Draft genome sequence of Microbacterium oleivorans strain Wellendorf implicates heterotrophic versatility and bioremediation potential.

Microbacterium oleivorans is a predominant member of hydrocarbon-contaminated environments. We here report on the genomic analysis of M. oleivorans strain Wellendorf that was isolated from an indoor door handle. The partial genome of M. oleivorans strain Wellendorf consists of 2,916,870 bp of DNA with 2831 protein-coding genes and 49 RNA genes. The organism appears to be a versatile mesophilic heterotroph potentially capable of hydrolysis a suite of carbohydrates and amino acids. Genomic analysis revealed metabolic versatility with genes involved in the metabolism and transport of glucose, fructose, rhamnose, galactose, xylose, arabinose, alanine, aspartate, asparagine, glutamate, serine, glycine, threonine and cysteine. This is the first detailed analysis of a Microbacterium oleivorans genome.

Introduction

The strain Wellendorf was isolated from a door handle surface with frequent human use in Stillwater, OK as part of the Student Initiated Microbial Discovery (SIMD) project (introduced in [1]). The Microbacterium genus is a phylogenetically and physiologically diverse genus with members ubiquitously found in polycyclic aromatic hydrocarbon (PAH)-contaminated [2], [3], as well as heavy metal-contaminated [4], [5] soils. PAHs and heavy metals are persistent environmental contaminants with both environmental and human health concerns [6], [7], [8]. Genomic analysis of strains belonging to the genus Microbacterium can contribute to our understanding of the molecular mechanisms of PAHs degradation and heavy metal mobilization and could potentially contribute to natural-attenuation-based, and engineered bioremediation schemes in multiple environments [9], [10]. Here we present the draft genomic sequence, and first detailed genomic annotation and analysis of a Microbacterium oleivorans strain.

Materials and methods

Genome project history

The draft assembly and annotation were completed in 2015–2016. Table 1 shows the genome project information.

Table 1

Project information.

MIGS ID	Property	Term
MIGS 31	Finishing quality	Draft
MIGS-28	Libraries used	2 × 300 paired end chemistry
MIGS 29	Sequencing platforms	Illumina Miseq
MIGS 31.2	Fold coverage	300 ×
MIGS 30	Assemblers	Velvet 2.0
MIGS 32	Gene calling method	Prodigal
	GenBank ID	MAYO00000000
	GenBank date of release	July 2016
	GOLD ID	Gp0126761
	BIOPROJECT	PRJNA327390
MIGS 13	Project relevance	Environmental

Growth conditions and genomic DNA preparation

M. oleivorans strain Wellendorf was grown overnight at 30 °C on tryptic soy agar plates. Genomic DNA of high sequencing quality was isolated using the MPBio PowerSoil® DNA extraction kit according to manufacturer's instructions. Negative stain TEM micrographs were obtained using the services of the Oklahoma State University Microscopy Lab. Briefly, the sample was placed on a carbon film TEM grid and allowed to incubate for 2 min, after which the excess liquid was wicked off. Phosphotungestic acid (PTA; 2% w/v) was then added to the grid followed by a 45-s incubation. Excess PTA was blotted off and the grid was allowed to dry before it was visualized using JOEL JEM-2100 transmission electron microscope.

Genome sequencing and assembly

The genome of M. oleivorans strain Wellendorf was sequenced using the Illumina MiSeq platform at the University of Georgia Genomics Facility using 2 × 300 paired end chemistry and an average library insert size of 700 bp. Quality filtered sequence data were assembled with the short read de Bruijn graph assembly program Velvet [11] using a kmer value of 101 bp and a minimum contig coverage value of 7 ×. The genome project is deposited in GOLD (Genomes On-Line Database) and this Whole Genome Shotgun (WGS) project has been deposited in GenBank under the accession MAYO00000000. The version described in this paper is version MAYO01000000.

Genome annotation

Gene models were created using the prokaryotic gene calling software package Prodigal [12]. A total of 2885 gene models were predicted. The average gene size was 961 bp. Translated protein sequences were functionally annotated using a combination of NCBI Blast C ++ homology search and HMMER 3.0 [13] hmmscan against the PFAM 26.0 database [14]. Additional gene analysis and functional annotation were carried out through the Integrated Microbial Genomes Expert Review (IMG-ER) platform.

