Genome features of a novel hydrocarbonoclastic Chryseobacterium oranimense strain and its comparison to bacterial oil-degraders and to other C. oranimense strains

Abstract

For the first time, we report the whole genome sequence of a hydrocarbonoclastic Chryseobacterium oranimense strain isolated from Trinidad and Tobago (COTT) and its genes involved in the biotransformation of hydrocarbons and xenobiotics through functional annotation. The assembly consisted of 11 contigs with 2,794 predicted protein-coding genes which included a diverse group of gene families involved in aliphatic and polycyclic hydrocarbon degradation. Comparative genomic analyses with 18 crude-oil degrading bacteria in addition to two C. oranimense strains not associated with oil were carried out. The data revealed important differences in terms of annotated genes involved in the hydrocarbon degradation process that may explain the molecular mechanisms of hydrocarbon and xenobiotic biotransformation. Notably, many gene families were expanded to explain COTT’s competitive ability to manage habitat-specific stressors. Gene-based evidence of the metabolic potential of COTT supports the application of indigenous microbes for the remediation of polluted terrestrial environments and provides a genomic resource for improving our understanding of how to optimize these characteristics for more effective bioremediation.

1. Introduction

The Chryseobacterium genus consists of gram-negative bacteria assigned to the family Weeksellaceae, phylum Bacteroidetes,¹ of which there are 160 species and two sub-species (http://www.bacterio.net/). Member species of this genus are adapted to survive in a range of ecosystems and three Chryseobacterium species have been reported as survivalists in oil-polluted soil, C. hungaricum,^2,3C. indoltheticum⁴ and C. nepalense.⁵

Recently and for the first time, the isolation and identification of a crude oil-degrading C. oranimense strain COTT from chronically polluted soil in Trinidad was reported.⁶ The novelty of the Trinidad C. oranimense strain COTT, compared to other C. oranimense strains, lies in the long-term exposure and evolution of enhanced capabilities for biotransformation of petroleum in an extreme terrestrial environment; one that has been chronically polluted with crude oil as a result of natural oil seeps that have been in existence for more than 100 years.⁷ Given the short generation times of some microbes that allow for up to tens of generations of evolution daily, indigenous microbes can exhibit selective enrichment and undergo genetic modifications, which enable higher degradation rates.⁸ Additionally, artificially introduced microbes have difficulty competing with the pre-existing/indigenous microbial community that have evolved to survive in a particular habitat.⁹

While it is expected that whole genome sequencing of different Chryseobacterium strains will reveal both niche-specific and convergent genome features, the COTT strain is hydrocarbonoclastic and it is not known what genomic traits support specific strategies employed in the metabolism and catabolism of hydrocarbons, as well as other xenobiotics.⁹ Additionally, to what degree these unique features are required to sustain long-term survival in naturally occurring oil seeps is also unknown.

The main objective of this study was to characterize the genome of the C. oranimense hydrocarbonoclastic strain, COTT. We sequenced, assembled, annotated and characterized the genome of COTT. An inventory of unique genes of the COTT genome compared to the high-quality genomes of 18 other crude oil-degrading bacterial species, and two other strains of C. oranimense was also generated. The findings of this study extend the current knowledge about the genome sequence diversity of the Chryseobacterium genus as well as intra-specific genome variation of C. oranimense. Genome analysis gave new insights into this strain’s capacity to degrade petrogenic and xenobiotic pollutants, in addition to some of its unique mechanisms for coping with different niche-specific stressors. We anticipate that this genome data will additionally provide an invaluable contribution towards the discovery of industrially important biocatalysts.

2. Materials and methods

2.1. Strain isolation and petroleum utilization

Strain COTT was initially isolated and stored from soil chronically contaminated with crude oil from an abandoned well in the Fyzabad Forest Reserve, south Trinidad. Details of the site, strain isolation, identification, characterization of crude oil utilization (2% crude oil bioassay) and initial identification can be found in Ramdass and Rampersad.⁶

2.2. Phylogenetic analysis

The bacterial Phylogenetic Tree Building Service on the BV-BRC (Bacterial and Viral Bioinformatics Resource Center) server (https://www.bv-brc.org/) was used for the construction of a custom phylogenetic tree built from selected genomes which included that of C. oranimense and 19 other Chryseobacterium species (Additional file 2: Supplementary Table S1). This Codon Tree method selects single-copy BV-BRC PGFams [both the protein (amino acid) and gene (nucleotide) sequences are used for each of the selected genes from the PGFams] and analyzes aligned proteins and coding DNA from single-copy genes using the program RAxML1A. Protein sequences are aligned using MUSCLE¹⁰ and the nucleotide coding gene sequences are aligned using the Codon_align function of BioPython.¹¹ Support values are generated using 100 rounds of the ‘Rapid’ bootstrapping option¹² of RaxML. The protein sequences of 1,000 genes aligned in MAFFT were used to produce the final tree and included 949,272 nucleotides and 316,424 amino acids.

