Cohnella cholangitidis sp. nov., a novel species of the genus Cohnella isolated from a clinical specimen in Korea

A Gram-positive, aerobic, rod-shaped bacterium, designated as strain 1605-214T, was isolated from the blood sample of a patient with cholangitis. Based on its 16S rRNA gene sequence, the strain 1605-214T belonged to the genus Cohnella and exhibited 97.9% sequence identity with Cohnella luojiensis DSM 24270T (GQ214052). DNA–DNA hybridization, digital DNA–DNA hybridization, and average nucleotide identity values between the two species were 23% ± 1.9, 21.1%, and 77.2%, respectively. The cellular fatty acids of strain 1605-214T were mainly comprised of anteiso-C15:0 (36.1%), iso-C16:0 (16.5%), and C16:0 (15.1%). The predominant quinone was menaquinone-7; predominant polar lipids were diphosphatidylglycerol, phosphatidylethanolamine, and aminophospholipid-1. The cell wall peptidoglycan of strain 1605-214T contained meso-diaminopimelic acid. DNA G + C content of strain 1605-214T was 50.6 mol%. 5187 genes out of a total of 5413 (94.6%) were assigned putative functions using eggNOG v5.0. Based on genotypic characteristics and genomic sequence analysis results, strain 1605-214T was confirmed to represent a novel species of genus Cohnella, for which the name Cohnella cholangitidis sp. nov., was proposed. Supplementary Information The online version contains supplementary material available at 10.1007/s00203-021-02565-3.


Introduction
The first species of the genus Cohnella was described as Cohnella thermotolerans in a report by Kämpfer et al (2006). Currently, the genus Cohnella comprises 37 species (LPSN: http:// www. bacte rio. net), including six that have not been validated. Most members of Cohnella were isolated from various environments such as soil (Cai et al. 2010;Kim et al. 2010Kim et al. , 2011, plants (Garcia-Fraile et al. 2008), water (Shiratori et al. 2010), and industrial materials (Kämpfer et al. 2006). Cohnella cellulosilytica (Khianngam et al. 2012) and Cohnella faecalis (Zhu et al. 2019) were isolated from animal excrements. Cohnella hongkongensis (Kämpfer et al. 2006) and Cohnella massiliensis (Abou Abdallah et al. 2019) were isolated from clinical samples (Table S1). In the present study, we have described strain 1605-214 T as a novel species of the genus Cohnella. To our knowledge, this is the

Phenotypic, morphological, and biochemical characterization
Gram staining was performed using Gram Stain Kits (BD), and a catalase test was performed by adding 3% hydrogen peroxide solution to bacteria smeared on slides. The growth conditions for strain 1605-214 T were determined at different pH values (4-10, at pH intervals of 0.5 unit) on BAP. For analysis of its biochemical and enzymatic characteristics, VITEK 2 GP (bioMérieux, France) was used according to the manufacturer's instructions.
To analyze its isoprenoid quinones, the cell biomass of strain 1605-214 T was obtained from cultures grown on BAP for 2 days at 30 °C. Quinones were extracted using the chloroform/methanol method [(C:M, 2:1, v/v)]. The extracted quinones were vacuum-evaporated and re-extracted using n-hexane-water (1:1, v/v). The purified quinones were analyzed using a reverse-phase HPLC system (Younglin, Korea), as described by Hiraishi et al. (1992).
The cellular fatty acid composition of the isolated strain was analyzed according to Miller's method (Miller 1982). Agilent Technologies 6890 Gas Chromatography was performed to analyze the prepared samples, and an A30 m × 0.320 mm × 0.25 μm crosslinked methyl siloxane column (HP-1) was used as a separation column. The profile was analyzed using Sherlock MIS Software. Peak identification, retention time, peak area, and area ratio were determined by comparison with the standard calibration solution.
The diaminopimelic acid in the cell wall was analyzed using a previously described method (Hasegawa et al. 1983).

Genomic DNA preparation and genome sequencing
Genomic DNA was extracted by digestion of the bacteria with proteinase K in 10% SDS, followed by purification using the phenol extraction and ethanol precipitation methods. The primary sequencing library was prepared according to the protocol of the SMRTbell Template Prep Kit 1.0 (Pacific Biosciences, USA). The secondary sequencing library was prepared according to the protocol of the Ion Xpress Plus Fragment Library kit (Thermo Fisher Scientific, USA). The genome was sequenced using PacBio RS II (Pacific Biosciences, USA) and Ion S5 (Thermo Fisher Scientific, USA) sequencing platforms.

