Carbon-wise utilization of lignin-related compounds by synergistically employing anaerobic and aerobic bacteria

Abstract

Background: Lignin is a highly abundant but strongly underutilized natural resource that could serve as a sustainable feedstock for producing chemicals by microbial cell factories. Because of the heterogeneous nature of the lignin feedstocks, the biological upgrading of lignin relying on the metabolic routes of aerobic bacteria is currently considered as the most promising approach. However, the limited substrate range and the inefficient catabolism of the production hosts hinder the upgrading of lignin-related aromatics. Particularly, the aerobic O-demethylation of the methoxyl groups in aromatic substrates is energy-limited, inhibits growth, and results in carbon loss in the form of CO₂.

Results: In this study, we present a novel approach for carbon-wise utilization of lignin-related aromatics by the integration of anaerobic and aerobic metabolisms. In practice, we employed an acetogenic bacterium Acetobacterium woodii for anaerobic O-demethylation of aromatic compounds, which distinctively differs from the aerobic O-demethylation; in the process, the carbon from the methoxyl groups is fixed together with CO₂ to form acetate, while the aromatic ring remains unchanged. These accessible end-metabolites were then utilized by an aerobic bacterium Acinetobacter baylyi ADP1. By utilizing this cocultivation approach, we demonstrated an upgrading of guaiacol, an abundant but inaccessible substrate to most microbes, into a plastic precursor muconate, with a nearly equimolar yields (0.9 mol/mol in a small-scale cultivation and 1.0 mol/mol in a one-pot bioreactor cultivation). The process required only a minor genetic engineering, namely a single gene knock-out. Noticeably, by employing a metabolic integration of the two bacteria, it was possible to produce biomass and muconate by utilizing only CO₂ and guaiacol as carbon sources.

Conclusions: By the novel approach, we were able to overcome the issues related to aerobic O-demethylation of methoxylated aromatic substrates and demonstrated carbon-wise conversion of lignin-related aromatics to products with yields unattainable by aerobic processes. This study highlights the power of synergistic integration of distinctive metabolic features of bacteria, thus unlocking new opportunities for harnessing microbial cocultures in upgrading challenging feedstocks.

Keywords: Acetobacterium woodii; Acinetobacter baylyi ADP1; Cis,cis-muconate; O-demethylation; Biological lignin upgrading; Guaiacol; Metabolic integration; Synergistic cocultures.

Fig. 1

The valorization of lignin-related compounds to products by A. woodii and/or ADP1. Ferulate and guaiacol are used as example model compounds. A A. woodii utilizes the methyl groups of the compounds as carbon and electron donors and double bond of the ferulate as electron acceptor in a process that fixes CO₂ and the cleaved methyl group into acetate. However, the modified aromatic compounds cannot be further utilized by A. woodii. B ADP1 cannot use guaiacol for its growth, so it is left unutilized. Ferulate is utilized as a substrate via β-ketoadipate pathway and can be directed to biomass or products of interest, but the O-demethylation produces toxic formaldehyde which is degraded into CO₂ (presented with dashed circle). C By modifying ferulate and guaiacol first anaerobically with A. woodii, ADP1 can utilize all the obtained products for its substrate. Theoretically, higher yields can be obtained compared to the process including only ADP1. In addition, by executing the O-demethylation of the aromatic compounds anaerobically, the formation of the toxic by-product formaldehyde can be avoided

Fig. 2

A Anaerobic vanillate O-demethylation in A. woodii in the presence of CO₂. 10 mM vanillate is converted into 10 mM PCA and 7.5 mM acetate. B The difference of ADP1 growth on 10 mM vanillate and on mixture containing 10 mM PCA and 7.5 mM acetate in aerobic conditions. The mixture represents the end products of vanillate O-demethylation and subsequent processing of the methyl group in the Wood–Ljungdahl pathway by A. woodii. The error bars indicate the standard deviations from the average values of the biologically independent triplicates. In some cases, the error bars are smaller than the size of the marker

Fig. 3

A A simplified scheme of the modification carried out by A. woodii for ferulate and coumarate. B. The growth of ADP1 on 15 mM of ferulate, caffeate, and dihydrocaffeate. C The growth of ADP1 on 15 mM coumarate and phloretate. The error bars in graph B and C indicate the standard deviations from the average values of the biologically independent triplicates. In some cases, the error bars are smaller than the size of the marker

Fig. 4

Two-stage cocultivation of A. woodii and ADP1 on acetobacterium medium supplemented with 6 mM guaiacol as the sole organic carbon source. The increase of acetate concentration at the beginning of the cultivation is a result of the carry-over acetate from A. woodii inoculation. ADP1 was inoculated later, at 141 h, marked in the graph with dashed vertical line. The error bars indicate the standard deviations from the average values of the biologically independent duplicates. In some cases, the error bars are smaller than the size of the marker

Fig. 5

Production of MA by ADP1ΔcatBC and A. woodii one-pot cocultivation. A Hypothetical schematics of the cultivation phases and the substrates and products in each phase. ED, Entner–Doudoroff pathway, TCA, Tricarboxylic acid cycle, WL, Wood–Ljungdahl pathway, DM, O-demethylation, BKA, β-ketoadipate pathway. B Experimental data of the three-phase one-pot coculture of A. woodii and ADP1ΔcatBC. The carry-over acetate resulted from the inoculation of A. woodii is subtracted from the acetate concentration, which therefore describes solely the acetate produced by A. woodii during the coculture.DO, deoxygenation phase. C. Close-up of the experimental data from 65.5 h to 85.5 h. Because of the spontaneous isomerization of ccMA to ctMA, the concentration of ccMA decreases during the aerobic phase in the graphs B and C