Marco N. Allemann

Horizontal gene transfer is an important evolutionary process by which DNA is exchanged between cells that are physically co-located but not direct evolutionary descendants. Horizontal transfer of highly divergent DNA is relatively easy to detect and can produce major phenotypic changes, exemplified by the acquisition of antibiotic resistance determinants. However, transfer of high-identity DNA, for example, between strains of the same species, is likely to be more frequent, harder to detect, and highly impactful in aggregate. In this work, we demonstrate that soil isolates of the alphaproteobacterium Novosphingobium aromaticivorans can exchange chromosomal DNA, leading to multiple unselected recombination events spanning approximately 10% of the chromosome. Chromosomal recombination was directional and more efficient near an integrative and conjugative element (ICE), and required a relaxase found in the ICE. Recombination could not be observed in strains from closely related Novosphingobium species. In combination, these results suggest that ICE-mediated recombination can efficiently recombine DNA within N. aromaticivorans , increasing the adaptive potential of the species while also enforcing species boundaries through preferential intraspecific recombination. IMPORTANCE: Horizontal gene transfer is a key process in bacterial evolution. Mechanisms for transfer of mobile genetic elements are well-characterized, but less is known about how chromosomal DNA is recombined. In this work, we demonstrate that integrative and conjugative elements can efficiently recombine chromosomal DNA between strains of Novosphingobium aromaticivorans but not between different Novosphingobium species. We conclude that integrative and conjugative element-mediated chromosomal recombination can be an important adaptive mechanism within a species, due to its ability to recombine nearby chromosomal alleles, but also serves to delineate species-specific gene pools as a result of its limited phylogenetic range.

Horizontal gene transfer is an important evolutionary process by which DNA is exchanged between cells that are physically co-located but not direct evolutionary descendants. Horizontal transfer of highly divergent DNA is relatively easy to detect and can produce major phenotypic changes, exemplified by acquisition of antibiotic resistance determinants. However, transfer of high-identity DNA, for example between strains of the same species, is likely to be more frequent, harder to detect, and highly impactful in aggregate. In this work, we demonstrate that recombination between soil isolates of the alphaproteobacterium Novosphingobium aromaticivorans can exchange chromosomal DNA, leading to multiple unselected recombination events spanning approximately 10% of the chromosome. Chromosomal recombination was directional, more efficient near an integrative and conjugative element (ICE), and required a relaxase found in the ICE. Recombination could not be observed into strains from closely related Novosphingobium species. In combination, these results suggest that ICE-mediated recombination can efficiently recombine DNA within N. aromaticivorans , increasing the adaptive potential of the species while also enforcing species boundaries through preferential intraspecific recombination. Importance: Horizontal gene transfer is a key process in bacterial evolution. Mechanisms for transfer of mobile genetic elements are well-characterized, but less is known about how chromosomal DNA is recombined. In this work, we demonstrate that integrative and conjugative elements can efficiently recombine chromosomal DNA between strains of Novosphingobium aromaticivorans but not between different Novosphingobium species. We conclude that ICE-mediated chromosomal recombination can be an important adaptive mechanism within a species, due to its ability to recombine nearby chromosomal alleles, but also serves to delineate species-specific gene pools as a result of its limited phylogenetic range.

A bacterial pinoresinol/lariciresinol reductase (PLR) homolog named NrPinZ was obtained from a Novosphingobium rhizosphaerae sp. LY bacterial strain, with NrPinZ being part of its 5‐step biochemical system catabolizing pinoresinol into coniferyl aldehyde and vanillin. Recombinant NrPinZ reduces racemic 8–8′ furanofuran lignans [(±)‐pinoresinols, medioresinols, and syringaresinols] with similar overall catalytic efficiencies. In those reductions, only one of the two furan ring systems is reduced. Two other bacterial PLR homologs, NaPinZ and SlPinZ, from N. aromaticivorans F199 and Sphingobium lignivorans SYK‐6, respectively, had comparable substrate versatilities and catalytic efficacies. Plant PLR homologs, by comparison, are either enantiospecific, enantioselective, or variants thereof, being able to reduce either one or both furan rings. For example, a recombinant enantioselective PLR (PLR_Tp2) from western red cedar ( Thuja plicata ) preferentially reduces both (+)‐pinoresinol furan rings to afford (−)‐secoisolariciresinol. BoltZ‐2 modeling of NrPinZ and PLR_Tp2, together with substrate docking of (+)‐ and (−)‐pinoresinols, medioresinols, and syringaresinols, was very instructive. The NrPinZ active site P1/P2 sub‐pockets allow for both racemic forms to be catabolized. Conversely, the smaller P1 pocket in PLR_Tp2 preferentially positions (+)‐pinoresinol for downstream metabolism into (−)‐secoisolariciresinol, thereby providing a biochemical explanation for the different stereochemical outcomes. NrPinZ, NaPinZ, and SlPinZ, catalyzing substrate versatile catabolism of both racemic forms, may have important ramifications for gymnosperm and angiosperm lignin and lignan biodegradation, including its evolutionary significance and potential in enzyme engineering.