2.3. Growth and DNA isolation

COTT was grown in Reasoner’s 2A agar (R2A) agar (HiMedia Laboratories LLC., West Chester, PA, USA) in the dark at 25°C for 24 to 48 h. The overnight culture was treated with 500–700 μl of TE buffer (10 mM Tris HCl, 1 mM EDTA, pH8; Sigma-Aldrich, St. Louis, MO, USA) and 100 μl of 50 mg/l each of lysozyme and proteinase K (Sigma-Aldrich, St. Louis, MO, USA) then incubated at 37 °C for 2 h in a water bath, with occasional mixing by inversion. Genomic DNA (gDNA) was extracted using the Maxwell 16 Cell DNA Purification Kit (Promega, Madison, WI, USA) according to the manufacturer’s protocol. Verification of sample quality and quantity, and purity was carried out prior to sequencing.

2.4. Whole genome sequencing

The genome of COTT was sequenced by Novogene Corporation Inc., Sacramento, CA, USA. Bacterial genome sequencing was performed using high-throughput Illumina technology on HiSeq using the paired-end (PE150) sequencing strategy. High-quality reads were obtained after quality control was performed in-house by Novogene. The software in the bioinformatics analysis pipeline of the whole genome sequencing (WGS) analyses by Novogene can be viewed in Additional file 2: Supplementary Table S2 and a detailed method in Additional file 4: Supplementary Method.

2.5. Genome assembly and annotation

Reads were assembled to contigs in Shovill (Galaxy v 1.1.0+galaxy0) using the Spades assembler. The obtained contigs were validated using QUAST (Galaxy v 5.0.2+galaxy3). Annotation was initially done using BV-BRC Annotations using Subsystems Technology tool kit (RASTtk)¹³ and genome data submitted to NCBI. The Integrated Microbial Genomes Expert Review (IMG/ER) (https://img.jgi.doe.gov/cgi-bin/mer/main.cgi) server of the Joint Genome Institute (JGI) was used for expert annotation of the assembled genome and was the primary source for genome predictions and functional analysis. This server is used for functional annotation and curation of microbial genomes of interest prior to release to GenBank.¹⁴ The detailed annotation system used by IMG/ER can be viewed in Additional file 2: Supplementary Table S3. Proksee (https://proksee.ca/) was used to generate the circular map of the COTT genome.

2.6. Genome analysis—COG and KEGG analysis and comparison

The cluster of orthologous groups (COG) and The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways in the COTT genome were annotated and analysed, using the abundance profile search tool and the statistical analysis tool on JGI IMG/ER. COG and KEGG CDSs of the COTT genome were compared to the three genomes available for C. oranimense: DSM 19055, G311 287168971/referred to as G311_A in text and G311 2602041528/referred to as G311_B in text (Additional file 2: Supplementary Table S4). Venn diagrams comparing the unique gene inventories in the C. oranimense genomes were done using OmicsBox (https://www.biobam.com/omicsbox). Fisher’s exact t-test was used to evaluate statistically significant differences in gene abundance in each COG and KEGG category between the COTT strain and other reference genomes deposited in the IMG bacteria genome database. The COTT genome was compared with 18 crude oil degrading specific bacterial genomes which were obtained via the Genomes by Ecosystem search tool on JGI (Additional file 2: Supplementary Table S5).

2.7. Antibiotic susceptibility

Antibiotic susceptibility testing (AST) was performed using HiMedia Antimicrobial Sensitivity Test using Single Test Discs according to the manufacture’s guidelines. Zone size was interpreted as per Clinical & Laboratory Standards Institute (CLSI) (https://clsi.org/) and European Committee on Antimicrobial Susceptibility Testing (EUCAST) (http://www.eucast.org). The experiments were performed in duplicate. Subsequently, antimicrobial resistance genes were identified in JGI. antiSMASH was used for rapid genome-wide identification, annotation and analysis of secondary metabolite biosynthesis gene clusters.¹⁵

3. Results and discussion

3.1. Phylogenetic analysis

An axenic COTT strain that was able to grow on 3% crude oil amended media in a previous study was used in this WGS analysis (Fig. 1). The 16S rRNA phylogenetic tree was constructed to show the phylogenetic position of the hydrocarbonoclastic COTT strain and to authenticate the COTT genome assembly (Additional file 1: Supplementary Fig. S1). 16S rRNA sequence comparison confirmed that COTT was most similar to C. oranimense type strain H8 (= LMG 24030 = DSM 19055; NR_044168; 100%/99.47% QC/ID)¹⁶ and C. oranimense strain FSB_HA2 (MG322209; 100%/99.05% QC/ID).¹⁷

The protein sequences of 1,000 housekeeping genes C. oranimense and other species were aligned using the bacterial phylogenetic tree building service in BV-BRC. The tree showed multiple highly supported clades (Fig. 2). Although the COTT strain had the closest similarity to G311 and DSM 19055, COTT was positioned separately within this C. oranimense clade; there seemed to be cladogenetic splitting into three distinct branches which may have arisen due to unique morphotype and/or microhabitat relationship within a given community structure.¹⁸ Cladogenesis as seen here, may indicate some evolutionary advantage to utilize new resources which perhaps, also implies rapid diversification of ecotypes as a trade-off for improved competitive ability in new ecological niches.