Genome assembly
SPAdes Genome Assembler (v3.1) was adopted for de novo assembly sequence reads generated by NGS platforms PacBio RS II and Ion S5, and produced contigs and scaffold sequences. SSPACE program was used for scaffolding contigs and scaffold sequences, and the remaining sequencing errors including gaps and low-quality region were corrected using Proovread (v2.14.0).

Genome annotation
The genome of strain 1605-214 T was initially annotated using the PROKKA (Seemann 2014) software package. The NCBI Prokaryotic Genome Annotation Pipeline (PGAP) (Tatusova et al. 2016) software package was used to generate the final annotation. The predicted protein sequences were classified into functional groups in Clusters of Orthologous Groups (COG) using eggNOG 5.0 (Huerta-Cepas et al. 2019). The resistance genes and virulence factors were identified using AMRFinderPlus (Feldgarden et al. 2019) and VFdb (Liu et al. 2018), respectively.

16S rRNA phylogenetic tree
An initial genomic distance calculation was conducted by searching for the genetically closest strains in EzTaxon Server (Chun et al. 2007) and Type Strain Genome Server (TYGS) (Meier-Kolthoff and Göker 2019). The 16S rRNA sequences of 37 type strains belonging to the Cohnella genus were downloaded from the list of prokaryotic names with standing in nomenclature (LPSN) (Parte 2018). The multiple sequence alignment was processed using MAFFT (Katoh and Standley 2013). Phylogenetic trees were constructed with 1000 bootstrap replicates using the neighborjoining (NJ) method by MEGA7 (Kumar et al. 2016) and the maximum-likelihood (ML) method by RAxML (Stamatakis 2014). Figtree software was used to visualize the trees (http:// tree. bio. ed. ac. uk/ softw are/ figtr ee).

Genomic sequence similarity comparison
Genomic sequence similarity comparison was conducted using the available genomes of the five closest Cohnella species. OrthoANI (Lee et al. 2016) and digital DNA-DNA hybridization (dDDH) (Meier-Kolthoff et al. 2013) were used to compare genome similarities. To calculate the average genomic identity of orthologous gene sequences (AGIOS) (Ramasamy et al. 2014) between genomes, the sets of orthologous proteins were first obtained using BLASTP, with the reciprocal-best-BLAST-hits (RBH) approach (minimal coverage of 50%, amino acid identity of 30%). The mean percentages of nucleotide sequence identity between the orthologous genes were then calculated.

Strain and sequence deposition
Strain 1605-214 T has been deposited in two microbial culture collections: the National Culture Collection for

Phenotype
The optimal conditions of strain 1605-214 T for growth were a temperature of 30 °C and a pH of 7 (Table 1). Based on VITEK 2 GP results, all four Cohnella spp. strains were positive for beta-galactosidase, beta galactopyranosidase, and alpha-galactosidase. In addition, strain 1605-214 T was positive for alpha-glucosidase and negative for D-trehalose, whereas Cohnella luojiensis, which is considered a genetically close species, was negative and positive, respectively, for the above-mentioned enzymes. The major lipid classes of strain 1605-214 T were identified as diphosphatidylglycerol (DPG), phosphatidylethanolamine (PE), and aminophospholipid-1 (APL1). The major quinone present in the strain was identified as MK-7. The cell wall peptidoglycan of strain 1605-214 T contained meso-diaminopimelic acid. The G + C content of strain 1605-214 T was 50.6 mol%, and the major fatty acids were anteiso-C 15:0 (36.1%), iso-C 16:0 (16.5%), and C 16:0 (15.1%) ( Table 2). The relatedness of DNA between strain 1605-214 T and C. luojiensis DSM 24270 T was     23.0% ± 1.9. Based on phenotypic and genotypic distinctness and DNA-DNA hybridization results, strain 1605-214 T was confirmed to be a novel pathogenic species similar to C. luojiensis.

Comparison with genomes of other Cohnella species
At the time of manuscript preparation, the 16S rRNA sequences of the type strains were analyzed as mentioned above; however, a comparison at the whole-genome level was not possible. Therefore, the Cohnella cholangitidis 1605-214 T was further compared to four type strains, including C. luojiensis (Table 3). The four strains were selected based on the results of the dDDH analysis from TYGS. Additionally, average nucleotide identity (ANI) analysis was also performed for the strains. The dDDH and ANI results for the assessed strains were, respectively, as follows: C. luojiensis (

Conclusion
The phenotypic, morphological, and biochemical characterizations, genome perspectives, and comparative genome analyses suggested that strain 1605-214 T represents a novel species of the genus Cohnella for which the name C. cholangitidis is proposed.
Strain 1605-214 T (= NCCP 16833 T , = DSM 112742 T ) was isolated from a clinical specimen at the Gyeongsang National University Hospital in Jinju, Gyeongsangnam-do, South Korea.