PREreview of The herbicide acetochlor causes lipid peroxidation by inhibition of glutathione peroxidase 4

This review arose out of a course for graduate students in the life sciences at UCSF, “Peer Review in the Life Sciences,” which aims to introduce junior scientists to peer review in a critical yet constructive way. The students selected preprints to review, led discussions of them, drafted reviews, and revised them based on feedback from peers and instructors.

Summary

This manuscript sought to answer whether rural exposure to certain herbicides may increase risk of obesity and identify the mechanism of such obesogens if so. First, the authors performed an in vitro screening of six commonly used herbicides for obesogenic properties and identified acetochlor as a potential culprit. This was followed by in vivo validation using zebrafish exposure to acetochlor as a method of verifying in vitro results. Next, the authors measured gene expression and lipidomics of the affected zebrafish to determine which metabolic pathways may be disrupted by acetochlor. Based on this, they proposed dysregulation of lipid oxidation, as mediated by glutathione peroxidase 4 (GPX4), as the main mode of action. To validate this pathway, they perform computational covalent docking and by in vitro GPX4 activity assays. 

The largest issue is the claimed covalent mechanism of action for acetochlor. Chloroacetamides are known to react quite rapidly in solution and due to the structural similarity of acetochlor to metolachlor, covalent reactivity rates should be similar. The initial assessments of obesogenicity instead observed significantly less pronounced effects of metolachlor, instead. Furthermore, while the activity of the peroxidase is decreased, the mechanism of action is unclear. Specifically, we are not sure why the loss of activity could not be attributed to competitive binding rather than covalent modification of the enzyme catalytic selenocysteine. Without using a known noncovalent acetochlor analogue (for example, an acetamide instead of a chloroacetamide), the claim that covalent bond formation is necessary for inhibition of GPX4 is difficult to support. Mass spectrometry analysis of an acetochlor adduct with GPX4 would provide the most direct evidence for the proposed covalent mechanism of action. 

In general, however, we find the authors appropriately use a wide variety of techniques to support their final conclusion that the herbicide acetochlor acts as a peroxidase inhibitor resulting in its obesogenic properties.Together, the authors present a strong case for the increased risk rural communities may face due to commonly used herbicides. This study lays the groundwork for environmental policymakers when considering rural versus urban exposure to pollutants and its link to health outcomes. The focus on a possible mechanism of action also encourages further research into possible overlooked obesogens with similar properties.

However, we would like to draw attention to a number of minor issues.

1. The background does not make a clear distinction between general lipid dysfunction and hyperlipidemia which detracts from clarity of justification in experimental design for assessing lipid accumulation.

2. Given claims in the conclusion about potential similarities between acetochlor and metolachlor, it is difficult to assess the relative effects of the two herbicides without a side-by-side comparison (e.g comparing metalochlor to acetochlor in the GPX4 activity and 4HNE accumulation assays, Figures 5C and 5D).  Figure S3 seems to suggest that the effects observed in metolachlor are significantly reduced as compared to acetochlor (there is also a typo in Figure S3 “ametolachlor”). 

3. The images for AdipoRed and DNA staining are unavailable, and we hope the authors would consider including those in the final manuscript as strong evidence for their control experiments verifying cellular integrity.

4. We were unsure why the positive control rosiglitazone was needed to normalize the AdipoRed and ORO staining in both mouse adipocytes and zebrafish (Figures 1 and 2).  We assumed samples were normalized as a percentage of the vehicle control sample, but additional clarification would be helpful to the readers.

5. We were also confused why EC50 was evaluated given that the measured sample would be inherently resistant to acetochlor due to their ability to survive for the days of treatment.

6. While we appreciate the authors’ qualitative interpretation to account for the surprising decrease in lipid accumulation at higher concentrations of acetochlor, comparison with the control group to evaluate food intake would help reinforce the conclusion. 

7. The organization of transcriptomic and lipidomic data is quite confusing as it was not immediately clear how production of reactive oxygen species is linked to lipid metabolism. 

8. We believe a short clarification at this point would also help improve the narrative and further reinforce the conclusions reached through the lipidomics experiment which identified increased lipid peroxidation caused by oxidative stress. 

9. There are some minor errors in referencing figures (e.g downstream analysis of NRF2-related genes should refer to figures S6D/S6E, not S5D/S5E)

10. Corrections to the chemical structure in Figure 5B (missing a carbon atom between the carbonyl of acetochlor and cysteine residue)

11. We are unsure about the interpretation of the GPX4 lysate results: specifically, what is the sensitivity of the assay? Some discussion would help contextualize these results.

Competing interests

The author declares that they have no competing interests.