MftG is crucial for ethanol metabolism of mycobacteria by linking mycofactocin oxidation to respiration

Reviewer #1 (Public review):

Using a knock-out mutant strain, the authors tried to decipher the role of the last gene in the mycofactocin operon, mftG. They found that MftG was essential for growth in the presence of ethanol as the sole carbon source, but not for the metabolism of ethanol, evidenced by the equal production of acetaldehyde in the mutant and wild type strains when grown with ethanol (Fig 3). The phenotypic characterization of ΔmftG cells revealed a growth-arrest phenotype in ethanol, reminiscent of starvation conditions (Fig 4). Investigation of cofactor metabolism revealed that MftG was not required to maintain redox balance via NADH/NAD+, but was important for energy production (ATP) in ethanol. Since mycobacteria cannot grow via substrate-level phosphorylation alone, this pointed to a role of MftG in respiration during ethanol metabolism. The accumulation of reduced mycofactocin points to impaired cofactor cycling in the absence of MftG, which would impact the availability of reducing equivalents to feed into the electron transport chain for respiration (Fig 5). This was confirmed when looking at oxygen consumption in membrane preparations from the mutant and would type strains with reduced mycofactocin electron donors (Fig 7). The transcriptional analysis supported the starvation phenotype, as well as perturbations in energy metabolism, and may be beneficial if described prior to respiratory activity data.
The data and conclusions support the role of MftG in ethanol metabolism.

Reviewer #3 (Public review):

Summary:

The work by Graca et al. describes a GMC flavoprotein dehydrogenase (MftG) in the ethanol metabolism of mycobacteria and provides evidence that it shuttles electrons from the mycofactocin redox cofactor to the electron transport chain.

Strengths:

Overall, this study is compelling, exceptionally well designed and thoroughly conducted. An impressively diverse set of different experimental approaches is combined to pin down the role of this enzyme and scrutinize the effects of its presence or absence in mycobacteria cells growing on ethanol and other substrates. Other strengths of this work are the clear writing style and stellar data presentation in the figures, which makes it easy also for non-experts to follow the logic of the paper. Overall, this work therefore closes an important gap in our understanding of ethanol oxidation in mycobacteria, with possible implications for the future treatment of bacterial infections.

Weaknesses:

I see no major weaknesses of this work, which in my opinion leaves no doubt about the role of MftG.

Reviewer #4 (Public review):

Summary:

The manuscript by Graça et al. explores the role of MftG in the ethanol metabolism of mycobacteria. The authors hypothesise that MftG functions as a mycofactocin dehydrogenase, regenerating mycofactocin by shuttling electrons to the respiratory chain of mycobacteria. Although the study primarily uses M. smegmatis as a model microorganism, the findings have more general implications for understanding mycobacterial metabolism. Identifying the specific partner to which MftG transfers its electrons within the respiratory chain of mycobacteria would be an important next step, as pointed out by the authors.

Strengths:

The authors have used a wide range of tools to support their hypothesis, including co-occurrence analyses, gene knockout and complementation experiments, as well as biochemical assays and transcriptomics studies.
An interesting observation that the mftG deletion mutant grown on ethanol as the sole carbon source exhibited a growth defect resembling a starvation phenotype.
MftG was shown to catalyse the electron transfer from mycofactocinol to components of the respiratory chain, highlighting the flexibility and complexity of mycobacterial redox metabolism.

Weaknesses:

Could the authors elaborate more on the differences between the WT strains in Fig. 3C and 3E? in Fig. 3C, the ethanol concentration for the WT strain is similar to that of WT-mftG and ∆mftG-mftG, whereas the acetate concentration in thw WT strain differs significantly from the other two strains. How this observation relates to ethanol oxidation, as indicated on page 12.
The authors conclude from their functional assays that MftG catalyses single-turnover reactions, likely using FAD present in the active site as an electron acceptor. While this is plausible, the current experimental set up doesn't fully support this conclusions, and the language around this claim should be softened.
The authors suggest in the manuscript that the quinone pool (page 24) may act as the electron acceptor from mycofactocinol, but later in in the discussion section (page 30) they propose cytochromes as the potential recipients. If the authors consider both possibilities valid, I suggest discussing both options in the manuscript.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1 (Public Review):

