Genome sequencing and description of Oerskovia enterophila VJag, an agar- and cellulose-degrading bacterium

A nonmotile, Gram-positive bacterium that shows an elongated and branching cell shape was isolated from soil samples from the botanical garden of Ulm University, Ulm, Germany. Here, the isolation procedure, identification, genome sequencing and metabolic features of the strain are described. Phylogenetic analysis allowed to identify the isolated strain as Oerskovia enterophila. The genus Oerskovia belongs to the family Cellulomonadaceae within the order Actinomycetales. The length of cells of O. enterophila ranges from 1 μm to 15 μm, depending on the growth phase. In the exponential growth phase, cells show an elongated and branching shape, whereas cells break up to round or coccoid elements in the stationary growth phase. The 4,535,074 bp long genome consists of 85 contigs with 3918 protein-coding genes and 57 RNA genes. The isolated strain was shown to degrade numerous complex carbon sources such as cellulose, chitin, and starch, which can be found ubiquitously in nature. Moreover, analysis of the genomic sequence revealed the genetic potential to degrade these compounds. Electronic supplementary material The online version of this article (doi:10.1186/s40793-017-0244-4) contains supplementary material, which is available to authorized users.


Introduction
Oerskovia enterophila was formerly characterized as Promicromonospora enterophila by Jàger et al. in 1983 [1]. Later, P. enterophila was re-classified as O. enterophila by Stackebrandt et al. [2], since only spore-like elements and no real spores are formed. Furthermore, a phylogenetic tree based on the 16S rRNA gene sequences of strains of the genera Cellulomonas and Promicromonospora shows that O. enterophila did not cluster with the type species of Promicromonospora, Promicromonospora citrea, or Promicromonospora sukumoe [2,3]. The genus Oerskovia was initially described in 1970 by Prauser et al. [4] and harbors currently four species with O. turbata as type species [2]. Bacteria of the genus Oerskovia belong to the phylum Actinobacteria, which is one of the largest taxonomic units among the domain Bacteria [5]. Bacteria belonging to Actinobacteria show a wide range of G + C-content, from 51% to more than 70% [5][6][7]. Actinobacteria are widely distributed in terrestrial as well as in aquatic habitats [8,9]. In general, members of the class Actinobacteria show a high morphological variety, which is also true for species of the genera Oerskovia and Cellulomonas [10]. Furthermore, members of the family Cellulomonadaceae are known for their ability to decompose plant-derived biopolymers such as starch, cellulose or chitin [11]. Due to the close relationship of members of the genera Oerskovia and Cellulomonas [12,13] it is likely that both share genetic features enabling them to degrade these biopolymers. To investigate the genetic potential for biopolymer degradation, the genome of the isolate was sequenced. Furthermore, a genome wide comparison of the isolated strain with other Oerskovia type strains was performed. Additionally, the isolated strain was aerobically grown on respective carbon sources to validate the functionality of the proposed degradation pathways.
In this contribution, the classification, the metabolic features, and the genome insights of the isolated strain are provided.