Phylogenetic analysis

A maximum likelihood phylogenetic tree was constructed using multiple sequence alignments of 16S rRNA genes sequences. Multiple sequence alignment was conducted in Mega, as were the selection of the best substitution model, and the maximum likelihood analysis [15]. The tree was obtained under “TN93 + G + I” model with, a proportion of invariable sites of 0.25, and a variable site γ shape parameter of 0.51. Escherichia coli partial 16S rRNA gene isolate ECSD9 was used as the outgroup. Bootstrap values, in percent, were based on 200 replicates.

Comparative genomics

We sought to compare the genome of Microbacterium oleivorans strain Wellendorf to 17 closely related genomes (IMG genome IDs: 2576861779, 2519899511, 2639762631, 2627854169, 2619619265, 2609459760, 2576861795, 2639762630, 2636415545, 2645728100, 2540341240, 2643221903, 2627854213, 2541047020, 2608642165, 2522572100, and 2526164566) using the “Genome clustering” function on the IMG-ER analysis platform based on the COG profile. We also used principal component analysis to compare the genomes based on several genomic features including the genome size, the number of genes, the number of transporters identified, the GC content, the number of non-coding bases, the number of genes belonging to COG categories, as well as the number of genes belonging to each COG category [16], [17]. The PCA analysis was conducted using the “princomp” function in the labdsv library of R [18]. The results were visualized using a biplot, where genomes were represented by stars and genomic features or COG categories used for comparison were represented by arrows, where the arrow directions follow the maximal abundance, and their lengths are proportional to the maximal rate of change between samples.

Results and discussion

Classification and features

Cells of M. oleivorans strain Wellendorf are Gram positive, non-motile, aerobic irregular rods that were arranged in pairs (Fig. 1). Colonies on TSA agar were orange-red.

Fig. 1

Negative stain TEM micrograph of Microbacterium oleivorans strain Wellendorf.

Within the genus Microbacterium, 94 species are described with validly published names. Strain Wellendorf shares 93.23–100% 16S rRNA gene identity with other species in the Microbacterium genus (Table 2). Compared to other Microbacterium oleivorans strains with sequenced genomes, Strain Wellendorf shares 99% 16S rRNA gene similarity with Microbacterium oleivorans strains CD11_3 (GenBank accession number LSTV00000000) and NBRC 103075 (GenBank accession number BCRG01000000), and 100% similarity to strain RIT293 [19].