3.2. Genome project history

A summary of the project and information about the genome sequence is shown in Additional file 2: Supplementary Table S6. A complete report on sequencing data quality control, mapping statistics and SNP, InDel, SV and CNV detection and annotation can be viewed in Additional file 3: Notes and Analysis S1.

3.3. Genome features

The genome annotation from JGI IMG/ER was used as the final and primary source for genome predictions and comparisons. As such, detailed data of the annotation data from JGI is presented in the main text and BV-BRC annotation data and analysis can be viewed in Additional file 3: Notes and Analysis S2.

The genome features are summarized in Table 1 and Fig. 3. The COTT genome has 11 contigs, a total length of 4,265,502 base pairs (bp), G+C content of 37.91% and a total of 4,040 genes. Biological roles were assigned to 2,794 of the 3,959 predicted protein coding genes (CDS), whereas 1,165 proteins were without a predicted function. 2,690 have CDS with clusters of orthologous groups (COGs) and 936 CDS are connected to KEGG pathways. No plasmids were detected in this strain, similar to the G311 and DSM 19055.¹⁹ All CDS can be viewed in Additional file 2: Supplementary Table S7. The list of annotated genes and their protein products as assigned to COGs and KEGG Orthology (KO) are presented in Additional file 2: Supplementary Tables S8 and S9, respectively.

3.4. Comparison of COTT to 18 crude oil-degrading bacteria

Comparative analysis was performed to determine the difference in gene abundance in the COTT genome and 18 crude oil-degrading bacteria for which high-quality crude oil-degrading bacterial genomes were available on JGI (Additional file 2: Supplementary Table S10 a-c; abundance of 0 means that a specific gene according to COG ID and COG Name was absent in the species and any number greater than 0 reflects the gene count for that specific gene). Among the COG categories, the COTT genome has a significantly higher mean abundance (Fisher’s Exact t-test; P < 0.001; P < 0.05 Additional file 2: Supplementary Table S11) pertaining to categories such as cell wall/membrane/envelope biogenesis, post-translational modification, protein turnover, chaperones, replication, recombination and repair, and translation, etc. compared to the other 18 genomes. Other COG categories, e.g. responsible for amino acid metabolism, energy production and conversion and lipid metabolism, etc., were significantly lower than the mean average levels in COTT (Fisher’s Exact t-test; P < 0.001; Additional file 2: Supplementary Table S11).

We then examined gene abundance and found that the COTT genome has 24 characterized CDS in COG, in addition to 16 ‘uncharacterized’ proteins, that were absent from the other 18 genomes (Table 2 and Additional file 2: Supplementary Table S12). There are also 21 proteins present in multiple copies (≥5 copies) in the COTT genome that were present in single or less than 5 copies in the other 18 genomes (Additional file 2: Supplementary Table S13). Statistically (Fisher’s Exact t-test; P < 0.001; P < 0.05), comparative analysis of COG genes between COTT and all 18 genomes showed that there were 28 genes that were significantly different, where 25 were significantly higher in mean abundance in COTT and 3 were significantly lower in COTT compared to the 18 genomes (Additional file 2: Supplementary Table S14).

The COTT genome has a significantly higher (Fisher’s Exact t-test; P < 0.001; P < 0.05) mean abundance of genes mapped by KEGG (Additional file 2: Supplementary Table S15); notably, proteins of higher abundance were related to multi-drug resistance ²⁰. Statistically (Fisher’s Exact t-test; P < 0.001; P < 0.05), comparative analysis of KO genes between COTT and all 18 genomes showed that there were 22 KO genes that were significantly different, all of which were significantly higher in mean abundance in COTT compared to the 18 genomes (Additional file 2: Supplementary Table S16); notably, proteins of higher abundance were related to multi-drug resistance.²⁰

3.5. Comparison of COTT to other C. oranimense genomes

The average nucleotide identity (ANI) score was calculated to be 98.25636% (ANI calculator on JGI IMG/ER) for each of the COTT-G311 and COTT-DSM19055 and for G311-DSM19055 pairwise comparisons.²¹

3.5.1 COG comparisons

Analysis of gene abundance in the COG categories between COTT and two C. oranimense genomes, DSM 19055 and two assemblies of G311, G311_A (G311 2871689712) and G311_B (G311 2602041528), is shown in Fig. 4(a)–(c) and in Additional file 2: Supplementary Table S17. A comparison of enzymes of industrial interest predicted for COTT is shown in Fig 4(d).

