The mycofactocin redox system.

(A) Schematic representation of the mft gene cluster of M. smegmatis. mftA-F: MFT biosynthetic genes. mftR: TetR-like regulator. mftG: GMC oxidoreductase (subject of this study). (B) Chemical structures of MMFT-n (oxidized methylmycofactocin) and MMFT-nH2 (reduced form) and hypothetical scheme of MFT reduction by the ethanol dehydrogenase Mdo/Mno. The proposed mycofactocin dehydrogenase is the subject of this study. X: Unknown electron acceptor.

Bioinformatics analysis of MftG.

(A) Structural model of MftG from M. smegmatis retrieved from the Alphafold database (26) with the FAD prosthetic group (yellow) modeled into the structure. Green: Rossman fold motif (GxGxxG), red: active site histidine (His411). (B) Collapsed phylogenetic tree (maximum likelihood) of GMC enzymes showing major MftG subfamilies. FastTree support values are shown on branches. The full tree is provided as Supplementary Figure S1 (C) Venn diagram representing the frequency of co-occurrence of mftC (left-medium blue) and mftG (right-dark blue) genes in 312 organisms that encode the MFT gene locus or MftG-like proteins.

Effect of mftG gene deletion on mycobacterial ethanol metabolism.

(A) Growth curve of M. smegmatis WT, ΔmftG, ΔmftG-mftG, and WT-mftG growing in HdB-Tyl with 10 g L−1 ethanol as the sole carbon source. (B) Growth curve of WT and ΔmftG growing on 10 g L−1 glucose (C, E). Ethanol and acetic acid quantification over time in M. smegmatis WT, ΔmftG, ΔmftG-mftG, and WT-mftG cultures in HdB-tyloxapol with 10 g L−1 of ethanol and in uninoculated media as control. (D) Growth curve of WT and ΔmftG on 10 g L−1 glucose and 10 g L−1 ethanol combined. (F) Acetaldehyde quantification in culture supernatants of the WT and ΔmftG strains grown with 10 g L−1 glucose and/or 10 g L−1 ethanol. () M. smegmatis WT; () M. smegmatis ΔmftG mutant; (-dashed) M. smegmatis ΔmftG mutant grown with starting OD600 1; () M. smegmatis ΔmftG-mftG complementation mutant; () M. smegmatis double presence of the mftG gene; () Medium with 10 g L−1 of ethanol without bacterial inoculation. Measurements were performed in biological replicates, (growth curves: n≥3, ethanol and acetate quantification: n=3). Error bars represent standard deviations. Statistical analysis was performed with ordinary one-way ANOVA with Tukey’s multiple comparison test, p-values depicted in the figure.

Phenotypic characterization of mycobacterial strains

grown on HdB-Tyl with glucose and/or ethanol or starvation. (A) Quantification of dead cells by flow cytometry using propidium iodide of M. smegmatis strains grown throughout 72 h. Biological replicates, starvation cells n=2, other conditions n=3. (B) Quantification of cells with normal transmembrane potential by flow cytometry of the M. smegmatis cultures throughout 72 h. Biological replicates n=3. (C) Super-resolution microscopy images of M. smegmatis strains at exponential phase or 48h of starvation, labeled with NADA (green), RADA (red), superposition of NADA and RADA (yellow). Bar size: 3 µm. (D) Cell size distribution obtained from super-resolution microscopy of the M. smegmatis strains at exponential phase or 48h of starvation. (E) Ratio of the number of replication sites to the number of cells of the M. smegmatis strains cultures at exponential phase or 48h of starvation, together with a microscopy image of a single ΔmftG cell at 48h grown on ethanol, with arrows pointing to the several septa stained with NADA (green) and RADA (red). Bar size: 3 µm. Color legend: (A,B): – sample at 24h; – sample at 48h; – 72h. (C) (D,E): orange – 10 g L−1 glucose; green – 10 g L−1 glucose and 10 g L−1 ethanol; blue – 10 g L− 1 ethanol; white – starvation for 48h. Statistical analysis was performed for PI, cell size and ratio of replication sites per cell with ordinary one-way ANOVA, for transmembrane potential with ordinary two-way ANOVA, all using Tukey’s multiple comparison test. The p-values are depicted on the figure, microscopy-based analysis performed with technical replicates (n=3).

Cofactor metabolism of M. smegmatis strains.

