Characterisation of the Paenarthrobacter nicotinovorans ATCC 49919 genome and identification of several strains harbouring a highly syntenic nic-genes cluster

Abstract

Background: Paenarthrobacter nicotinovorans ATCC 49919 uses the pyridine-pathway to degrade nicotine and could provide a renewable source of precursors from nicotine-containing waste as well as a model for studying the molecular evolution of catabolic pathways and their spread by horizontal gene transfer via soil bacterial plasmids.

Results: In the present study, the strain was sequenced using the Illumina NovaSeq 6000 and Oxford Nanopore Technology (ONT) MinION platforms. Following hybrid assembly with Unicycler, the complete genome sequence of the strain was obtained and used as reference for whole-genome-based phylogeny analyses. A total of 64 related genomes were analysed; five Arthrobacter strains showed both digital DNA-DNA hybridization and average nucleotide identity values over the species threshold when compared to P. nicotinovorans ATCC 49919. Five plasmids and two contigs belonging to Arthrobacter and Paenarthrobacter strains were shown to be virtually identical with the pAO1 plasmid of Paenarthrobacter nicotinovorans ATCC 49919. Moreover, a highly syntenic nic-genes cluster was identified on five plasmids, one contig and three chromosomes. The nic-genes cluster contains two major locally collinear blocks that appear to form a putative catabolic transposon. Although the origins of the nic-genes cluster and the putative transposon still elude us, we hypothesise here that the ATCC 49919 strain most probably evolved from Paenarthrobacter sp. YJN-D or a very closely related strain by acquiring the pAO1 megaplasmid and the nicotine degradation pathway.

Conclusions: The data presented here offers another snapshot into the evolution of plasmids harboured by Arthrobacter and Paenarthrobacter species and their role in the spread of metabolic traits by horizontal gene transfer among related soil bacteria.

Keywords: Catabolic transposon; Evolution; Nicotine metabolism; Paenarthrobacter; Plasmids.

Fig. 1

Phylogenetic tree and pairwise comparisons based on both dDDH (d₄) and ANI values of 64 genomes of Paenarthrobacter strains and of other Arthrobacter, Nocardioides, and Rhodococcus strains which possess the nic-genes. The tree was inferred with FastME 2.1.6.1 [33] from GBDP distances calculated from genome sequences. The branch lengths were scaled in terms of GBDP distance formula d₅. Values in blue represent GBDP pseudo-bootstrap support values > 60% from 100 replications, with an average branch support of 67.3%. The tree was rooted at the midpoint. The same tree clustering was used for both columns and rows in the heatmap. Cut-off values for species clustering was 70% for dDDH (d₄) [34] and 96% for ANI. Paenarthrobacter strains mislabelled as nicotinovorans are marked with a red asterisk. Strains previously reported to harbour nic-genes are marked with a black asterisk. Strains first reported here to possess the nic-genes are marked with a blue asterisk

Fig. 2

BRIG comparative genomic analysis showing the high similarity of the P. nicotinovorans ATCC 49919 genome with nine genomes of related Arthrobacter and Paenarthrobacter strains at the (A) chromosome and (B) plasmid level. The assembly IDs used for each strain are listed in Supplementary Table 2

Fig. 3

Nine highly syntenic nic-genes clusters identified in various Arthrobacter and Paenarthrobacter genomes. A BRIG comparative genomic analysis of the pAO1 megaplasmid against the strains harboring the nic-genes listed in the legend. External black circle – the localisation of the nic-genes cluster on pAO1; blue labels – pAO1 genes experimentally related to nicotine metabolism; red labels – genes related to recombination events. B Mauve alignments of the nic-genes clusters. Plasmids and genomes are listed on the left with the corresponding location of the nic-genes below. Coloured boxes – linear collinear blocks (LCB); white gaps – insertions and deletions; position atop or below the horizontal line represents the direction of LCB; red rectangles and text – recombination related ORFs; blue rectangles and text – pAO1 genes experimentally related to nicotine metabolism; nbr – nicotine blue oxidoreductase; mao – monoamine-oxidase; sad – succinic semi aldehyde dehydrogenase; folD—methylene-tetrahydrofolate dehydrogenase/ cyclohydrolase; abo—γ-N-methylaminobutyrate oxidase; purU—formyl-tetrahydrofolate deformylase; pmfR—transcriptional regulator; mobA—MobA, related to molybdopterin cytosine dinucleotide cofactor biosynthesis; nit—ω-amidase; hph—2,6-dihydroxypyridine-3-hydroxylase; pkc—putative polyketide cyclase; kdhL—ketone dehydrogenase, large subunit; pnh—2,6-dihydroxypseudooxynicotine hydrolase; kdhMS—ketone dehydrogenase, medium and subunits; 6hlno—6-hydroxy-L-nicotine oxidase; ndhLSM—nicotine dehydrogenase, large, small and medium subunits; moaA—molybdopterin cofactor synthesis protein; 6hdno—6-hydroxy-D-nicotine oxidase; hdnoR—transcriptional regulator; LCB2—locally colinear block containing genes associated with the synthesis of the molybdopterin cytosine dinucleotide cofactor; LCB1—locally colinear block containing genes for processing 6-hydroxy-D-nicotine. For an overview of the nicotine degradation pathway and role of each gene product, please see Supplementary Fig. 2

Fig. 4

Maximum Likelihood phylogeny of the microbial pyridine pathway for nicotine degradation (A) and of the plasmids harbouring the nic-genes cluster B. The nicotine pathway phylogeny was reconstructed from the concatenated purU, pnh, pmfR, ndhL, kdhL gene sequences. The plasmids phylogeny was reconstructed from features postulated to be shared by Paenarthrobacter [19]: the syntenic T4SS system and DNA repeats located 5′ of 3 key ORFs—Duf4192, DprA and ParB. Numbers on branches indicate bootstrap support percentages and values > 70% are shown. The size bar corresponds to 0.1 nucleotide substitutions per site; the length of the dashed lines is not true to scale. Blue colour indicates identical nic-genes cluster; green indicates syntenic nic-genes cluster and orange indicates the presence of 5 nic genes, but the nic-cluster is not syntenic; red indicates plasmids that share a syntenic T4SS and DNA repeats located 5′ of 3 key ORFs—Duf4192, DprA and ParB