Investigation of the COGs revealed genes that were only found in the COTT genome (Table 3 and Additional file 2: Supplementary Table S18). Fisher’s Exact t-test did not reveal any significant differences (P < 0.05) in gene abundance according to COG category (Additional file 2: Supplementary Table S19). However, there were higher abundances for genes in COTT compared to the other C. oranimense genomes in 9 out of 24 COG categories (Additional file 2: Supplementary Table S19). There were significant differences (P < 0.05) in abundance of COG0454 N-acetyltransferase, GNAT superfamily (includes histone acetyltransferase HPA2), COG0477 MFS family permease and COG0457 tetratricopeptide (TPR) repeat (Additional file 2: Supplementary Table S20). The reason may lie in COTT’s need for metabolic self-regulation and survival in soil chronically contaminated with crude oil.²²

3.5.2 KEGG comparisons

Genes associated with functional KEGG categories were also examined (Fig. 4(b) and Additional file 2: Supplementary Table S21) and it was found that differences in gene abundance according to KEGG category and KO function were not significant (Additional file 2: Supplementary Table S22 and S23, respectively).

3.5.3 Genes only found in the COTT genome

Bacteria involved in xenobiotic degradation have robust bio-catalytic systems that allow them to function as microbial factories.²³ The ability of COTT to survive in its environment was predicted based on its unique gene inventories (Fig. 5, Table 3 and Additional file 2: Supplementary Table S18). Genes found in COTT but not in the other Chryseobacterium genomes according to the COG category are general function prediction only—1; replication, recombination and repair—6; post-translational modification, protein turnover, chaperones—1; inorganic ion transport and metabolism—3; carbohydrate transport and metabolism—3; coenzyme transport and metabolism—5; cell wall/membrane/envelope biogenesis—2; amino acid transport and metabolism—2; nucleotide transport and metabolism—2; mobilome: prophages, transposons—2; function unknown—8 (Additional file 3: Notes and Analysis S3—Table N12 contains a description of each of the above functions). These gene-based predictions suggest specific areas for functional validation.

3.6. Hydrocarbon degradation

3.6.1 Xenobiotic and petrogenic hydrocarbon degradation in COTT

The COTT genome specifies 34 genes encoding proteins associated with xenobiotic biodegradation and metabolism (Additional file 2: Supplementary Table S24) including those associated with benzoate degradation, chloroalkane and chloroalkene degradation, chlorocyclohexane and chlorobenzene degradation, ethylbenzene degradation, metabolism of xenobiotics by cytochrome P450, naphthalene degradation, nitrotoluene degradation, steroid degradation and styrene degradation. Genes encoding enzymes associated with non-specific alkane metabolism were also detected in the COTT genome (Additional file 2: Supplementary Table S25).

Bacteria are known to engage in ring cleavage in the oxidative bio-transformation of crude oil.²⁴ The COTT genome encodes numerous proteins involved in the degradation of aromatic compounds: in the phenylpropionate degradation pathway, 2-hydroxy-6-oxonona-2,4-dienedioate hydrolase (mhpC) similar to the petroleum oil-degrader Pseudomonas aeruginosa²⁵; alcohol dehydrogenase (adh) and propanol-preferring alcohol dehydrogenase (adhP) that act upon aliphatic alcohols, xenobiotic aromatic and aliphatic hydroxyls via similar pathways²⁶; gluconolactonase (gnl, RGN) that is involved in the degradation of cyclohexanols which are moderately resistant to biodegradation.²⁷ 2-Haloacid dehalogenase was found to be encoded in the COTT genome, which suggests that this strain can potentially degrade and/or detoxify recalcitrant halogenated xenobiotic pollutants.^28,29 Novel sources of haloacid dehalogenase may offer biotechnological applications ranging from waste treatment to synthesis of stereoisomers³⁰ and bioremediation.³¹Additional file 2: Supplementary Table S24 contains this data.

Members of the NAD(P)H:FMN oxidoreductase family are thought to participate in oxidative stress responses and in the degradation of xenobiotic compounds including a wide array of nitroalkanes and nitroaromatics.³² Evidence of ­2,4,6-trinitrotoluene (TNT) degradation via an anaerobic pathway has been revealed for the COTT strain (http://eawag-bbd.ethz.ch/tnt2/tnt2_map.html). It is possible that COTT adopted this anaerobic mechanism as its genome possesses genes encoding Fe-S oxidoreductase, sulphite reductase, alpha subunit (flavoprotein) and sulphite reductase, beta subunit (haemoprotein), carbon monoxide dehydrogenase and two genes encoding dienelactone hydrolase for anaerobic TNT degradation. Additional file 2: Supplementary Tables S8 and S24 contain this data.

