Functional Analysis of Methylomonas sp. DH-1 Genome as a Promising Biocatalyst for Bioconversion of Methane to Valuable Chemicals

Abstract

Methylomonas sp. DH-1, newly isolated from the activated sludge of a brewery plant, has been used as a promising biocatalytic platform for the conversion of methane to value-added chemicals. Methylomonas sp. DH-1 can efficiently convert methane and propane into methanol and acetone with a specific productivity of 4.31 and 0.14 mmol/g cell/h, the highest values ever reported, respectively. Here, we present the complete genome sequence of Methylomonas sp. DH-1 which consists of a 4.86 Mb chromosome and a 278 kb plasmid. The existence of a set of genes related to one-carbon metabolism and various secondary metabolite biosynthetic pathways including carotenoid pathways were identified. Interestingly, Methylomonas sp. DH-1 possesses not only the genes of the ribulose monophosphate cycle for type I methanotrophs but also the genes of the serine cycle for type II. Methylomonas sp. DH-1 accumulated 80 mM succinate from methane under aerobic conditions, because DH-1 has 2-oxoglutarate dehydrogenase activity and the ability to operate the full TCA cycle. Availability of the complete genome sequence of Methylomonas sp. DH-1 enables further investigations on the metabolic engineering of this strain for the production of value-added chemicals from methane.

1. Introduction

Methane is the principal component of natural/shale gas and biogas, and recently, has attracted much attention as a chemical feedstock [1]. The chemical conversions of methane to other chemicals generally require the input of high amounts of energy because of the high stability of the carbon–hydrogen bond (C–H bond), while the biological conversion of methane to chemicals using methanotrophs can be conducted in ambient conditions [2]. In addition, the bioconversion of methane showed higher conversion of up to 75% [1,3].

In an effort to develop a biocatalytic platform strain, we have isolated a type I methanotroph, Methylomonas sp. DH-1, from the activated sludge of a brewery plant [4]. Methylomonas sp. DH-1 was reported as a highly efficient biocatalyst for the bioconversion of methane to methanol, which can be directly used as alternative fuels, antifreeze and as a precursor to other compounds, with a specific productivity of 4.31 mmol/g cell/h [4]. Furthermore, Methylomonas sp. DH-1 has high potential for methanol production due to its high tolerance to methanol of up to 7% (v/v) [4]. Methylomonas sp. DH-1 has also been evaluated for its catalytic capability to convertpropane to acetone, which is used as an industrial solvent for polymers, in acetylene storage and in the pharmaceutical industry [5]. Moreover, the accumulation of acetone in the absence of chemical inhibitors is advantageous for biocatalytic gas-to-liquid conversion technology. Additionally, Methylomonas sp. DH-1 can produce yellow-to-red pigments which are expected to be numerous carotenoids (unpublished report). Thus, Methylomonas sp. DH-1 can be an important biocatalyst for methane bioconversion to chemicals/fuels. In this study, we sequenced, assembled and annotated the whole genome sequence of Methylomonas sp. DH-1 as the first step for the development of a methanotrophic platform strain. Methylomonas sp. DH-1 was also used for the production of succinate as a model compound from methane.

2. Results

2.1. Genome Statistics and General Features

The complete genome of Methylomonas sp. DH-1 consists of a circular chromosome of 4,849,532 bp (56.5% G + C) and a plasmid of 277,875 bp (51.7% G + C) (Figure 1).

Methylomonas sp. DH-1 was shown to be phylogenetically closely related to Methylomonas koyamae Fw12E-Y^T based on 16S sequence similarity [4]. Electronic DNA–DNA hybridization (DDH) estimate between DH-1 and Fw12E-Y^T (=JCM 16701^T), calculated by the Genome-to-Genome Distance Calculator (http://ggdc.dsmz.de/distcalc2.php), was 73.9%, which suggests that DH-1 belongs to the Methylomonas koyamae species, while average nucleotide identity (ANI) between these two strains was calculated to be 97.76% using JSpecies [6]. Moreover, MUMMER whole-genome alignment [7] between DH-1 and Fw12E-Y^T showed the close similarity of these two strains, aligning 298 out of 382 scaffolds of the Fw12E-Y^T genome assembly (96.68% of the total length) on the DH-1 reference genome sequence (Figure 2A).

