Dithiothreitol causes toxicity in C. elegans by modulating the methionine–homocysteine cycle

  1. Gokul G
  2. Jogender Singh

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    Thiol agents, such as dithiothreitol (DTT), are toxic to many species, but the mechanisms of toxicity is incompletely understood. In this work, the authors use the animal C. elegans, a small worm, to propose a new mechanisms for how DTT causes organismal growth arrest. Specifically, they suggest that DTT causes reduction in the key molecule S-adenosyl methionine (SAM), which is used as a methyl donor to modify proteins, lipid, and/or other macromolecules. The genetic and supplementation experiments by the authors are compelling, but no direct evidence is provided that SAM levels are indeed lower following exposure of C. elegans to DTT.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

The redox reagent dithiothreitol (DTT) causes stress in the endoplasmic reticulum (ER) by disrupting its oxidative protein folding environment, which results in the accumulation and misfolding of the newly synthesized proteins. DTT may potentially impact cellular physiology by ER-independent mechanisms; however, such mechanisms remain poorly characterized. Using the nematode model Caenorhabditis elegans , here we show that DTT toxicity is modulated by the bacterial diet. Specifically, the dietary component vitamin B12 alleviates DTT toxicity in a methionine synthase-dependent manner. Using a forward genetic screen, we discover that loss-of-function of R08E5.3, an S -adenosylmethionine (SAM)-dependent methyltransferase, confers DTT resistance. DTT upregulates R08E5.3 expression and modulates the activity of the methionine–homocysteine cycle. Employing genetic and biochemical studies, we establish that DTT toxicity is a result of the depletion of SAM. Finally, we show that a functional IRE-1/XBP-1 unfolded protein response pathway is required to counteract toxicity at high, but not low, DTT concentrations.

Article activity feed

  1. Version published to 10.7554/elife.76021 on eLife
  2. eLife

    Author Response

    Evaluation Summary:

    Thiol agents, such as dithiothreitol (DTT), are toxic to many species, but the mechanisms of toxicity is incompletely understood. In this work, the authors use the animal C. elegans, a small worm, to propose a new mechanisms for how DTT causes organismal growth arrest. Specifically, they suggest that DTT causes reduction in the key molecule S-adenosyl methionine (SAM), which is used as a methyl donor to modify proteins, lipid, and/or other macromolecules. The genetic and supplementation experiments by the authors are compelling, but no direct evidence is provided that SAM levels are indeed lower following exposure of C. elegans to DTT.

    We thank the reviewers and the editor for this very nice summary of our work. In the revised manuscript, we have measured the SAM levels and provided direct evidence that SAM levels are indeed lowered upon exposure of C. elegans to DTT.

    Reviewer #1 (Public Review):

    The current manuscript investigates the mechanisms of DTT toxicity in C. elegans. In a veritable detective story, the authors show that developmental DTT toxicity is determined by the bacterial food source. They realize that the toxicity might be linked to vitamin B12 content of the food and can indeed show that low B12 levels in OP50 bacteria lead to the strongest DTT toxicity while their data suggest that wild-type worms on high B12 bacteria are protected against DTT toxicity. Indeed, B12 supplementation suppresses DTT toxicity on OP50 bacteria and this is dependent on a functional methionine synthase gene. The authors then perform a forward genetic mutagenesis screen to identify DTT resistance loci and hone in on a particular locus encoding a SAM-dependent methyltransferase they name drm-1. drm-1 loss of function protects against DTT toxicity providing support to the idea that it is the depletion of SAM that leads to DTT toxicity in worm development. This is further supported by methionine and choline supplementation experiments. Finally, the authors address the relative contribution of ER stress and SAM depletion in the DTT developmental resistance. Interestingly, they find that UPR signaling mutants affect become DTT hypersensitive only at high but not at low DTT levels. This suggests that SAM depletion is responsible for DTT toxicity at lower concentrations while only at high DTT levels, its effect on the ER becomes toxic.