3.6.2 Mono- and di-oxygenases for aromatic degradation

The degradation of aromatic hydrocarbons requires ring activation which is catalyzed by monooxygenases or dioxygenases/dehydrogenases.³³ Genes encoding nitronate monooxygenase (ncd2, npd) were detected in COTT. Detoxification of nitroalkane compounds that are used in industry is of considerable interest.³⁴ Another example is cholesterol oxidase which is involved in styrene degradation, and is encoded in the COTT genome. Cholesterol oxidase has an application as biocatalysts in industry.³⁵

The COTT genome encodes catechol 2,3-dioxygenase (C23O), which catalyzes the meta (extradiol) ring cleavage of catechol and its alkyl derivatives under aerobic conditions. C23O genes are essential for the degradation of monoaromatic hydrocarbons e.g., BTEX and phenol^36–38 as well as the degradation of PAHs e.g., naphthalene, phenanthrene³⁹ and pyrene.^40,41 Six quercetin 2,3-dioxygenase genes/redox-sensitive bicupin YhaK belonging to the pirin family were detected in COTT and these genes have been implicated as associated with hydrocarbon-pollutant degradation.⁴²Additional file 2: Supplementary Tables S8 and S24 contain this data.

In addition, oxygenase-type enzymes required for pollutant detoxification, e.g. catalases (katG and katE) and dye-decolourizing peroxidase (DyP-type peroxidases) were identified in the COTT genome. Porphyrinogen peroxidase (yfeX) and chloroperoxidase (cpo), which are of industrial interest as biological decolourizers of synthetic dyes were also encoded. Additional file 2: Supplementary Tables S7, S8 and S24 contain this data.

3.6.3 Propanoate metabolism

Propionate, a short chain fatty acid (SCFA), is a major component of subsurface petroleum reservoirs.⁴³ Propanoate metabolism was found to be stimulated by the presence of crude oil.⁴⁴ SCFAs produced by hydrocarbonoclastic microbes are involved in biodegradation of petroleum and has been shown to be crucial for petroleum recovery and energy extraction.^43,45,46 In COTT, the succinate pathway appears to be the dominant pathway for propionate production. Additional file 2: Supplementary Table S24 contains this data.

3.6.4 Lipid metabolism

The COTT genome has 72 lipid metabolism genes (Additional file 2: Supplementary Table S24). Lipid metabolism plays an integral role in glycolipid biosurfactant production and in the formation of the lipopolysaccharide and phospholipid scaffolds of the outer membrane of Gram-negative bacteria.⁴⁷ Notably, the esterase and lipase superfamily are economically important enzymes to a range of industries.^48,49 The COTT genome contains two genes encoding esterase/lipase (COG4782) which were not found in the DSM 19055 strain and only one copy was found in G311. Additional file 2: Supplementary Table S24 contains this data.

3.7. Chaperones in protein folding

Chaperones, also known as heat shock proteins, assist in the correct three-dimensional folding of proteins, acting to stabilize or protect disassembled polypeptides, for normal cell growth; they are stress-induced under conditions of high temperatures and in the presence of pollutants and heavy metals.⁵⁰ As expected, the COTT genome harbours a comparatively higher number of genes coding for chaperones, e.g. chaperone modulatory protein CbpM, heat shock proteins GroEL, GroES, HtpG/Hsp90, HSP20, and DnaK, DnaJ, GrpE. Additional file 2: Supplementary Table S24 contains this data.

3.8. Quorum sensing and biofilm formation

Quorum sensing (QS) enables communication and exchange of chemical signals in dense populations of bacteria especially in heavily polluted soil^51,52 and helps to regulate antibiotic production.⁵³ QS output with respect to biofilm formation is advantageous as it provides greater resistance to stress, antimicrobial agents, predation and toxic chemicals.⁵⁴ The removal of xenobiotic compounds, including crude oil and PAHs, by biofilm-forming strains has been reported⁵⁵; biofilms of Burkholderia sp. NK8 and P. aeruginosa PA01 have been reported to be involved in the degradation of chlorinated benzoates; biofilm of P. stutzeri T102 is involved in the emulsification of naphthalene⁵⁴ phenanthrene, and pyrene in activated sludge.⁵⁶

Twenty-six genes encoding enzymes that function in QS were detected in the COTT genome. In discriminating the lipid environment during QS to facilitate optimal binding to fulfil specific functions and target certain cell types, COTT has thiol-activated cytolysin (slo).⁵⁷ In the family of two-component systems (TCSs), extracellular polysaccharide/exopolysaccharides EspA/wza, EspP and EspB/etk-/wzc, were identified in the COTT genome. These gene products are required for bacterial attachment to solid surfaces, to other bacteria and in biofilm formation.⁵⁸ Thirty-two related biofilm formation genes were detected in the COTT including exopolysaccharide (EPS) genes that are known to provide the three-dimensional structure of the biofilm⁵⁹ (Additional file 2: Supplementary Table S18 and S24). Cells use EPS to navigate into the more nutrient and oxygen-rich regions at the air-liquid interface.⁶⁰