Table 2

M. oleivorans strain Wellendorf 16S rRNA gene percentage similarity to other Microbacterium species.

Microbacterium species	Type strain	Wellendorf strain % similarity
M. aerolatum	V-73	98.27%
M. agarici	CC-SBCK-209	94.05%
M. amylolyticum	N5	93.23%
M. aoyamense	KV-492	97.82%
M. aquimaris	JS54-2	97.97%
M. arabinogalactanolyticum	ATCC 51926	97.44%
M. arborescens	ATCC 4358	97.20%
M. arthrosphaerae	CCM 7681	97.48%
M. aurantiacum	ATCC 49090	97.89%
M. aurum	ATCC 51345	97.51%
M. awajiense	YM13-414	97.66%
M. azadirachtae	AI-S262	97.95%
M. binotii	CIP 101303	97.42%
M. chocolatum	BUCSAV 207	97.72%
M. deminutum	KV-483	97.66%
M. dextranolyticum	M-73	97.89%
M. enclense	NIO-1002	97.81%
M. endophyticum	PA15	97.35%
M. esteraromaticum	ATCC 8091	97.51%
M. flavescens	ATCC 13348	98.12%
M. flavum	YM18-098	98.58%
M. fluvii	YSL3-15	97.89%
M. foliorum	P 333/02	98.43%
M. ginsengisoli	Gsoil 259	96.35%
M. ginsengiterrae	DCY37	98.65%
M. gubbeenense	DPC 5286	93.25%
M. halimionae	PA36	97.58%
M. halophilum	N° 76	96.02%
M. halotolerans	YIM 70130	95.63%
M. hatanonis	FCC-01	97.51%
M. hominis	CIP 105731	98.27%
M. humi	CC-12309	94.12%
M. hydrocarbonoxydans	BNP48	98.35%
M. hydrothermale	0704C9-2	97.58%
M. immunditiarum	SK 18	96.50%
M. imperial	ATCC 8365	97.28%
M. indicum	BBH6	94.53%
M. insulae	DS-66	98.12%
M. invictum	DSM 19600	97.51%
M. jejuense	THG-C31	97.20%
M. keratanolyticum	ATCC 35057	98.42%
M. ketosireducens	CIP 105732	97.66%
M. kitamiense	C2	97.81%
M. koreense	JS53-2	97.74%
M. kribbense	MSL-04	95.88%
M. kyungheense	THG-C26	97.82%
M. lacticum	ATCC 8180	98.12%
M. lacus	A5E-52	97.67%
M. laevaniformans	ATCC 15953	97.88%
M. lemovicicum	ViU22	97.74%
M. lindanitolerans	MNA2	93.77%
M. luteolum	ATCC 51474	98.34%
M. luticocti	SC-087B	94.99%
M. mangrove	MUSC 115	96.82%
M. marinilacus	YM11-607	95.80%
M. marinum	H101	98.04%
M. maritypicum	ATCC 19260	98.42%
M. mitrae	M4-8	97.04%
M. murale	01-Gi-001	97.88%
M. nanhaiense	OAct400	93.98%
M. natoriense	TNJL143-2	98.80%
M. neimengense	7087	97.20%
M. oleivorans	BAS69	100%
M. oryzae	MB10	95.64%
M. oxydans	DSM 20578	98.42%
M. paludicola	US15	95.57%
M. panaciterrae	DCY56	97.67%
M. paraoxydans	CF36	98.73%
M. petrolearium	LAM0410	96.81%
M. phyllosphaerae	P 369/06	98.65%
M. populi	10-107-8	94.67%
M. profundi	Shh49	98.12%
M. proteolyticum	RZ36	97.97%
M. pseudoresistens	CC-5209	96.66%
M. pumilum	KV-488	97.74%
M. pygmaeum	KV-490	97.67%
M. radiodurans	GIMN 1.002	97.43%
M. rhizomatis	DCY100	95.17%
M. saccharophilum	K-1	98.04%
M. saperdae	ATCC 19272	98.27%
M. schleiferi	ATCC 51473	98.42%
M. sediminicola	YM10-847	96.81%
M. sediminis	YLB-01	96.27%
M. shaanxiense	CCNWSP60	97.90%
M. soli	DCY 17	95.25%
M. suwonense	M1T8B9	96.27%
M. terrae	ATCC 51476	97.65%
M. terregens	ATCC 13345	97.74%
M. terricola	KV-448	97.74%
M. thalassium	CIP 105728	98.12%
M. trichothecenolyticum	ATCC 51475	97.82%
M. ulmi	XIL02	96.66%
M. xylanilyticum	S3-E	97.05%
M. yannicii	G72	97.89%

Phylogenetic analysis based on the 16S rRNA gene placed strain M. oleivorans BAS69 as the closest taxonomic relative of M. oleivorans strain Wellendorf (Table 3 and Fig. 2).

Table 3

Classification and general features of M. oleivorans strain Wellendorf [30].

MIGS ID	Property	Term	Evidence codea
	Classification	Domain Bacteria	TAS [22]
		Phylum Actinobacteria	TAS [22]
		Class Actinobacteria	TAS [22]
		Order Micrococcales	TAS [22]
		Family Microbacteriaceae	TAS [22]
		Genus Microbacterium	TAS [22]
		Species oleivorans	TAS [22]
		(Type) strain: Wellendorf
	Gram stain	Positive	TAS [22]
	Cell shape	Irregular rods	TAS [22]
	Motility	Non-motile	TAS [22]
	Sporulation	Non-spore forming	TAS [22]
	Temperature range	Mesophile	TAS [22]
	Optimum temperature	30 °C	TAS [22]
	pH range; optimum	Unknown
	Carbon source	l-arabinose, d-cellobiose, d-fructose, d-galactose, gluconate, d-glucose, d-maltose, d-mannose, α-d-melibiose, l-rhamnose, d-ribose, d-sucrose, salicin, d-trehalose, l-xylose, d-mannitol, sorbitol, fumarate, dl-lactate, l-malate, pyruvate, l-aspartate, l-histidine, putrescine and 4-hydroxybenzoate	TAS [22]
MIGS-6	Habitat	Indoor environment, door handle	TAS [22]
MIGS-6.3	Salinity	2–4% NaCl (w/v)	TAS [22]
MIGS-22	Oxygen requirement	Obligate aerobe	TAS [22]
MIGS-15	Biotic relationship	free-living	IDA
MIGS-14	Pathogenicity	Unknown
MIGS-4	Geographic location	USA	IDA
MIGS-5	Sample collection	March 2016	IDA
MIGS-4.1	Latitude	36.1157	IDA
MIGS-4.2	Longitude	− 97.0586	IDA
MIGS-4.4	Altitude	1 M	IDA

Evidence codes - IDA: inferred from direct assay; TAS: traceable author statement (i.e., a direct report exists in the literature); NAS: non-traceable author statement (i.e., not directly observed for the living, isolated sample, but based on a generally accepted property for the species, or anecdotal evidence). These evidence codes are from the Gene Ontology project [31].