Monolignol-derived dimers containing β–β linkages are synthesized by vascular plants and can be released during lignin depolymerization. In this work, we isolated a bacterium, Novosphingobium rhizosphaerae LY, that grows with the β–β lignan (+)-pinoresinol as a sole growth substrate. Sequence analysis suggested that this strain encodes a broad range of pathways for assimilation of aromatic monomers as well as one enzyme implicated in pinoresinol catabolism but lacks other known pathways for aromatic dimer catabolism. We constructed a genome-wide barcoded transposon library and identified genes required for pinoresinol catabolism. Using feeding studies, compound isolation, targeted synthesis, and analysis of purified enzymes, we elucidated the biochemical intermediates and reaction pathway involved in pinoresinol catabolism. We demonstrated that the first enzymatic reaction is the reductive cleavage of a furan ring in (±)-pinoresinol with retention of configuration to yield lariciresinol. We additionally confirmed that the final pathway enzyme, PinU, is related to lignostilbene dioxygenases and oxidatively cleaves a diguaiacylbutadiene intermediate to yield vanillin and coniferaldehyde. Finally, based on the enzyme characterization, we demonstrated that the strain can grow with a second β–β lignan, (–)-syringaresinol, as a sole growth substrate. In combination, these results demonstrate a new biocatalytic route for transforming a widely occurring group of plant phenylpropanoid natural products. IMPORTANCE: Plants synthesize a variety of aromatic phenylpropanoid compounds containing β–β linkages, including lignin, a major structural polymeric component of the vascular plant cell wall, and lignans, biochemically related secondary metabolites with a wide range of bioactivities. Although microbial catabolic pathways have been described for dimeric phenylpropanoids featuring other interunit linkages, relatively little is known about pathways for catabolism of β–β-linked compounds. In this work, we isolated a Novosphingobium strain capable of degrading the β–β-linked lignan (+)-pinoresinol and elucidated the catabolic pathway. Understanding how bacteria catabolize β–β-linked compounds provides a basis for new biocatalytic transformations of lignans and oligolignols and has the potential to improve bacterial lignin valorization.

Lignin contains a variety of interunit linkages, leading to a range of potential decomposition products that can be used as carbon and energy sources by microbes. β-O-4 linkages are the most common in native lignin, and associated catabolic pathways have been well characterized. However, the fate of the mono-aromatic intermediates that result from β-O-4 dimer cleavage has not been fully elucidated. Here, we used experimental evolution to identify mutant strains of Novosphingobium aromaticivorans with improved catabolism of a model aromatic dimer containing a β-O-4 linkage, guaiacylglycerol-β-guaiacyl ether (GGE). We identified several parallel causal mutations, including a single nucleotide polymorphism in the promoter of an uncharacterized gene that roughly doubled the growth yield with GGE. We characterized the associated enzyme and demonstrated that it oxidizes an intermediate in GGE catabolism, β-hydroxypropiovanillone, to vanilloyl acetaldehyde. Identification of this enzyme and its key role in GGE catabolism furthers our understanding of catabolic pathways for lignin-derived aromatic compounds. IMPORTANCE: Lignin degradation is a key step for both carbon cycling in nature and biomass conversion to fuels and chemicals. Bacteria can catabolize lignin-derived aromatic compounds, but the complexity of lignin means that full mineralization requires numerous catabolic pathways and often results in slow growth. Using experimental evolution, we identified an uncharacterized enzyme for the catabolism of a lignin-derived aromatic monomer, β-hydroxypropiovanillone. A single nucleotide polymorphism in the promoter of the associated gene significantly increased bacterial growth with either β-hydroxypropiovanillone or a related lignin-derived aromatic dimer. This work expands the repertoire of known aromatic catabolic genes and demonstrates that slow catabolism of lignin-derived aromatic compounds may be due to misregulation under laboratory conditions rather than inherent catabolic challenges.

A bacterial isolate, B1D3A^T, was isolated from river sediment collected from the Hiwassee River near Calhoun, TN, by enrichment culturing with a model 5–5′ lignin dimer, dehydrodivanillate, as its sole carbon source. B1D3A^T was also shown to utilize several model lignin-derived monomers and dimers as sole carbon sources in a variety of minimal media. Cells were Gram-stain-negative, aerobic, motile, rod-shaped and formed yellow/cream-coloured colonies on rich agar. Optimal growth occurred at 30 °C, pH 7–8, and in the absence of NaCl. The major fatty acids of B1D3A^T were C_18 : 1 ω7c and C_17 : 1 ω6c. The predominant hydroxy fatty acids were C_14 : 0 2-OH and C_15 : 0 2-OH. The polar lipid profile consisted of a mixture of phosphatidylethanolamine, phosphatidylglycerol, diphosphatidylglycerol, phosphatidyldimethylethanolamine and sphingoglycolipid. B1D3A^T contained spermidine as the only major polyamine. The major isoprenoid quinone was Q-10 with minor amounts of Q-9 and Q-11. The genomic DNA G+C content of B1D3A^T was 65.6 mol%. Phylogenetic analyses based on 16S rRNA gene sequences and coding sequences of 49 core, universal genes defined by Clusters of Orthologous Groups gene families indicated that B1D3A^T was a member of the genus Sphingobium . B1D3A^T was most closely related to Sphingobium sp. SYK-6, with a 100 % 16S rRNA gene sequence similarity. B1D3A^T showed 78.1–89.9 % average nucleotide identity and 19.5–22.2% digital DNA–DNA hybridization identity with other type strains from the genus Sphingobium . On the basis of phenotypic and genotypic properties and phylogenetic inference, strain B1D3A^T should be classified as representing a novel species of the genus Sphingobium , for which the name Sphingobium lignivorans sp. nov. is proposed. The type strain is strain B1D3A^T (ATCC TSD-279^T=DSM 111877^T).