In addition to biofilm formation, genes in the COTT genome such as anthranilate synthase component 1 (trpE) and component 2 (trpG) can also serve as novel drug targets, offer antibiotic tolerance and virulence.^61,62 Genes encoding multidrug efflux system outer membrane protein (oprM) were detected in the QS pathway for COTT. Resistance-nodulation-division efflux pumps have versatile physiological functions, e.g. multidrug resistance, microbial environmental adaptability, pathogenesis and organic solvent tolerance.^63,64 In the PAH-degrading strain P. putida B6-2, efflux pumps were critical for releasing the toxicity caused by intermediates of PAH degradation.⁶⁵ Although the functions of multidrug resistance efflux pumps (MDREPs) in a clinical sense is well-understood, their roles in the degradation of PAHs require additional characterization.⁶⁶Additional file 2: Supplementary Table S24 contains this data.

3.9. Membrane transport

Nutrient uptake and elimination of toxic by-products via transporters, e.g., ABC transporters, are essential to microbial survival in polluted environments.⁶⁷ The genome of COTT has a total of 27 genes that encode various ABC transporters (Additional file 2: Supplementary Table S24) in addition to ABC subfamily B multi-drug efflux pump (mdlA, mdlB), which plays a role in lipid A and possibly glycerophospholipid transport.⁶⁸

3.10. Secretion systems

Bacterial secretion systems facilitate protein export⁶⁹ attachment to eukaryotic cells, and scavenging for energy resources.⁷⁰ Among the genes responsible for type I, II, V and VI secretion systems present in the COTT genome (Additional file 2: Supplementary Table S24), type VI secretory protein (vgrG), is implicated in (i) bacterial interactions and community structure in a range of environmental niches, (ii) in the ability of many pathogenic bacteria to outcompete rival bacteria, and (iii) in the direct interaction of symbionts with their eukaryotic hosts.⁷¹

3.11. Regulation patterns

Bacteria can sense and respond to environmental stimuli which is important to survival particularly in highly dynamic, competitive niches. There are several recognized classes of sensory signal transduction systems in prokaryotes.⁷² Based on KEGG analysis, 91 CDSs in the COTT genome were classified into signal transduction pathways and of these, 66 were classified into TCSs which consist of two types of signal transducers, a sensor kinase and its cognate response regulator⁷³ (Additional file 2: Table S24). The COTT genome contains 18 genes that encode proteins of the OmpR family for TCS responding to phosphate limitation and assimilation, cell envelope protein folding and protein degradation, serine protease, copper/silver efflux, potassium transport, DNA replication, iron acquisition, lantbiotics biosynthesis and immunity and adhesion, autolysis, multidrug resistance and virulence genes. Phosphorous plays a major role in bacterial biological activity in oil reservoirs.⁷⁴ PhoP and PhoR phosphate regulon response and sensor, respectively, were detected in COTT’s genome.

Response regulator YesN is also present in the COTT genome. YesN proteins belong to the AraC/XylS family of transcriptional regulators that control the expression of genes with diverse biological functions.⁷⁵ The apparent redundancy of regulatory genes may indicate a flexible mechanism associated with nitrogen fixation and assimilation.⁷⁶

radC, DNA repair protein, is commonly associated with microorganisms that inhabit harsh environments.⁷⁷ The radC gene was more abundant in the COTT genome than in the other C. oranimense genomes. The atoB gene, which is present in COTT, encodes acetyl-CoA C-acetyltransferase that is involved in short-chain fatty acid metabolism and induces modulation of chemotactic behaviour, as well as other yet undefined responses.⁷⁸ Also detected in this family were C4-dicarboxylate transport proteins DctA (similar to DctP above) and RNA polymerase σ (sigma)-54 factor RpoN.⁷⁹ COTT has four gloA genes, which code for a type I glyoxalase, which may serve as an oxygen sensor and regulator in transcription of the Fe-Mo transport operon for nitrogen fixation,⁸⁰ as well as three type II glyoxalases (gloB) which may be important to minimize the effects of environmental stressors on protein stability or enzymatic activity in bacteria.⁸¹

cbb₃ oxidases were found in COTT. cbb₃ oxidases possess a high affinity for oxygen and are encoded by the tetracistronic ccoNOQP-1 and ccoNOQP-2 operons.^82,83 The enzymatic and transcriptional characteristics of cbb₃ oxidases assist bacteria (e.g. P. aeruginosa) to grow in low-oxygen environments.^84,85 Cytochrome bd quinol oxidase was present in COTT and was reported to become active under hypoxic conditions^86,87 and under high H₂S concentration.⁸⁸

3.12. Nutrient uptake (Fe, S, P, N)

3.12.1 Iron

Most prokaryotes regulate transcription of metal ion-responsive genes to coordinate and regulate metal homeostasis, e.g. ferric uptake regulator, Fur, responds to changes in iron availability.⁸⁸ The fur/zur gene is present in the COTT genome, and studies suggest that Fur downregulates the production of iron sequesters, e.g., siderophore biosynthesis.⁸⁸ The ferric enterobactin receptors (fepA, pfeA, iroN, pirA) are also present in the COTT genome. Additionally, COTT has several genes essential to iron uptake into the cytoplasm, including energy transduction genes exbB, exbD, tonB, tolQ/exbB and the tonB-dependant receptor (TBDR).⁸⁹Additional file 2: Supplementary Table S24 contains this gene inventory.