Fig. 2

A maximum likelihood phylogenetic tree constructed using multiple sequence alignments of 16S rRNA genes. “Microbacterium oleivorans strain Wellendorf” sequence is shown in bold. GenBank accession numbers are given in parentheses. The tree was obtained under “TN93 + G + I” model with, a proportion of invariable sites of 0.25, and a variable site γ shape parameter of 0.51. The tree was rooted using Escherichia coli partial 16S rRNA gene isolate ECSD9 (not shown). Bootstrap values, in percent, are based on 200 replicates and are shown for branches with > 50% bootstrap support. Multiple sequence alignment, model selection, and maximum likelihood analysis using MEGA [15].

Genome properties

The genome assembly produced a contig N50 of 2,860,671 bp with a total genome size of 2,916,870 bp. The GC content was 69.57%. Forty nine RNA genes were identified in the genome including 4 ribosomal RNA and 45 tRNA genes. The ribosomal RNA operon showed an atypical organization. Of the 2885 detected, 2831 were protein-coding, of which 76.26% had a function prediction, 65.34% represented a COG functional category, and 4.99% were predicted to have a signal peptide. Psort [20] classified proteins as 49.45% cytoplasmic, 0.85% extracellular, and 31.54% associated with the membrane. Based on the presence of 139 single copy genes [21], the genome is predicted to be 77.69% complete. Genome statistics are shown in Table 4. The distribution of genes into COG functional categories is shown in Table 5.

Table 4

Genome statistics.

Attribute	Value	% of Total
Genome size (bp)	2,916,870	100%
DNA coding (bp)	2,726,938	93.49%
DNA G + C (bp)	2,029,207	69.57%
DNA scaffolds	2	100%
Total genes	2885	100%
Protein coding genes	2831	98.13%
RNA genes	54	1.87%
Pseudo genes	0
Genes in internal clusters	527	18.27%
Genes with function prediction	2159	74.84%
Genes assigned to COGs	1889	65.48%
Genes with Pfam domains	2271	78.72%
Genes with signal peptides	144	4.99%
Genes with transmembrane helices	807	27.97%
CRISPR repeats	0

Table 5

Number of genes associated with general COG functional categories.

Code	Value	% age	Description
J	163	7.66%	Translation, ribosomal structure and biogenesis
A	1	0.05%	RNA processing and modification
K	191	8.98%	Transcription
L	96	4.51%	Replication, recombination and repair
B	0	0%	Chromatin structure and dynamics
D	22	1.03%	Cell cycle control, cell division, chromosome partitioning
V	40	1.88%	Defense mechanisms
T	88	4.14%	Signal transduction mechanisms
M	98	4.61%	Cell wall/membrane biogenesis
N	16	0.75%	Cell motility
U	29	1.36%	Intracellular trafficking and secretion
O	82	3.85%	Posttranslational modification, protein turnover, chaperones
C	106	4.98%	Energy production and conversion
G	230	10.81%	Carbohydrate transport and metabolism
E	217	10.2%	Amino acid transport and metabolism
F	76	3.75%	Nucleotide transport and metabolism
H	123	5.78%	Coenzyme transport and metabolism
I	93	4.37%	Lipid transport and metabolism
P	108	5.08%	Inorganic ion transport and metabolism
Q	38	1.79%	Secondary metabolites biosynthesis, transport and catabolism
R	203	9.54%	General function prediction only
S	95	4.46%	Function unknown
–	1000	34.66%	Not in COGs

The total is based on the total number of protein coding genes in the genome.

Insights from the genome sequence

Genome analysis of M. oleivorans strain Wellendorf identified a Gram positive microorganism with an atypical cell wall structure, with genomic evidences of a peptidoglycan layer lacking pentaglycine bridges and with meso-diaminopimelic acid (meso-DAP) as the second amino acid in the peptide linkage. This is different from Microbacterium oleivorans type strain whose cell wall was shown to be devoid of meso-DAP [22]. We identified genes encoding for the biosynthesis of the phosphoglycerolipid CDP-diacyl-glycerol in the genome. The analysis also revealed the absence of flagellar assembly genes and the presence of extracellular structures including Flp and Type IV pilus.