    In all, this is a well-executed paper that is clear and well written. The finding is relevant as it sheds new light on the DTT mechanism, which is broadly considered an ER stressor acting on disulfide bond formation, which needs to be reconsidered now. The DTT effect on SAM is surprising and important.

    We thank the reviewer for an excellent summary of our work and for highlighting the importance of the findings.

    Reviewer #2 (Public Review):

    Gokol et al. use C. elegans as a model to explore links between stress caused by the compound DTT, diet and growth. They show that effects of DTT toxicity are dependent on diet and link vitamin b12 and the met/SAM cycle through dietary rescue and use of Met/SAM cycle mutants. We do not find that the authors results support their claims. Since DTT is a compound used in labs to induce ER stress and is not naturally present, the general impact is lessened.

    We thank the reviewer for summarizing our work. We have carried out additional experiments (described below in response to the reviewer’s comments) that further support our model of DTT toxicity via the methionine-homocysteine cycle.

    Strengths:

    1. C. elegans is a good model for investigating links between stress and diet. 2) This work includes a mutant screen for animals that regain viability on DTT

    We thank the reviewer for highlighting the strengths of our work.

    Weaknesses:

    1. DTT can affect protein folding in general. While it clearly induces ER stress by disrupting protein folding, it could be affecting a myriad of other processes. Although the authors have a figure to show that DTT toxicity appear to correlate with acdh-1 expression, acdh-1 is part of a pathway that detoxifies propionate (multiple papers from the Walhout lab). Thus, the idea that there is a specific link between the Met/SAM cycle is difficult to sustain. The authors also show that both ER and mito stress reporters are activated, showing the non-specificity of the stress response. An alternate possibility is that SAM is necessary for this histone modifications to activate the stress response to DTT, this is not explored experimentally.

    We would like to underscore that we have multiple lines of evidence that support that DTT toxicity is specifically mediated via the methionine-homocysteine cycle:

    i) Supplementation of vitamin B12 on E. coli OP50 diet completely rescues DTT toxicity (Figure 2). In this context, vitamin B12 works only via methionine synthase providing specificity to the methionine-homocysteine cycle in DTT toxicity. We have now clarified in the revised manuscript that acdh-1 expression levels are sensitive to propionate levels and are only an indirect reporter of vitamin B12 levels.

    ii) Our genetic screen for DTT-resistant mutants resulted in the isolation of 12 alleles of a SAM-dependent methyltransferase (Figure 3). These results (only mutations in one gene recovered) again demonstrate that DTT causes toxicity specifically via the methioninehomocysteine cycle.

    iii) Supplementation of methionine rescues DTT toxicity (Figure 5).

    iv) Supplementation of vitamin B12 rescues toxicity to 5 mM DTT in all the mutants of ER UPR pathways (Figure 6). These results highlight that DTT causes toxicity via the methionine-homocysteine cycle before resulting in lethal ER proteotoxicity.

    v) The mitochondrial UPR induced by DTT is fully rescued by vitamin B12 supplementation (Figure 3—figure supplement 4), suggesting that DTT induces mitochondrial UPR via the methionine-homocysteine cycle.

    Also, Figure 1F is mostly data that has been published previously by the Walhout lab (Watson et al. Cell 2014). Although paper is cited early that section of the results, this figure simply re-presents the previously published data which is not cited in the figure legend or when the data is directly discussed.

    We acquired acdh-1p::GFP data on different bacterial diets so that we could compare acdh-1p::GFP expression levels under our conditions with the development retardation by DTT on different bacterial diets. Moreover, our data also included some bacterial strains that were not part of the Watson et al. 2014 study. Nevertheless, in the revised manuscript, we have now cited Watson et al. 2014 both in the figure legend and the text that describes the results on acdh-1p::GFP.

    As the Apfeld lab (Schiffer, et al. 2020, eLife) have also recently shown, different bacteria can produce metabolites that affect oxidative stress phenotypes, thus conclusions based on dietary effects are complex.