3.12.2 Sulphur

The COTT genome has 15 genes encoding enzymes involved in sulphur utilization and one sulphate permease encoding gene of the SulP family.⁹⁰ Less attention has been paid to the mechanisms for sulphur (sulphate) uptake from the environment and how it is subsequently assimilated into organic compounds such as cysteine.⁹¹ There is no data on the mechanism of sulphate uptake in crude oil environments where high concentrations of sulphate exist, as would be the case for the oil-polluted habitat of the COTT strain. The cysH,D,N,E,K,G,J,I genes are implicated in the assimilation of sulphur into cysteine in the COTT genome. Additional file 2: Supplementary Table S24 contains this gene inventory.

3.12.3 Phosphorus

phoA,B,H,L,R,B1,P phosphate response regulators belonging to the OmpR family, are present in COTT. These pho genes may play a vital role in phosphate/phosphonate transport systems by way of organophosphonate biosynthesis and catabolism. Phosphonates are catabolized by phn genes and COTT had one phosphodiesterase (phnP), one phosphonoacetate hydrolase (phnA) and two PhnB protein (phnB) genes.^92,93 The genome also had phosphate transport system substrate-binding protein (pstS) which encodes a periplasmic phosphate binding protein, exopolyphosphatase (ppx) which, when conditions change and phosphate becomes limiting, allows phosphate to be liberated with no expenditure of energy; ppx has been previously described to be involved in heavy metal resistance and efflux and polyphosphate kinase (ppk), which is present in COTT, is involved in polyphosphate storage.⁹⁴ Phosphate starvation-inducible phoH-like protein (phoH/L) was also detected in COTT. GDP-α-D-mannose is a key substrate in glycoprotein formation and is produced by the enzyme mannose-1-phosphate guanylyltransferase (manC/cpsB). It is responsible for transferring phosphorus-containing groups and was more abundant on COTT genome than the other C. oranimense genomes. cpsB expression is increased by osmotic shock, physiological response for enhanced environmental application.⁹⁵Additional file 2: Supplementary Table S24 contains this gene inventory.

3.12.4 Nitrogen

In the nitrogen regulatory protein C (NtrC) family, three genes glnA, ntrY and ntrX are involved in nitrogen metabolism when nitrogen availability is low and were found in the COTT genome. Similarly, COTT glutamine synthetase (glnA), is involved in glutamate metabolism under limiting conditions.⁹⁶ For symbiotic nitrogen-fixing bacteria, Azorhizobium caulinodans and Azospirillum brasilense ntrY gene expression is upregulated in response to hypoxic conditions which in turn signals regulator ntrX to increase expression of nitrogen respiration enzymes.⁹⁷ This system thus functions as a redox sensor. A drag gene encoding ADP-ribosyl-[dinitrogen reductase] hydrolase was identified in COTT. draG is a key player in the regulation of nitrogenase activity.⁹⁸ Glutaminase, encoded by glsA, present in COTT may be integral to supplying nitrogen required for the biosynthesis of a variety of metabolic intermediates.^99,100 COTT had the highest abundance of serine protease, subtilisin family, and it is predicted that these exoproteases break down proteins present in the environment in response to low levels of available nitrogen.¹⁰¹Additional file 2: Supplementary Table S24 contains this gene inventory.

3.13. Heavy metal resistance

Crude oil and heavy metal pollution occur simultaneously in soil.¹⁰² Toxic concentrations of cobalt (Co), nickel (Ni) and zinc (Zn) are detoxified by cation efflux mechanisms in bacteria.^103,104 The COTT genome contains genes associated with cation antiporter cobalt–zinc–cadmium CzcCBA. Czc in C. metallidurans allowed its survival in metal-rich environments.¹⁰⁵ Zinc and cadmium transporter, ZipB, is a ZIP zinc transporter protein, which controls the influx of zinc into the cytoplasm from outside of the cell and transport of cadmium^106,107 and it was identified in COTT.

One means of relieving copper excess is through Cu²⁺ export facilitated by different systems, e.g. Cus, Cop and Cut, which contribute to copper homeostasis (Cu²⁺ import, export, detoxification) in bacteria.^108,109 Importantly, high levels of Cu²⁺ disrupt Fe–S clusters which in turn affects disulphide bond formation and correct protein folding.¹¹⁰ The multi-protein complex that controls copper efflux, CusCBA, was detected in COTT. Copper resistance phosphate regulon response regulator (CusR) was also detected in COTT, which under anaerobic conditions, activates transcription of the cus operon.¹⁰⁹ COTT also possesses genes in the copper inducible cop operon which aids in regulating copper resistance and detoxification, as well as heavy metal tolerance.¹¹¹Additional file 2: Supplementary Table S24 contains this data.