As of February 2018, there are 17 publicly available genome sequences in the genus Methylomonas (four at the ‘Complete Genome’ assembly level), including three strain types: M. denitrificans FJG1^T, M. koyamae Fw12E-Y^T, and, M. methanica NCIMB 11130^T. The three strains Methylomonas sp. LW13, Methylomonas sp. MK1, and Methylomonas sp. 11b were all isolated from a single geographical location [8]. The seven genome sequences of M. methanica (NCIMB 11130^T, R-45363, and R-45371), M. koyamae (R-45378, R-45383, and R-49807), and M. lenta (R-45370), isolated from different terrestrial ecosystems, were reported by a single research group [9]. The genome sizes range from 4.70 Mb (M. lenta R-45370, 171 contigs) to 5.48 Mb (M. methanica R-45371, 120 contigs).

The ANI-based genome analysis clustered 17 strains in eleven species, the M. koyamae group being the largest one, accommodating strains Fw12E-Y^T, DH-1, LM6, and R-49807 (Figure 2B,C). Multiple non-type strains originally labeled as M. koyamae or M. methanica were classified into separate groups, which implies that repositioning is required for these strains. It was found that two strain types, M. methanica and M. denitrificans, appear to form a conspecific group (100.0% ANI; 92.3% DDH estimate), while 16S rRNA sequence similarity is at 98.9%. The complete genome sequences of DH-1 and LM6 could be aligned with each other collinearly without any indication of gross rearrangement of insertion/deletion both in chromosome and in plasmid (data not shown).

The genome annotation predicted 4669 protein-coding genes, 47 tRNA and 9 rRNA (

Table 1. The general genome features of Methylomonas sp. DH-1.

Feature	Chromosome	Plasmid 1
Size (bp)	4,849,532	277,875
G + C content (%)	56.47	51.66
Protein coding genes 2	4441	228
Pseudogenes	85	13
tRNAs	47	0
rRNAs	3, 3, 3 (16S, 23S, 5S)	0
ncRNAs	4	0
CRISPR arrays	4	0
GenBank accession	CP014360	CP014361

¹ For the prediction of plasmidic genes, only Prokaryotic Genome Annotation Pipeline (PGAP) annotation was accepted without integration of multiple predictions. ² Including pseudogenes.

). Furthermore, there were 3638 genes assigned to different function categories based on the clusters of orthologous genes (COG) designation (

Table 2. The COG function classification of the Methylomonas sp. DH-1 genome.

Category	Functional Classification	Chromosome	Plasmid
A	RNA processing and modification	1	1
B	Chromatin structure and dynamics	2	0
C	Energy production and conversion	200	1
D	Cell cycle control, cell division, chromosome partitioning	53	4
E	Amino acid transport and metabolism	191	3
F	Nucleotide transport and metabolism	58	0
G	Carbohydrate transport and metabolism	111	0
H	Coenzyme transport and metabolism	157	1
I	Lipid transport and metabolism	73	0
J	Translation, ribosomal structure and biogenesis	172	0
K	Transcription	183	12
L	Replication, recombination and repair	242	32
M	Cell wall/membrane/envelope biogenesis	248	13
N	Cell motility	123	0
O	Posttranslational modification, protein turnover, chaperones	162	5
P	Inorganic ion transport and metabolism	223	9
Q	Secondary metabolites biosynthesis, transport and catabolism	63	2
	General function prediction only	366	15
S	Function unknown	329	10
T	Signal transduction mechanisms	340	9
U	Intracellular trafficking, secretion, and vesicular transport	127	9
V	Defense mechanisms	83	5

) [10]. The most abundant COG category was “General function prediction only” (381 CDSs), followed by “Signal transduction mechanisms” (349 CDSs). A single gigantic gene (AYM39_10365, 32.6 kb) was found to encode a hypothetical transmembrane protein with repetitive domains which have Ca²⁺ and carbohydrate binding property. Eight contigs from Fw12E-Y^T were aligned consecutively with this sequence and complete gene sequences were found from strains LM6, R-49807, and R-45378, while no homologous sequence could be found from other genomes, which suggests that this is a common characteristic of the Methylomonas koyamae species that could increase bacterial fitness under specific environmental niches [11].