    While it is true that conclusions based on dietary effects could be complex, we were able to demonstrate that the single bacterial metabolite, vitamin B12, could fully recapitulate the dietary effects on DTT. Supplementation of vitamin B12 on E. coli OP50 diet fully rescues DTT toxicity (Figure 2A-B and Figure 1—figure supplement 1).

    1. Although the authors isolate a mutant encoding a putative methyltransferase that is resistant to DTT toxicity, this is of limited use as there is no data showing what this methyl transferase does. Figure 3 also shows the development of drm-1 only on DTT, no wild type is shown.

    We have added data to better characterize the methyltransferase RIPS-1. We provide evidence that loss-of-function (multiple premature stop-codon alleles and knockdown by RNAi) of rips-1 provides resistance to DTT. Overexpression of rips-1 sensitizes the animals to DTT toxicity (Figure 3—figure supplement 3). Further, we demonstrate that rips-1 is required for DTT-mediated SAM depletion (Figure 5F-G). For Figure 3, we have now added data for rips-1 mutants on control (0 mM DTT) bacteria also (Figure 3C-D). These studies establish that the loss-of-function of rips-1 imparts DTT resistance while its overexpression sensitizes the animals to DTT toxicity.

    1. The authors show that methionine rescues DTT effect in wt, and metr-1 backgrounds, but not sams-1. This could also be due to multiple effects. sams-1 animals have defects in membranes, that have not been reported in metr-1. Thus, DTT could simply be more toxic to these animals.

    Both wild-type and metr-1 animals can convert methionine into SAM, but sams-1 mutants cannot. Therefore, methionine supplementation rescues DTT toxicity in the wildtype and metr-1 animals but not in the sams-1 animals. sams-1 animals have defects in the membrane due to low levels of SAM (Walker et al., 2011, PMID: 22035958 ). Compared to sams-1 animals, metr-1 animals do not have defects in membranes, most likely because methionine is not a limiting factor in the diet and metr-1 animals have higher SAM levels than sams-1 animals.

    1. The partial choline rescue was done at 80mM, this is much higher than the previously published amounts (30mM, Brendza, et al. 2007). Even at this high level of choline, the rescue is partial, which brings the rescue in question.

    Several previous studies have also carried out choline supplementation experiments at 80 mM or higher concentrations (Walker et al., 2011, PMID: 22035958; Koh et al., 2018, PMID: 30333136; Giese et al., 2020, PMID: 33016879). Therefore, we decided to use choline rescue at 80 mM. The rescue was complete in N2 and metr-1 animals and partial in sams-1 animals. This suggests that phosphatidylcholine is not the sole SAM product in combating DTT toxicity. It is likely that other SAM-related functions are also involved in attenuating DTT toxicity. We have discussed this possibility in the manuscript (lines 295297): “These results suggested that phosphatidylcholine is a major, but not the sole, SAM product responsible for combating DTT toxicity.”

    Reviewer #3 (Public Review):

    This manuscript studies mechanisms of DTT toxicity in C. elegans, using larval development as readout. The authors find that DTT is not toxic to C. elegans when exogenous vitamin B12 is provided i.e. animals successfully develop. This depends one only one of the two B12 dependent enzymes, methionine synthase metr-1 but not on MMCoA dismutase mmcm-1. A forward genetic screen for mutations that suppress DTT toxicity identified 12 alleles in drm-1 (R08E5.3). An independently generated mutation in drm-1 also showed resistance to DTT, and this was blocked by expression of drm-1 from its own promoters. mRNA of drm-1 and of its homologs R08E5.1, R08F11.4, but not K12D9.1, are induced by DTT. Using metabolite supplementation and mutant analysis, the authors pinpoint SAM deficiency as the key consequence of DTT exposure; in part, this is rescued by choline, suggesting PC deficiency as a key issue. Because ER stress is linked to the 1-carbon cycle, the authors next studied the UPR and found that its activation by DTT is reduced by B12, Met, or choline. Functionally, ire-1 and xbp-1 mutation, but none of the other UPR genes tested, rescued the developmental delay, but only at intermediate (5mM) concentration of DTT, not at a high concentration. The authors propose a model whereby DTT activated drm-1 expression causes SAM depletion, which contributes to DTT toxicity and results in larval arrest.