3.14. Oxidative stress

Environmental pollutants can serve as oxidants that undergo biotransformation to produce free radical oxides, cations and anions.¹¹² The COTT genome has three genes that code for Cu/Zn superoxide dismutase SOD2, and one Fe/Mn superoxide dismutase SOD1. After SOD converts radical species to H₂O₂, catalase neutralizes H₂O₂ into harmless H₂O and O₂.¹¹³ An alkyl hydroperoxide reductase (ahpC) encoded gene was found in COTT genome; it is known to be directly involved in H₂O₂ detoxification and it also plays an important role in biofilm formation.¹¹⁴ The COTT genome had three diheme cytochrome c peroxidase genes which are responsible for H₂O₂ reduction.¹¹⁵ The presence of oxidative stress proteins may enable COTT to cope with shifts in sediment oxygen concentration.¹¹⁶

3.15. Antibiotic resistome of COTT

There has been an ongoing, dedicated search for new antibiotics for the treatment of multidrug-resistant (MDR) infections caused by bacteria. Based on antiSMASH analysis, there were genes encoding novel antimicrobial compounds in the COTT genome (Additional file 2: Supplementary Table S26). Flexirubin and carotenoid clusters were identified in antiSMASH (Additional file 1: Supplementary Fig. S2) and are likely expressed to produce pigments to give colonies of this genus (chryseos = golden) a yellow to orange colour.¹¹⁷

The COTT strain was resistant to aztreonam, kanamycin, chloramphenicol, ampicillin and erythromycin but was susceptible to imipenem-EDTA, streptomycin, ciprofloxacin and trimethoprim in in vitro disc assays (Additional file 1: Supplementary Fig. S3 and Additional file 2: Supplementary Table S27). The resistome of COTT consisted of 21 β-lactam resistance genes, 13 cationic antimicrobial peptide (CAMP) resistance genes, 5 vancomycin resistance genes and 2 tetracycline resistance genes (Additional file 2: Supplementary Table S28). The COTT genome was the only C. oranimense genome containing the bla regulator protein BlaR1 (K02172), which largely controls β-lactam antibiotic resistance¹¹⁸ and membrane carboxypeptidase (penicillin-binding protein) which would render COTT resistant to penicillin (Additional file 2: Supplementary Table S18 and S28). There were also three phenicol and two bacteriocin genes identified in COTT (Additional file 2: Table S28).

The COTT genome has 16 gene copies encoding the major facilitator superfamily (MFS) of permeases (Additional file 2: Table S18). Notably, among these were the DHA1 and DHA2 family of multidrug resistance proteins which confer resistance to tetracycline, bicyclomycin/chloramphenicol; the FSR family of transporters which enables resistance to fosmidomycin, EmrB/QacA subfamily drug resistance transporter and PAT family β-lactamase induction signal transducer, AmpG, which allow ampicillin resistance (Additional file 2: Supplementary Tables S7 and S9). A summary of the AMR and multi-drug genes annotated in this genome and corresponding mechanisms are provided in Table 4. In this context, this report may have additional clinical value.

This work predicted distinctive metabolic capabilities that are unique to COTT (Fig. 5) and analysis of the COTT resistome indicated that these findings have clinical value. Notably, we also provided gene-based evidence that COTT is capable of producing several biocatalysts that may be important to different industrial processes.

Acknowledgements

(A portion of) These data were produced by the U.S. Department of Energy Joint Genome Institute (https://ror.org/04xm1d337; operated under Contract No. ­DE-AC02-05CH11231) in collaboration with the user community. The sequencing was performed by Novogene and the ­annotation and bioinformatic analysis was supported by the U.S. Department of Energy Joint Genome Institute.

Funding

This study was funded by The University of the West Indies Campus Research and Publications Grant (CRP.5.OCT18.77).

Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

A.C.R. and S.N.R. prepared the manuscript, performed data analysis, conceptualized and designed experiments. All authors read and approved the final manuscript.

Data Availability

The whole genome sequence of C. oranimense TT (COTT) is available in DDBJ/ENA/GenBank repository [JAPDSL000000000, https://www.ncbi.nlm.nih.gov/search/all/?term=JAPDSL000000000], Genomes OnLine Database (GOLD) [Project ID: Gp0615513, https://gold.jgi.doe.gov/projects?id=Gp0615513] and JGI IMG/M database [https://img.jgi.doe.gov/cgi-bin/m/main.cgi?section=TaxonDetail&page=taxonDetail&taxon_oid=2945865373].

References