDH values between A. salavatliensis DSM 22626T and A. kamchatkensis DSM 14988T, A. amylolyticus DSM 15939T, they did not determine DDH values between A. salavatliensis DSM 22626T and the other closely related Anoxybacillus species.

In the present study, we determined the pairwise nucleotide sequence alignment (16S rRNA gene sequence) between A. salavatliensis DSM 22626T and A. gonensis G2T was 99.3% with a mismatch of ten nucleotides. Also, in the present study, we reconstructed the phylogenetic trees based on 16S rRNA gene sequences and determined that A. salavatliensis DSM 22626T and A. gonensis G2T clustered together in the neighbour-joining phylogenetic tree with high bootstrap resampling values of 97% (Fig. 1). Topologies of phylogenetic trees built according to the maximum-likelihood and maximum-parsimony algorithms also supported the results of the neighbour-joining algorithm (Fig. S1, S2). Further, in the phylogenomic tree (Fig. 2) A. salavatliensis DSM 22626T and A. gonensis G2T formed a robust branch different from other type strains of this genus with high bootstrap resampling values of 98%. The ANI value between A. salavatliensis DSM 22626T and A. gonensis G2T was 97.98% which was greater than the threshold value (95–96%) for species demarcation (Richter and Rosselló-Móra 2009), confirming that A. salavatliensis DSM 22626T and A. gonensis G2T were highly phylogenetically closely related. The calculated AAI value between the A. salavatliensis DSM 22626T and A. gonensis G2T was 98.1% and this value is also clearly above the suggested cut-offs for species delineation (AAI > 95%) (Luo et al. 2014), confirming that they belong to the same species. Also, digital DNA–DNA hybridization (DDH) analyses indicated that A. salavatliensis DSM 22626T and A. gonensis G2T exhibited 81.0% dDDH value which is higher than the cut-off (70%) used to classify bacterial strains to the same species (Wayne et al. 1987), further confirming that A. salavatliensis DSM 22626T and A. gonensis G2T should belong to the same genomic species. Pangenomic analysis of the Anoxybacillus species, including A. salavatliensis DSM 22626T and A. gonensis G2T, revealed 6,138 orthologous clusters that constituted the pangenome. The numbers of core genes, strain-specifc genes (singleton) and accessory genes (partial) were 1867, 2623, and 1648, respectively. According to the pangenome-based phylogenomic analysis, A. salavatliensis DSM 22626T and A. gonensis G2T formed a monophyletic clade and shared 2381 core genes (Fig. 3).

Fig. 1
figure 1

Neighbour-joining (NJ) tree constructed based on 16S rRNA gene sequences available from the GenBank database. Bootstrap values (expressed as percentages of 1000 replications) greater than 50% are shown at branch points. Bar, 0.01 represents substitutions per nucleotide position. Paenibacillus polymyxa DSM 36T was used as the outgroup

Full size image
Fig. 2
figure 2

Phylogenetic tree based on whole-genome sequences of A. salavatliensis DSM 22626T and A. gonensis G2T and related reference strains. The tree was inferred with FastME 2.1.6.1 (Lefort et al. 2015) from genome blast distance phylogeny (GBDP) distances calculated from genome sequences using the TYGS server (https://tygs.dsmz.de) (Meier-Kolthoff and Göker 2019) The branch lengths are scaled in terms of GBDP distance formula d5. The numbers at branches are GBDP pseudo-bootstrap support values ≥ 64% from 100 replications with an average branch support of 97.7%. The tree was rooted at the midpoint (Farris 1972)

Full size image
Fig. 3
figure 3

Pangenome-based phylogenomic analysis of Anoxybacillus species. Orthologous gene sets within a pangenome are partitioned into three categories: core (blue), singleton (red), and partial pangenome (pink). Pangenome-based phylogenomic analysis was created by the OrthoMCL and phylogenetic pangenome accumulation (v1.4.0) app. (Color figure online)

Full size image

In addition, this conclusion has been also confirmed by a comparation of phenotypic and chemotaxonomic features between A. salavatliensis DSM 22626T and A. gonensis G2T. In API 20E, API 50CH and Vitek2 BCL system, A. salavatliensis DSM 22626T and A. gonensis G2T shared similar biochemical features with few exceptions (Table 1). For example, acid production from D-galactose, D-mannose, esculin ferric citrate and, ellman, palatinose, β-glucosidase were negative for A. gonensis G2T, while positive for A. salavatliensis DSM 22626T. β-galactosidase, α-galactosidase and acid production from methyl-alpha-D-mannopyranoside and maltotriose were positive for A. gonensis G2T, while negative for A. salavatliensis DSM 22626T. Both species were shown positive for ONPG hydrolysis, tyrosine arylamidase, nitrate reduction, Leucine-arylamidase, gelatinase, phenylalanine arylamidase, Ala-Phe-Pro-Arylamidase, α-glucosidase, acid production from d-Xylose, d-Mannitol, d-Glucose, d-Fructose, methyl-a-d-glucopyranoside, salicin, d-Cellobiose, d-Maltose, d-Melibiose, d-Sucrose, d-Trehalose, inulin, d-Melezitose, d-Raffinose, starch, glycogen, d-Turanose. A. salavatliensis DSM 22626T and A. gonensis G2T were shown negative for arginine dihydrolase, tryptophan deaminase, citrate utilization, urease, lysine decarboxylase, Voges–Proskauer, l-Lysine arylamidase, ornithine decarboxylase, alanine arylamidase, indole production (tryptophanase), pyruvate, hydrogen sulfide production, β-xylosidase, l-Aspartate arylamidase, l-Proline arylamidase, l-Pyrrolydonyl arylamidase, phosphoryl choline, β-N-acetyl-glucosaminidase, cyclodextrine, Methyl-d-Xyloside, α-mannosidase, glycine arylamidase, N-acetyl-glucosamine, kanamycin resistance, β-mannosidase, growth in 6.5% NaCI, oleandomycin resistance, polymyxin B and acid production from inositol, l-Rhamnose, d-Tagatose, d-Ribose, putrescine, glycerol, erythritol, d-Arabinose, l-Arabinose, l-Xylose, d-Xylose, methyl-d-Xylopyranoside, l-Sorbose, dulcitol, d-Sorbitol, amygdalin, arbutin, d-Lactose, xylitol, gentiobiose, d-Lyxose, d-Fucose, l-Fucose, d-Arabitol, l-Arabitol, potassium gluconate, potassium 2-ketogluconate, potassium 5-ketogluconate. A total of 91 of phenotypic test performed using the API 50CH, API 20E and Vitek2 BCL system. It was determined that there was a difference between A. salavatliensis DSM 22626T and A. gonensis G2T in only 10 tests and the difference value was 11%.