    The mechanisms identified here is to my knowledge novel and appears very interesting. The experiments in this manuscript are well done and well controlled, and the authors' conclusions are (mostly) well justified by the data. The study provides new insights into the action of DTT toxicity, and pinpoints drm-1 as a new gene implicated in thiol resistance; identifying 12 alleles is extremely compelling as to the key role of this gene (but see below on other methylases). The paper is also well written and explains well the rationale and the reasoning behind the experiments.

    However, I think the authors need to measure SAM levels in the various contexts to actually support the main conclusion drawn her. They also should examine more broadly both the role of thiol agents as well as of methylases related to drm-1, to better define what the specificity of the discovery pathway is, as well as probe more deeply into the role of drm-1 function.

    We thank the reviewer for a very good summary of our work, for highlighting the importance of the findings, and for appreciating the clarity of the manuscript. In the revised manuscript, we have provided additional data on quantifying SAM levels, the toxicity mechanism of two other thiol reagents (β-mercaptoethanol and NAC) with respect to the identified methyltransferase, and the role of rips-1-related methyltransferases on the toxic effects of DTT.

  3. eLife

    Evaluation Summary:

    Thiol agents, such as dithiothreitol (DTT), are toxic to many species, but the mechanisms of toxicity is incompletely understood. In this work, the authors use the animal C. elegans, a small worm, to propose a new mechanisms for how DTT causes organismal growth arrest. Specifically, they suggest that DTT causes reduction in the key molecule S-adenosyl methionine (SAM), which is used as a methyl donor to modify proteins, lipid, and/or other macromolecules. The genetic and supplementation experiments by the authors are compelling, but no direct evidence is provided that SAM levels are indeed lower following exposure of C. elegans to DTT.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)

  4. eLife

    Reviewer #1 (Public Review):

    The current manuscript investigates the mechanisms of DTT toxicity in C. elegans. In a veritable detective story, the authors show that developmental DTT toxicity is determined by the bacterial food source. They realize that the toxicity might be linked to vitamin B12 content of the food and can indeed show that low B12 levels in OP50 bacteria lead to the strongest DTT toxicity while their data suggest that wild-type worms on high B12 bacteria are protected against DTT toxicity. Indeed, B12 supplementation suppresses DTT toxicity on OP50 bacteria and this is dependent on a functional methionine synthase gene. The authors then perform a forward genetic mutagenesis screen to identify DTT resistance loci and hone in on a particular locus encoding a SAM-dependent methyltransferase they name drm-1. drm-1 loss of function protects against DTT toxicity providing support to the idea that it is the depletion of SAM that leads to DTT toxicity in worm development. This is further supported by methionine and choline supplementation experiments. Finally, the authors address the relative contribution of ER stress and SAM depletion in the DTT developmental resistance. Interestingly, they find that UPR signaling mutants affect become DTT hypersensitive only at high but not at low DTT levels. This suggests that SAM depletion is responsible for DTT toxicity at lower concentrations while only at high DTT levels, its effect on the ER becomes toxic.

    In all, this is a well-executed paper that is clear and well written. The finding is relevant as it sheds new light on the DTT mechanism, which is broadly considered an ER stressor acting on disulfide bond formation, which needs to be reconsidered now. The DTT effect on SAM is surprising and important.

  5. eLife

    Reviewer #2 (Public Review):

    Gokol et al. use C. elegans as a model to explore links between stress caused by the compound DTT, diet and growth. They show that effects of DTT toxicity are dependent on diet and link vitamin b12 and the met/SAM cycle through dietary rescue and use of Met/SAM cycle mutants. We do not find that the authors results support their claims. Since DTT is a compound used in labs to induce ER stress and is not naturally present, the general impact is lessened.