Using a knock-out mutant strain, the authors tried to decipher the role of the last gene in the mycofactocin operon, mftG. They found that MftG was essential for growth in the presence of ethanol as the sole carbon source, but not for the metabolism of ethanol, evidenced by the equal production of acetaldehyde in the mutant and wild type strains when grown with ethanol (Fig 3). The phenotypic characterization of ΔmftG cells revealed a growth-arrest phenotype in ethanol, reminiscent of starvation conditions (Fig 4). Investigation of cofactor metabolism revealed that MftG was not required to maintain redox balance via NADH/NAD+, but was important for energy production (ATP) in ethanol. Since mycobacteria cannot grow via substrate-level phosphorylation alone, this pointed to a role of MftG in respiration during ethanol metabolism. The accumulation of reduced mycofactocin points to impaired cofactor cycling in the absence of MftG, which would impact the availability of reducing equivalents to feed into the electron transport chain for respiration (Fig 5). This was confirmed when looking at oxygen consumption in membrane preparations from the mutant and would type strains with reduced mycofactocin electron donors (Fig 7). The transcriptional analysis supported the starvation phenotype, as well as perturbations in energy metabolism, and may be beneficial if described prior to respiratory activity data.

We thank the reviewer for their thorough evaluation of our work. We carefully considered whether transcriptional data should be presented before the respirometry data. However, this would disrupt other transitions and the flow of thoughts between sections, so that we prefer to keep the order of sections as is.

While the data and conclusions do support the role of MftG in ethanol metabolism, the title of the publication may be misleading as the mutant was able to grow in the presence of other alcohols (Supp Fig S2).

We agree that ethanol metabolism was the focus of this work and that phenotypes connected to other alcohols were less striking. We, therefore, changed “alcohol” to “ethanol” in the title of the manuscript.

Furthermore, the authors propose that MftG could not be involved in acetate assimilation based on the detection of acetate in the supernatant and the ability to grow in the presence of acetate. The minimal amount of acetate detected in the supernatant but a comparative amount of acetaldehyde could point to disruption of an aldehyde dehydrogenase.

We do not agree that MftG might be involved in acetaldehyde oxidation. According to our hypothesis, the disruption of an acetaldehyde dehydrogenase would lead to the accumulation of acetaldehyde. However, we observed an equal amount of acetaldehyde in cultures of M. smegmatis WT and ∆mftG grown on ethanol as well as on ethanol + glucose. Furthermore, the amount of acetate detected in the supernatants is not “minimal” as the reviewer points out but higher as or comparable to the acetaldehyde concentration (Figure 3 E and F, note that acetate concentration are indicated in g/L, acetaldehyde concentrations in µM). Furthermore, the accumulation of mycofactocinols in ∆mftG mutants grown on ethanol is not in agreement with the idea of MftG being an aldehyde dehydrogenase but very well supports our hypothesis that MftG is involved in cofactor reoxidation.

The link between mycofactocin oxidation and respiration is shown, however the mutant has an intact respiratory chain in the presence of ethanol (oxygen consumption with NADH and succinate in Fig 7C) and the NADH/NAD+ ratios are comparable to growth in glucose. Could the lack of growth of the mutant in ethanol be linked to factors other than respiration?

Indeed, by using NADH and succinate as electron donors we show that the respiratory chain is largely intact in WT and ∆mftG grown on ethanol. Also, when mycofactocinols were used as an electron donor, we observed that respiration was comparable to succinate respiration in the WT. However, respiration was severely hampered in membranes of ∆mftG when mycofactocinols were offered as reducing agent. These findings support our hypothesis very well that MftG is necessary to shuttle electrons from mycofactocin to the respiratory chain, while the rest of the respiratory chain stayed intact. The fact that NADH/NAD+ ratios are comparable between ethanol and glucose conditions are interesting but indirectly support our hypothesis that mycofactocin and not NAD is the major cofactor in ethanol metabolism. Therefore, we do not see any evidence that the lack of growth of the mutant in ethanol is linked to factors other than respiration.