Table 1 The biochemical characteristics of A. salavatliensis DSM 22626T and A. gonensis G2T
Full size table

In the original articles, the polar lipids of A. salavatliensis DSM 22626T and A. gonensis G2T were not determined (Belduz et al. 2003; Cihan et al. 2011). In the present study, the polar lipids found in A. salavatliensis DSM 22626T were diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylcholine (PC), phosphatidylethanolamine (PE), unidentified phospholipid-1 (PL1), and unidentified phospholipid-2 (PL2), whereas A. gonensis G2T consisted of DPG, PG, PC, PE and PL1. Polar lipid composition showed very similar profile between two species (Fig. 4). The respiratory quinone of A. salavatliensis DSM 22626T and A. gonensis G2T was menaquinone MK-7. Most of the chemotaxonomic and phenotypic features between A. salavatliensis DSM 22626T and A. gonensis G2T were almost identical except for a few exceptions as is shown in Table 1 and Fig. 4. The disagreement for phenotypic and chemotaxonomic was probably due to their different ecological niches.

Fig. 4
figure 4

Two-dimensinal thin-layer chromatogram of polar lipids of A A. salavatliensis DSM 22626T and B A. gonensis G2T. Abbreviations: DPG, diphosphatidylglycerol; PG, phosphatidyeglycerol; PE, phosphatidylethanolamine; PC, phosphatidylcholine; APL, unidentified aminophospolipid; PL, unidentified phospholipid

Full size image

The present results demonstrate the synonym between A. salavatliensis DSM 22626T and A. gonensis G2T. Based on genomic, phylogenetic and chemotaxonomic comparison, we propose that A. salavatliensis DSM 22626T Cihan et al. 2011 should be reclassified as a later heterotypic synonym of A. gonensis G2T Belduz et al. 2003. The type strain is G2T (= NCIMB 13933 T = NCCB 100040 T) and A343 (DSM 22,626 = NCIMB 14,579) is an additional strain of A. gonensis.

Emended description of A. gonensis Belduz et al. (2003)

The description is the same as given by Belduz et al. (2003) with the following modification.

The respiratory quinone is menaquinone MK-7. Major polar lipids include diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylcholine (PC), phosphatidylethanolamine (PE) and unidentified phospholipid-1 (PL1). In API 50CH, API 20E and Vitek2 BCL system, the following activities were positive for tyrosine arylamidase, ONPG hydrolysis, Leucine-arylamidase, nitrate reduction, Ala-Phe-Pro-Arylamidase, gelatinase, phenylalanine arylamidase, β-galactosidase, α-galactosidase, α-glucosidase, acid production from d-Xylose, d-Mannitol, d-glucose, d-Fructose, Methyl-α-d-Glucopyranoside, salicin, d-Cellobiose, d-Maltose, d-Melibiose, d-Sucrose, d-Trehalose, inulin, d-Melezitose, d-Raffinose, starch, glycogen, d-Turanose, methyl-alpha-D-mannopyranoside and maltotriose. Negative for urease, arginine dihydrolase, tryptophan deaminase, citrate utilization, Voges–Proskauer, ornithine decarboxylase, indole production (tryptophanase), β-xylosidase, l-Lysine arylamidase, lysine decarboxylase, l-Aspartate arylamidase, phosphoryl choline, l-Proline arylamidase, ellman, l-Pyrrolydonyl arylamidase, pyruvate, alanine arylamidase, hydrogen sulfide production, β-N-acetyl-glucosaminidase, cyclodextrine, kanamycin resistance, palatinose, β-glucosidase, Methyl-d-Xyloside, α-mannosidase, N-acetyl-glucosamine, β-mannosidase, glycine arylamidase, growth in 6.5% NaCI, oleandomycin resistance, polymyxin B and acid production from inositol, d-Galactose, d-Mannose, esculin ferric citrate, l-Rhamnose, d-Tagatose, d-Ribose, putrescine, glycerol, erythritol, d-Arabinose, l-Arabinose, l-Xylose, d-Xylose, Methyl-d-Xylopyranoside, l-Sorbose, dulcitol, d-Sorbitol, amygdalin, arbutin, d-Lactose, xylitol, gentiobiose, d-Lyxose, d-l-Fucose, d-and l-Arabitol, potassium gluconate, potassium 2-ketogluconate, potassium 5-ketogluconate. The DNA G + C content of the type strain G2T (= NCIMB 13933 T = NCCB 100040 T) is 41.46 mol%.