Classification and features
The isolated strains were identified as Oerskovia enterophila based on 16S rRNA gene sequence identities of more than 99% compared to the type strain of O. enterophila DSM 43852 [14]. All subsequent analyses were performed using the strain designated as O. enterophila VJag. Information regarding the enrichment and isolation procedures as well as identification of Oerskovia strains are described in the Additional files 1 and 2: S1 and S2.
Investigations of the cell morphology of the isolated strain O. enterophila VJag (Table 1) using scanning electron microscopy revealed that cells show different morphologies in exponential and stationary growth stage. In the exponential growth phase, cells show extensive branches with an overall length up to 15 μm, whereas the cells are smaller and less branched in the stationary growth phase (Fig. 1). These different cell morphologies were also previously observed by Stackebrandt et al. [2].
The 16S rRNA gene sequence (OJAG_11220, LRIE01000058.1) of O. enterophila VJag was blasted [15] and used for subsequent phylogenetic analysis. Therefore, 16S rRNA reference sequences of 17 closely related type strains were aligned using MAFFT version 7.215 [16,17] and was performed using EMBL-EBI web services. The length of the 17 references ranged from 1395 to 1612 bp and had average length of 1486 bp. The phylogenetic tree was reconstructed using the software MrBayes version 3.2.6 [18]. The recommended settings in the manual for tree reconstruction use a generalized time reversible evolutionary model. The quick start instructions were followed to run Bayesian phylogenetic analysis. The run was stopped since the standard deviation of split frequencies was below 0.0042 after 1,000,000 generations. Phylum: ' Actinobacteria' TAS [5] Class: Actinobacteria TAS [12] Order: Actinomycetales TAS [40][41][42] Family: Cellulomonadaceae TAS [11,19] Genus: Oerskovia TAS [4] Species: Oerskovia enterophila TAS [1,2] Strain: VJag (LRIE00000000) TAS [5,14] Gram Evidence code -IDA: Inferred from Direct Assay; TAS: Traceable Author Statement; These evidence codes are from the Gene Ontology project [43]. If the evidence is IDA, then the property was directly observed for a live isolate by one of the authors The resulting phylogenetic tree is shown in Fig. 2. Described species of the genera Oerskovia and Cellulomonas belong to the same family of Cellulomonadaceae. On the other hand, Sanguibacter belongs to the family of Sanguibacteriaceae which is defined as a neighboring group to Cellulomonadaceae [19]. Sanguibacter is the only described genus within the respective family with currently six species [20][21][22][23][24].

Genome project history
The genome of O. enterophila VJag was sequenced to get insights in the genomic features and the metabolic potential of this strain. Furthermore, no genomes of members of this species were available at the time of writing. A draft sequence is available at NCBI for the species O. turbata NRRL B-8019 (JOFV00000000) [25]. The complete genome of O. enterophila VJag has a size of 4,535,074 bp and consists of 85 contigs. In this contribution the version LRIE01000000 is described. The genome sequencing and gene annotation was performed by Goettingen Genomics Laboratory (Germany). The sequence can be found under the accession number LRIE00000000. Table 2 shows the project information according to MIGS specification [26].

Growth conditions and genomic DNA preparation
O. enterophila VJag was cultivated in 5 ml TSYEmedium (medium 92, DSMZ) at 28°C overnight in an orbital shaker at 120 rpm for the isolation of genomic DNA. Genomic DNA was isolated using MasterPure Gram positive DNA Purification kit (Epicentre, Madison, WI, USA) according to the manufacturer's instructions. DNA concentrations and purity were analyzed using the UV-Vis spectrophotometer NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA). The genomic DNA yield was 2463 ng/μl. The DNA purity was determined using the UV absorbance ratio 260/208 nm and 260/230 nm and revealed ratios of 2.01 and 2.17, respectively.  The phylogenetic tree was created using MrBayes [18] version 3.2.6, sequences were aligned using MAFFT [16,17]. Numbers at the nodes present the posterior probability Genome sequencing and assembly A combined approach was used for the whole-genome sequencing of O. enterophila VJag using the 454 GS-FLX TitaniumXL system (titanium GS70 chemistry, Roche Life Science, Mannheim, Germany) and the Genome Analyzer II (Illumina, San Diego, CA). According to the manufacturer's protocols, the shotgun libraries were prepared, which resulted in 97,681 reads for 454 shotgun sequencing (11.46 × coverage) and 4,756,630 112-bp paired end Illumina reads (68.28 × coverage). Illumina reads were trimmed using Trimmomatic 0.32 [27] to remove sequences with quality scores lower than 20 (Illumina 1.9 encoding) and remaining adaptor sequences, respectively. The initial hybrid de novo assembly was performed using the MIRA 3.4 [28] and Newbler 2.9 (Roche Life Science, Mannheim, Germany) software. The final assembly resulted in 85 contigs with an average coverage of 79.60, an N50 value of 96,617 bp and an N90 value of 28,097 bp, respectively.