All genes required for a type I methanotrophic lifestyle were identified. One functional operon encoding particulate methane monooxygenase (pMMO, pmoCAB), and the pxm operon (pxmABC), encoding the copper membrane monooxygenase [12] was determined in the DH-1 genome. All genes for carbon fixation via the ribulose monophosphate pathway were predicted. Genes encoding PQQ-dependent methanol dehydrogenases (mxaFJGIRSACKLDEK) along with the PQQ biosynthesis gene cluster (pqqBCDE) for methanol oxidation were detected. The tetrahydrofolate (H4F)- and tetrahydromethanopterin (H4MPT)-mediated formaldehyde oxidation pathways and formate dehydrogenase were encoded. Notably, the genome of DH-1 possesses two types of the gene cluster encoding 3-hexulose-6-phosphate synthase (hps) and 6-phospho-3-hexuloisomerase (phi) including a hps-phi operon and another hpsi gene encoding an hps-phi fused protein [13].

2.2. Functional Analysis of the Complete Genome Sequence of Methylomonas sp. DH-1 and the Production of Succinate from Methane

The existence of the complete Embden–Meyerhof–Parnas (EMP) pathway, the pentose-phosphate pathways (PPPs), and the Entner–Doudoroff pathway (EDD) along with a complete TCA cycle were confirmed. Interestingly, a complete set of genes for the serine cycle was identified together with the gene encoding phosphoenolpyruvate carboxylase (ppc) which plays a key role in the serine cycle by converting phosphoenolpyruvate (PEP) to oxaloacetate with the addition of CO₂. Some type I methanotrophs have been predicted to have a partial serine cycle due to the absence of ppc [8,14]. The existence of PEP carboxylase together with pyruvate carboxylase and acetyl-CoA carboxylase indicates that DH-1 possesses more potential for CO₂ fixation compared to other type I methanotrophs. The ability to convert PEP to oxaloacetate, a key intermediate in the TCA cycle, is also advantageous in the production of TCA-derived products such as succinic acid. Additionally, most type I methanotrophs have no remarkable accumulation of succinate in aerobic conditions because 2-oxoglutarate cannot be converted to succinyl-CoA due to poor activity of 2-oxoglutarate dehydrogenase. It forms an incomplete “horseshoe” shaped TCA cycle [15], and consequently succinate could not be accumulated in aerobic conditions. Unusually, Methylomicoribum buryatense, Type I methanotroph, accumulated a detectable amount of succinate in aerobic conditions, because it has three different genes for succinate synthesis [16]. Even though DH-1 has only a TCA cycle as a succinate generation pathway, a large amount of succinate of up to 80 mM was successfully accumulated under aerobic growth conditions, indicating that DH-1 has 2-oxoglutarate dehydrogenase activity and the ability to operate a full TCA cycle (Figure 3).

The existence of a set of genes related to various secondary metabolite biosynthesis pathways via the methylerythritol 4-phosphate (MEP) pathway, including isoprenoid and carotenoid pathways, was confirmed (Figure 4). Notably, the DH-1 genome contains two genes encoding 1-Deoxy-D-xylulose 5-phosphate synthase catalyzing the first step of the MEP pathway. However, the carotenoids biosynthesis pathway in DH-1 has not been fully discovered. From the genome mining analysis, squalene/phytoene synthase (sqs) which is committed in the carotenoid synthesis pathway and the gene cluster related to 4,4′-diapolycopene biosynthesis including diapolycopene oxygenase (crtP), phytoene desaturase (crtI) and aldehyde dehydrogenase (ald) were identified in DH-1. Other potential secondary metabolites that can be synthesized by Methylomonas sp. DH-1 were identified using antiSMASH [17]. The results indicated that eight possible gene clusters encoding secondary metabolites were identified in Methylomonas sp. DH-1 including aryl polyene, bacteriocins, terpene, hserlactone, and T1pks-Nrps. Among them, aryl polyene can protect the bacterium from reactive oxygen species, similar to the functionality of carotenoids [18].