Further genomic analysis identified almost compete to complete catabolic KEGG pathways for each of the following carbon sources; glucose, fructose, rhamnose, galactose, xylose, arabinose, alanine, aspartate, asparagine, glutamate, serine, glycine, threonine and cysteine, and fatty acids as carbon and energy sources. The genome also encodes a complete TCA cycle and electron transport chain with P/V/-type ATPase subunits confirming the aerobic nature of the microorganism. While lactate and acetate fermentation capabilities were also identified in the genome, the facultative nature of this organism was not confirmed in the lab. Genomic analysis suggested auxotrophy for arginine, asparagine, thiamine, ubiquinone and biotin. In agreement with this observation, comparison of the protein-coding genes against the transporter database [23] identified several ABC and secondary transporters that could potentially import these elements.

When compared against the virulence factor database [24], the genome of M. oleivorans strain Wellendorf showed 668 virulence factor hits (19% of the protein-coding genes). These included secretion systems Type I and Type VII, among others.

The Wellendorf genome also encoded several proteins with bioremediation potential. These include enzymes for 4-hydroxyphenylacetate degradation via the meta-cleavage pathway, as well as for detoxification of nitronate [25], a known plant-secreted toxin [26], and of nitriloacetate [27], a chelating agent used in industry and frequently encountered in soil [28]. The genome also encodes for enzymes that can salvage S from organo-S-compounds (e.g. alkanesulfonates) in cases of limiting inorganic S [29].

Insights from comparative genomics

When the genome of M. oleivorans strain Wellendorf was compared to 17 closely related genomes based on their COG profile, the genome clustered with Microbacterium olievorans strain RIT293 (Fig. 3A). A closer look at the COG function profile of M. oleivorans strain Wellendorf in comparison to only Microbacterium oleivorans strains is shown in Table S1. Similarity to M. oleivorans strains at the functional level was in agreement with the phylogenetic position of the isolate as a member of the genus (Fig. 2). We used genomic features including the genome size, the number of genes, the number of transporters identified, the GC content, the number of non-coding bases, the number of genes belonging to COG categories, as well as the number of genes belonging to each COG category to cluster M. oleivorans strain Wellendorf genome in comparison to the 17 other closely related genomes. Results are shown in Fig. 3B. The genome of M. oleivorans strain Wellendorf clustered with the other M. oleivorans genome based on the enrichment in the number of transporters identified in the genomes.

Fig. 3

(A) COG profile clustering of the genomes compared in this study. (B) Principal component analysis biplot based on the genomic features and COG category distribution in the genomes compared. Genomes are represented by stars (strain names are shown). Strain Wellendorf is shown in blue. Arrows represent genomic features or COG categories used for comparison. The arrow directions follow the maximal abundance, and their lengths are proportional to the maximal rate of change between genomes. The first two components explained 75% of variation.

Conclusions

This study presents the genome sequence and annotation of Microbacterium oleivorans strain Wellendorf. The genome revealed an extensive sugar and amino acid degradation machinery (for glucose, fructose, rhamnose, galactose, xylose, arabinose, alanine, aspartate, asparagine, glutamate, serine, glycine, threonine and cysteine). Comparison to the virulence factor database identified 668 genes in the genome with potential virulence-associated function including type Type I, and Type VII secretion systems. The genome also suggests the capability of degradation of fatty acid and the detoxification of several environmental contaminants including phenylacetate, nitronate, and nitriloacetate. Comparative genomics using general genomic features as well as the COG function profile coincided with the phylogenetic position predicted based on the 16S rRNA gene sequence and clustered the strain Wellendorf with another representative of the M. oleivorans species.

The following are the supplementary data related to this article.

Table S1

Comparison of the COG function profile of strain Wellendorf and two other M. oleivorans strain. Only COG families with a representative in at least one of the three genomes are shown. Pearson correlations based on the abundances of the different COG families in the three genomes are shown below the table for all possible pairwise compaisons.

Transparency document

Transparency document.

Competing interests

All authors declare no competing interests.

Authors' contributions

APA, ELH, CL, MBC, and NY contributed to the analysis. APA, WDH, DPF, and NY wrote the manuscript. RW, CB, and RAH performed the lab experiments.