To this end, bioinformatic investigation or other evidence to identify the membrane-bound respiratory partner would strengthen the conclusions.

We generally agree that it is an important next step to identify the direct interaction partners of MftG. However, we are convinced that experimental evidence using several orthogonal approaches is required to unequivocally identify interaction partners of MftG. Nevertheless, we agree that a preliminary bioinformatics study, could guide follow-up studies. We therefore attempted to predict interaction partners of MftG using D-SCRIPT and Alphafold 2. However, our approach did not reveal any meaningful results. Thus, we prefer not to integrate this approach into the manuscript but briefly summarize our methodology here: To predict potential interaction partners of M. smegmatis mc2 155 MftG (MSMEG_1428), D-SCRIPT (Sledzieski et al. 2021, https://doi.org/10.1016/j.cels.2021.08.010) with the Topsy-Turvy model version 1 (Singh et al. 2022, https://doi.org/10.1093/bioinformatics/btac258) was employed to screen every combination of the MSMEG_1428 amino acid sequence with the amino acid sequence of every potential interaction partner from the M. smegmatis mc2 155 predicted total proteome (total 6602 combinations, UniProt UP000000757, Genome Accession CP000480). Predictions failed for eight potential interaction partners due to size constraints (MSMEG_0019, MSMEG_0400, MSMEG_0402, MSMEG_0408, MSMEG_1252, MSMEG_3715, MSMEG_4727, MSMEG_4757; all amino acids sequences ≥ 2000 AA). Afterward, the top 100 predicted interaction partners, ranked by D-SCRIPT protein-protein-interaction score, were subjected to an Alphafold 2 multimer prediction using ColabFold batch version 1.5.5 (AlphaFold 2 with MMseqs2, Mirdita et al. 2022, https://doi.org/10.1038/s41592-022-01488-1) on a Google Colab T4 GPU with a Python 3 environment and the following parameters (msa_mode: MMseqs2 (UniRef+Environmental), num_models = 1, num_recycles = 3, stop_at_score = 100, num_relax = 0, relax_max_iterations = 200, use_templates = False). As input, the MSMEG_1428 amino acid sequence was used as protein 1 and the amino acid sequence of the potential interaction partner was used as protein 2. In addition, proteins of the electron transport chain and the dormancy regulon (dos regulon) were included as potential interaction partners. In total, 222 unique potential MftG interactions were predicted. The AlphaFold 2 model interface predicted template modelling (ipTM) score peaked at 0.45 for MftG-MftA. This score, however, lies below the threshold of 0.75, which indicates a likely false prediction of interaction (Yin et al. 2022, https://doi.org/10.1002/pro.4379). Nonetheless, the models with the highest ipTM scores (MftG with MftA, MSMEG_3233, MSMEG_4260, MSMEG_0419, MSMEG_5139, MSMEG_5140) were inspected manually using ChimeraX version 1.8 (Meng et al. 2023, https://doi.org/10.1002/pro.4792). However, no reasonable interaction was found.

Reviewer #2 (Public Review):

Summary

Patrícia Graça et al., examined the role of the putative oxidoreductase MftG in regeneration of redox cofactors from the mycofactocin family in Mycolicibacerium smegmatis. The authors show that the mftG is often co-encoded with genes from the mycofactocin synthesis pathway in M. smegmatis genomes. Using a mftG deletion mutant, the authors show that mftG is critical for growth when ethanol is the only available carbon source, and this phenotype can be complemented in trans. The authors demonstrate the ethanol associated growth defect is not due to ethanol induced cell death, but is likely a result of carbon starvation, which was supported by multiple lines of evidence (imaging, transcriptomics, ATP/ADP measurement and respirometry using whole cells and cell membranes). The authors next used LC-MS to show that the mftG deletion mutant has much lower oxidised mycofactocin (MFFT-8 vs MMFT-8H2) compared to WT, suggesting an impaired ability to regenerate myofactocin redox cofactors during ethanol metabolism. These striking results were further supported by mycofactocin oxidation assays after over-expression of MftG in the native host, but also with recombinantly produced partially purified MftG from E. coli. The results showed that MftG is able to partially oxidise mycofactocin species, finally respirometry measurements with M. smegmatis membrane preparations from WT and mftG mutant cells show that the activity of MftG is indispensable for coupling of mycofactocin electron transfer to the respiratory chain. Overall, I find this study to be comprehensive and the conclusions of the paper are well supported by multiple complementary lines of evidence that are clearly presented.