(A) NADH/NAD+ ratio of M. smegmatis WT and ΔmftG grown on HdB-Tyl with either 10 g L−1 glucose or 10 g L−1 ethanol at exponential phase. (B) NADH and NAD+ quantification of M. smegmatis WT and ΔmftG grown on HdB-Tyl with either 10 g L−1 glucose or 10 g L−1 ethanol at exponential phase. (C) ADP/ATP ratio of M. smegmatis WT and ΔmftG grown on HdB-Tyl with 10 g L−1 ethanol at 24 h, 48 h and 72 h. (D, E, F) Targeted comparative metabolomics of M. smegmatis WT, ΔmftG, ΔmftG-mftG, and WT-mftG strains. The most representative MFT species, methylmycofactocinone with 8 glucose moieties (MMFT-8H2, sum formula: C62H99NO43, RT: 6.82 min, m/z 1546.5665 [M+H]+) and methylmycofactocinol with 8 glucose moieties (MMFT-8, sum formula: C62H97NO43, RT: 7.18 min, m/z 1544.5507 [M+H]+), was used to reflect MFT obtained from M. smegmatis strains. The bacteria were grown in HdB-Tyl with either (D) 10 g L−1 glucose, (E) 10 g L−1 ethanol, or (F) 10 g L−1 glucose combined with 20 g L−1 ethanol. Samples of the different growth phases are represented in the chart. A sampling at 60h of ΔmftG was chosen to sample the residual growth of the strain on ethanol as the sole carbon source. Statistical analysis was performed with one-or two-way ANOVA with Dunnett’s multiple comparison test for NADH/NAD+ ratio and, Tukey’s test for the rest, with most relevant p-values depicted on the figure. Measurements were performed in biological replicates (n=3).

MftG assays with recombinant enzymes and MFTs as substrates.

(A) Mycofactocinol oxidation assay with semi-purified cell-free extract of M. smegmatis ΔmftG or ΔmftG-mftGHis6 using ΔmftG metabolome extract as substrate (naturally enriched in reduced MFT as described in Figure 5F). Result showing the oxidation of MMFT-8H2 to MMFT-8 after overnight incubation (ON) when semi-purified cell-free extract from ΔmftG-mftGHis6 is used. t0 – start of the assay. (B) Assay with semi-purified cell-free extract of ΔmftG-mftGHis6 using synthetic PMFT as a control substrate showed no relevant reaction. (C) Successful oxidation of PMFTH2 when synthetic PMFTH2 was used as substrate (C13H17NO3, RT: 7.84 min, m/z 236.1281 [M+H]+) to PMFT (C13H15NO3, RT: 8.40 min, m/z 234.1125 [M+H]+). Black and grey lines depict reactions performed in an anaerobic chamber. (D) Successful oxidation of MMFT-2H2 (C26H39NO13, RT: 7.10 min, m/z: 574.2494 [M+H]+) to MMFT-2 (C26H37NO13, RT: 7.47 min, m/z 572.2338 [M+H]+). Black and grey lines depict assays performed in the anaerobic chamber (E) Mycofactocinol (MMFT-2H2) oxidation using MftG heterologously produced in E. coli and DCPIP, NAD+, NDMA, and PMS as potential electron acceptors. Control – no MftG added; Enzyme – MftG added. (F) Dose-dependent effect of heterologously expressed MftG on the oxidation of MMFT-2H2 to MMFT-2 was observed after a 24-hour incubation period. Sample size of all experiments n=3. Error bars represent standard deviations.

Respiratory activity (oxygen consumption) of M. smegmatis WT and ΔmftG mutants.

(A) Respiration of intact cells. Average of n=3 (B) Respiration of WT isolated cell membranes and addition of electron donors as indicated in the figure. (C) Respiration of ΔmftG isolated cell membranes and addition of electron donors as indicated in the figure. NADH and succinate served as positive controls, and water as a negative control. KCN treatment served as inhibitor control. MMFT-2H2 and PMFTH2 were added to confirm MFT’s role as an electron donor. (D) Oxidation of PMFTH2 to PMFT in WT isolated membranes (combined LC-MS profiles). (E) Oxidation of PMFTH2 to PMFT of the ΔmftG isolated membranes (combined LC-MS profiles). Each inset depicts the profile after KCN treatment. Representative data was selected from independent experiments n≥3.

Representation of the main metabolic activities affected by mftG deletion in M. smegmatis grown on 10 g L−1 ethanol as the sole carbon source compared to WT.

(A,B) Functional annotation chart (Gene ontology enrichment analysis) of the (A) up- and (B) down-regulated processes. Gene ratio denotes the ratio of the involved genes (count) to the quantity of the genes making up the enriched terms. (C) Impact on the respiration of the mutant strain ΔmftG grown on 10 g L−1 ethanol compared to WT strain. Blue represents genes downregulated p<0.05 and log2FC< −2. Red represents genes upregulated p<0.05 and log2FC> 2. Protein figures retrieved from public databases NADH-II: A0QYD6, NADH-I: 8e9g, Cyt bd: 7d5i, Cyt bcc-aa3: 7rh5, SDH-II: 6LUM, Sdh-I: 7d6x, ATP synthase: 7NJK.

List of M. smegmatis and E. coli strains, vectors, plasmids and primers used and generated on the course of this study.