Genome annotation
The Prodigal software tool [29] was used for automatic gene prediction [29], rRNA and tRNA gene identification was performed using RNAmmer [30] and tRNAscan [31], respectively. The automatic gene-annotation was performed by using the IMG-ER system [32,33]. The annotation was manually curated using the Swiss-Prot, TrEMBL, and InterPro databases [34].

Genome properties
The genome of O. enterophila VJag is 4,535,074 bp in length and has an average G + C content of 72.4% (Fig. 3). The genome sequence shows 3975 genes in total, 3918 are protein-coding genes, 57 are RNA genes, of which 6 code for rRNA. The remaining genes code for proteins with unknown function or hypothetical proteins. All statistics and properties are listed in Table 3, the number of protein-coding genes associated with general COG functional categories is shown in Table 4.
A circular representation of the O. enterophila VJag genome sequence and comparison to O. enterophila DFA-19 T [14] and O. turbata NRRL B-8019 genome sequences is shown in Fig. 3.  Fig. 3). The two inner most plots represent the GC content and the GC skew (circle 7-8). Furthermore, a pairwise ANI analysis of the VJag strain and type strain O. enterophila DFA-19 [14] showed a similarity value of 99.36%, whereas a respective analysis of VJag strain and O. turbata NRRL B-8019 resulted in 89.31% similarity.

Insights from the genome sequence
Because of the close relationship to members of the genus Cellulomonas, O. enterophila VJag was expected to use cellulose as carbon source. According to the KEGG pathway, genes coding for enzymes probably responsible for the degradation of cellulose to cellobiose and β-D-glucose were found in O. enterophila VJag. Cellulose is one of the main components of plant material and is one of the most abundant biopolymers in the environment [35]. Plate assays revealed that O. enterophila VJag is able to utilize cellulose [Additional file 3: Figure S1]. The used plates contained CMC as sole carbon source and Congo red to stain CMC. O. enterophila VJag hydrolyzed CMC to glucose whereby the Congo red was eluted, the red color got lost and resulted in formation of bright halos around cell spots.
Starch is also ubiquitous in nature as it accumulates in plants as storage compound [36]. The genome sequence of O. enterophila VJag harbors genes coding for αamylases (OJAG_12050; OJAG_09450) and a starch phosphorylase (OJAG_12070). Thus, starch is either converted to glycogen, dextrin, or amylose by O. enterophila VJag. Starch or glycogen could also be degraded to trehalose by respective enzymes (glycogen debranching enzyme encoded by OJAG_00790 or OJAG_12120). Subsequently, trehalose would be further converted to β-D-glucose-1-phosphate or D-glucose via an αtrehalose phosphorylase (encoded by OJAG_12210). Dextrin would be converted to α-D-glucose by an oligo-1,6-glucosidase (encoded by OJAG_08510). A plate assay using Jag-MM-agar plates containing starch (2% w/v) as carbon source showed that starch is utilized during  Figure S2]. After incubation, starch was stained using Lugol's solution and bright halos around cell spots showed starch consumption by O. enterophila VJag (see Additional file 3: Figure S2).
Additionally, genes encoding enzymes for xylose degradation were found in the O. enterophila VJag genome sequence. D-xylose could be converted to D-xylulose by a xylose isomerase (encoded by OJAG_26770). Furthermore, D-xylulose would be phosphorylated to D-xylulose-5-phosphate via a xylulokinase (OJAG_26780). D-xylulose-5phosphate would be converted to D-ribulose-5-phosphate by a ribulose-5-phosphate 3-epimerase (OJAG_00210), and then metabolized via the pentose phosphate pathway, or D-xylulose-5-phosphate would be converted to L-ribulose -5-posphate via a L-ribulose-5-phosphate 4-epimerase (OJAG_27380). This also fits into the overall picture since xylose is a main part of hemicellulose and makes up a part of plant materials [38].

Conclusions
The genome of O. enterophila VJag, which was isolated from forest soil, is described. Furthermore, the phylogenetic  The total is based on the total number of protein coding genes in the genome