Strengths

The major strengths of the paper are that it is clearly written and presented and contains multiple, complementary lines of experimental evidence that support the hypothesis that MftG is involved in the regeneration of mycofactocin cofactors, and assists with coupling of electrons derived from ethanol metabolism to the aerobic respiratory chain. The data appear to support the authors hypotheses.

We thank the reviewer for their thorough evaluation of our work.

Weaknesses

No major weaknesses were identified, only minor weaknesses mostly surrounding presentation of data in some figures.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

(1) In Fig 6 C and D, would it not be expected that MMFT-2H2 would be decreasing over time as MMFT-2 is increasing?

This is true. MMFT-2H2 is indeed decreasing while MMFT-2 in increasing, however, since the y-axis is drawn in logarithmic scale the visible difference is not proportional to the actual changes. The increase of MMFT-2 against a very low starting point is more clearly visible than the decrease of MMFT-2H2, which was added in high quantities.

(2) It would be beneficial to include rationale regarding the electron acceptors tested and why FAD was not included.

FAD is a prosthetic group of the enzyme and was always a component of the assay. The other electron acceptors were chosen as potential external electron acceptors.

(3) Bioinformatic analysis to capture possible interacting partners of MftG

See our response to the previous review.

Reviewer #2 (Recommendations For The Authors):

Questions:

(1) The co-occurrence analysis showed that one genome encoded mftG, but not mftC - do the authors think that this is a mftG mis-annotation?

This is a good question. We have investigated this case more closely and conclude that this particular mftG is not a misannotation. Instead, it appears that the mftC gene underwent gene loss in this organism. We added on page 8, line 15: “Only one genome (Herbiconiux sp. L3-i23) encoding a bona fide MftG did not harbor any MftC homolog. However, close inspection revealed the presence of mftD, mftF, and a potential mftA gene but a loss of mftB,C and E in this organism.”

(2) Figure 3A - the complemented mutant strain shows enhanced growth on ethanol when compared to the WT strain with the same mftG complementation vector, suggesting that dysregulation from the expression plasmid may not be responsible for this phenotype. Have the authors conducted whole genome sequencing on the mutant/complement isolate to rule out secondary mutations?

This is an interesting point. We have not conducted further investigations into the complement mutant. However, we can confidently state that the complementation was successful in that it restored growth of the ∆mftG mutant on ethanol, thus confirming that the growth arrest of the mutant was due to the lack of mftG activity and not due to any secondary mutation. We also observed that both the complement strain and the overexpression strain, both of which are based on the same overexpression plasmid, exhibited shorter lag phases, faster growth and higher final cell densities compared to the wild type. We interpret these data in a way that overexpression of mftG might lift a growth limited step. Notably, this is only an interpretation, we do not make this claim. What we cannot explain at the moment, is the observation that the complement mutant grew to a higher OD than the overexpression strain. This is indeed interesting, and it might be due to an artefact or due to complex regulatory effects, which are hard to study without an in-depth characterization of the different strains involved. While this goes beyond the scope of this study, we are convinced that our main conclusions are not challenged by this phenomenon.

(3) Figure 4C - could the yellow fluorescence that suggests growth arrest be quantified in these images similar to the size and septa/replication sites?

In principle, this is a good suggestion. However, the amount of yellow fluorescence only differed in the starvation condition between genotypes. Since this condition was not a focus of this study, we preferred not to discuss these differences further.

(4) Figure 4E - the complemented mutant strain has very high error, why is that? Could this phenotype not be complemented?