2.3. Nucleotide Sequence Accession Number

The completed genome sequence of Methylomonas sp. DH-1 was deposited at GenBank under accession number CP014360 and CP014361. In addition, the strain was deposited at the Korean Collection for Type Culture under the KCTC number 13004BP.

3. Conclusions

We sequenced and analyzed the whole genome of a newly isolated type I methanotroph, Methylomonas sp. DH-1 consisting of a 4.86 Mb chromosome and a 278 kb plasmid. Methylomonas sp. DH-1 accumulated a large amount of succinate (up to 80 mM) under aerobic conditions most probably due to 2-oxoglutarate dehydrogenase activity, showing its biocatalytic potential for methane bioconversion. The existence of PEP carboxylase, pyruvate carboxylase and acetyl-CoA carboxylase can enable the DH-1 strain to fix CO₂ more efficiently compared to other type I methanotrophs. A set of genes related to various secondary metabolite biosynthesis pathways via the MEP pathway was also identified. The availability of a complete genome sequence of Methylomonas sp. DH-1 contributes to a system-level understanding of methanotrophic metabolism which provides valuable resources for metabolic engineering of this strain for overproduction of value-added chemicals from methane.

4. Materials and Methods

4.1. Bacterial Growth, DNA Isolation, Genome Assembly and Annotation

Methylomonas sp. DH-1 was isolated from the activated sludge of a brewery plant based in a nitrate mineral salts (NMS) medium using enrichment culture with methane as a sole carbon source as described by Hur et al. [4]. Liquid pre-cultures were grown in a 180 mL baffled-flask with a 10 mL NMS medium containing 10 μM CuSO₄ with a supplement of 30% methane (v/v) as a sole carbon source at 30 °C and 230 rpm, sealed with a screw cap. The pre-cultures were then inoculated into 50 mL of fresh medium in a 500 mL baffled-flask for large-scale cultivation.

The genomic DNA was extracted using a Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA). The library construction and sequencing were carried out at NICEM, Seoul, Republic of Korea. The general genome properties were first obtained using Illumina HiSeq 2500 platform-based draft genome sequencing (2 × 151 nt; 1.6 Gb) and then, the PacBio RS II platform was used to obtain the complete genome sequence, which was polished and verified using the previously generated Illumina reads. The 807.6 Mb filtered polymerase reads, produced from the PacBio RS II sequencing using P6-C4 chemistry with 119-fold average coverage was assembled into two contigs using the hierarchical genome assembly process RS_HGAP.3 [19]. The identification of sequence overlap at both ends and the alignments with the Illumina assemblies revealed their circular structures. The genome annotation was performed by integrating results from Prokaryotic Genome Annotation Pipeline (PGAP) (http://www.ncbi.nlm.nih.gov/genome/annotation_prok/), Integrated Microbial Genomes (IMGs) (http://jgi.doe.gov/data-and-tools/img/), Rapid Annotation using Subsystem Technology (RAST) (http://rast.nmpdr.org/) and PROKKA [20] on the basis of stop codons. The priority for choosing functional annotation was in the order of PGAP, IMG, PROKKA, and RAST (Tables S1 and S2). The genome sequence information for comparative and phylogenomic analyses was downloaded from the RefSeq database. The ANI-based genome comparison and clustering were done using DREP [21]. The universal prokaryotic marker gene sequences, identified using the PHYLOSIFT [22], were concatenated into one and an approximately maximum-likelihood tree was constructed using the FASTTREE 2 [23].

4.2. Analytical Methods

The supernatant of cultures was separated by centrifugation. The succinate was quantified using a HPLC equipped with an Aminex HPX-87 column (Bio-Rad, Hercules, CA, USA) and a refractive index detector. Sulfuric acid to the amount of 0.005 M was used as the mobile phase with a flow rate of 0.7 mL/min at 60 °C. The bioreactor off-gas was connected to the GC (Agilent 7890A, Santa Clara, CA, USA) and the methane composition was analyzed by a GC equipped with a TCD detector.