    Strengths:

    1. C. elegans is a good model for investigating links between stress and diet.
    2. This work includes a mutant screen for animals that regain viability on DTT

    Weaknesses:

    1. DTT can affect protein folding in general. While it clearly induces ER stress by disrupting protein folding, it could be affecting a myriad of other processes. Although the authors have a figure to show that DTT toxicity appear to correlate with acdh-1 expression, acdh-1 is part of a pathway that detoxifies propionate (multiple papers from the Walhout lab). Thus, the idea that there is a specific link between the Met/SAM cycle is difficult to sustain. The authors also show that both ER and mito stress reporters are activated, showing the non-specificity of the stress response. An alternate possibility is that SAM is necessary for this histone modifications to activate the stress response to DTT, this is not explored experimentally.

    Also, Figure 1F is mostly data that has been published previously by the Walhout lab (Watson et al. Cell 2014). Although paper is cited early that section of the results, this figure simply re-presents the previously published data which is not cited in the figure legend or when the data is directly discussed.

    As the Apfeld lab (Schiffer, et al. 2020, eLife) have also recently shown, different bacteria can produce metabolites that affect oxidative stress phenotypes, thus conclusions based on dietary effects are complex.

    1. Although the authors isolate a mutant encoding a putative methyltransferase that is resistant to DTT toxicity, this is of limited use as there is no data showing what this methyl transferase does. Figure 3 also shows the development of drm-1 only on DTT, no wild type is shown.

    2. The authors show that methionine rescues DTT effect in wt, and metr-1 backgrounds, but not sams-1. This could also be due to multiple effects. sams-1 animals have defects in membranes, that have not been reported in metr-1. Thus, DTT could simply be more toxic to these animals.

    3. The partial choline rescue was done at 80mM, this is much higher than the previously published amounts (30mM, Brendza, et al. 2007). Even at this high level of choline, the rescue is partial, which brings the rescue in question.

  6. eLife

    Reviewer #3 (Public Review):

    This manuscript studies mechanisms of DTT toxicity in C. elegans, using larval development as readout. The authors find that DTT is not toxic to C. elegans when exogenous vitamin B12 is provided i.e. animals successfully develop. This depends one only one of the two B12 dependent enzymes, methionine synthase metr-1 but not on MMCoA dismutase mmcm-1. A forward genetic screen for mutations that suppress DTT toxicity identified 12 alleles in drm-1 (R08E5.3). An independently generated mutation in drm-1 also showed resistance to DTT, and this was blocked by expression of drm-1 from its own promoters. mRNA of drm-1 and of its homologs R08E5.1, R08F11.4, but not K12D9.1, are induced by DTT. Using metabolite supplementation and mutant analysis, the authors pinpoint SAM deficiency as the key consequence of DTT exposure; in part, this is rescued by choline, suggesting PC deficiency as a key issue. Because ER stress is linked to the 1-carbon cycle, the authors next studied the UPR and found that its activation by DTT is reduced by B12, Met, or choline. Functionally, ire-1 and xbp-1 mutation, but none of the other UPR genes tested, rescued the developmental delay, but only at intermediate (5mM) concentration of DTT, not at a high concentration. The authors propose a model whereby DTT activated drm-1 expression causes SAM depletion, which contributes to DTT toxicity and results in larval arrest.

    The mechanisms identified here is to my knowledge novel and appears very interesting. The experiments in this manuscript are well done and well controlled, and the authors' conclusions are (mostly) well justified by the data. The study provides new insights into the action of DTT toxicity, and pinpoints drm-1 as a new gene implicated in thiol resistance; identifying 12 alleles is extremely compelling as to the key role of this gene (but see below on other methylases). The paper is also well written and explains well the rationale and the reasoning behind the experiments.

    However, I think the authors need to measure SAM levels in the various contexts to actually support the main conclusion drawn her. They also should examine more broadly both the role of thiol agents as well as of methylases related to drm-1, to better define what the specificity of the discovery pathway is, as well as probe more deeply into the role of drm-1 function.

  7. Version published to 10.1101/2021.11.16.468906v2 on bioRxiv
  8. Version published to 10.1101/2021.11.16.468906v1 on bioRxiv