It is true that the standard deviation (SD) is relatively high in this experiment. This is due to the fact that single-cell analyses based on microscopic images were conducted here - not bulk measurements of the average fluorescence. This means that the high variance partially reflects phenotypic heterogeneity of the population, rather than inefficient complementation. While it is interesting that not all cells behaved equally, a finding that deserves further investigations in the future, we conclude that the mean value is a good representative for the efficiency of the complementation.

(5) While the whole cell extract experiment presented in Figure 6A is very clear, could the authors include SDS page or MS results of their partially purified MftG preparations used for figure B-F in the supplementary data to rule out any confounding factors that may be oxidising mycofactocin species in these preparations?

We did not include SDS-Page or MS results since the enzyme preparations obtained were not pure. This is why we refer to the preparation as “partially purified fraction”. Since we were aware of the risk of confounding factors being potentially present in the preparation, we used two different expression hosts (M. smegmatis ∆mftG and E. coli) and included negative controls, i.e., a reaction using protein preparations from the same host that underwent the exact same purification steps but lacked the mftG gene. For instance, Figure 6A shows the negative control (M. smegmatis ∆mftG) and the verum (M. smegmatis ∆_mftG-mftG_His6). Although this control is not shown in panels BCD for more clarity, we can assure that the proposed activity of MftG as never been detected in any extract of M. smegmatis ∆mftG. Concerning MftG preparations obtained from heterologous expression in E. coli, we also performed empty vector controls and inactivated protein controls. We added a new Supplementary Figure S4 to show one example control. Taken together, the usage of two different expression hosts along with corresponding background controls clearly demonstrates that mycofactocinol oxidation only occurred in protein extracts of bacterial strains that contained the mftG gene. Taken together, these data indicate that the observed mycofactocinol dehydrogenase activity is connected to MftG and not to any background activity.

Recommendations:

• A suggestion - revise sub-titles in the results section to be more 'results-oriented' e.g. rather than 'the role of MftG in growth and metabolism of mycobacteria' consider instead 'MftG is critical for M. smegmatis capacity to utilise ethanol as a sole carbon source for growth' or something similar.

In principle this is a good idea for many manuscripts. However, we have the impression that this approach does not reflect the complexity and additive aspect of the sections of our manuscript.

• For clarity, revise all figures to include p-values in the figure legend rather than above the figures (use asterisks to indicate significance).

We are not sure whether the deletion of p-values in the figures would enhance clarity. We would prefer to leave them within figures.

• Figure 5B -revise colour legend, it is unclear which bar on the graph corresponds to which strain.

The figure legend was enlarged to enhance readability.

• Page 8 - MftG and MftC should be lowercase and italicised as the authors are writing about the co-occurrence of genes encoded in genomes, not proteins.

Good point, we changed some instances of MftG / MftC to mftG / mftC, to more specifically refer to the gene level. However, in some cases, the protein level is more appropriate, for instance, the phylogenies are based on protein sequences. That is why we used the spelling MftG / MftC in these cases.

• Page 9 - for clarity move Figure 3 after first in text citation.

We moved Figure 3.

• Page 17 - for clarity move Figure 5 after first in text citation.

We moved Figure 5. We furthermore reformatted figure legend to fit onto the same page as the figures.

• Page 20, line 17 - 'was attempted' change to 'was performed'. The authors did more than attempt purification, they succeeded!

Since purification of MftG was not successful, we prefer the term “attempted” here. However, activity assays indeed indicate successful production of MftG.

• Page 20, line 19-21 - data showing that the MftG-HIS6 complements ∆mftG could be included in supplementary information.

Complementation was obvious by growth on media containing ethanol as a sole carbon source.

• Page 26 line 25 - 'we also we' delete duplicated we.

Thank you for the hint, we deleted the second instance of “we” in the manuscript.

• Page 26 Line 26 - 'mycofactocinols were oxidised to mycofactocinols', should this read mycofactocinols were oxidised to mycofactocinones?

Correct. We changed “mycofactocinols” to “mycofactocinones”

• Page 28 line 17, huc hydrogenase operon

We added (“huc operon”).

• Page 38 line 24, 'Two' not 'to'.

This is a misunderstanding. “To” is correct