Reducing the Periplasmic Glutathione Content Makes Escherichia coli Resistant to Trimethoprim and Other Antimicrobial Drugs

ABSTRACT Although glutathione (GSH) has been shown to influence the antimicrobial effects of many kinds of antibiotics, little is known about its role in relation to trimethoprim (TMP), a widely used antifolate. In this study, several genes related to glutathione metabolism were deleted in different Escherichia coli strains (i.e., O157:H7 and ATCC 25922), and their effects on susceptibility to TMP were tested. The results showed that deleting gshA, gshB, grxA, and cydD caused TMP resistance, and deleting cydD also caused resistance to other drugs. Meanwhile, deleting gshA, grxA, and cydD resulted in a significant decrease of the periplasmic glutathione content. Supplementing exogenous GSH or further deleting glutathione importer genes (gsiB and ggt) restored TMP sensitivity to ΔcydD. Subsequently, the results of quantitative-reverse transcription PCR experiments showed that expression levels of acrA, acrB, and tolC were significantly upregulated in both ΔgrxA and ΔcydD. Correspondingly, deleting cydD led to a decreased accumulation of TMP within bacterial cells, and further deleting acrA, acrB, or tolC restored TMP sensitivity to ΔcydD. Inactivation of CpxR and SoxS, two transcriptional factors that modulate the transcription of acrAB-tolC, restored TMP sensitivity to ΔcydD. Furthermore, mutations of gshA, gshB, grxA, cydC, and cydD are highly prevalent in E. coli clinical strains. Collectively, these data suggest that reducing the periplasmic glutathione content of E. coli leads to increased expression of acrAB-tolC with the involvement of CpxR and SoxS, ultimately causing drug resistance. To the best of our knowledge, this is the first report showing a linkage between periplasmic GSH and drug resistance in bacteria. IMPORTANCE After being used extensively for decades, trimethoprim still remains one of the key accessible antimicrobials recommended by the World Health Organization. A better understanding of the mechanisms of resistance would be beneficial for the future utilization of this drug. It has been shown that the AcrAB-TolC efflux pump is associated with trimethoprim resistance in E. coli clinical strains. In this study, we show that E. coli can sense the periplasmic glutathione content with the involvement of the CpxAR two-component system. As a result, reducing the periplasmic glutathione content leads to increased expression of acrA, acrB, and tolC via CpxR and SoxS, causing resistance to antimicrobials, including trimethoprim. Meanwhile, mutations in the genes responsible for periplasmic glutathione content maintenance are highly prevalent in E. coli clinical isolates, indicating a potential correlation of the periplasmic glutathione content and clinical antimicrobial resistance, which merits further investigation.

3 The authors claimed that after knocking out the cydD gene, the susceptibility to other antibiotics is reduced. Does that determine the sensitivity of other antibiotics in gene complement strains? 4 Efflux pump is a general antimicrobial mechanism against many kinds of antibiotics. So, the resistance of drugs that are the substrate of this mechanism would be all influenced by interrupting the GSH metabolism? 5 In the analysis of the mutations of genes involved in GSH metabolism of the clinical E. coli strain, it should be emphasized that mutations may not necessarily mean the loss the function of the gene.

Reviewer #2 (Comments for the Author):
Deng and colleagues reported the reduced accumulation of periplasmic GSH may lead to antibiotic resistance by up-regulating several resistance genes such as acrAB-tolC. This work updates the understanding towards how GSH metabolism affects the antimicrobial resistance of bacteria and sheds the light on the potential association between bacterial metabolic progress and their non-metabolic phenotypes. The experiments were well designed and the manuscript is of interest to the researchers in the microbiology field. However, there are some major concerns to be addressed. Major: 1. The authors provided convincing but only phenotypical evidence to elucidate the linkage between periplasmic GSH and drug resistance. How will be the potent machinery of bacterial GSH enhance the antibiotic tolerance via the SOS-efflux pathway? As we know soxS is highly-related to the intracellular ROS level, which generally quenched by GSH. Will the ROS be a checkpoint between periplasmic GSH and upregulated expression of efflux? The authors better perform experiments to see whether ROS involved in this mechanism. 2. The authors indicate that it is the periplasmic not cytoplasmic GSH functionalized the antibiotic resistance. It will be of great significance to decipher this site-dependent effect of GSH. 3. I noticed that there are the slight differences in resistant phenotypes among mutant from different E. coli strains. Could author explain why choose W3100 as the model strain throughout the manuscript. 4. The manuscript was not given in proper scientific English and should be tidied up prior to resubmission. Minor: Line 54: 'living creatures? Line 61: occurring to Line 67-68: synthesized through 2-step reaction Line 77: besides its role Line 144: 'behaved'? Line 125-136: this section appears with weak consistency with other parts, and associated discussion may be necessary Line 195: what does 'slight' stand for? Is it of statistical significance or just a numerical decrease?
Staff Comments:

Preparing Revision Guidelines
To submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex. Go to Author Tasks and click the appropriate manuscript title to begin the revision process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Here are a few examples of required updates that authors must address: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER. • Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file. • Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me. If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If you would like to submit an image for consideration as the Featured Image for an issue, please contact Spectrum staff.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publication Fees, including supplemental material costs, please visit our website.
Corresponding authors may join or renew ASM membership to obtain discounts on publication fees. Need to upgrade your membership level? Please contact Customer Service at Service@asmusa.org.
Thank you for submitting your paper to Microbiology Spectrum.

Reducing the Periplasmic Glutathione Content Makes Escherichia coli Resistant to Trimethoprim and Other Antimicrobial Drugs
Major concerns: Deng and colleagues reported the reduced accumulation of periplasmic GSH may lead to the antibiotic resistance by up-regulating several resistance genes such as acrAB-tolC. This work updates the understanding towards how GSH metabolism affects the antimicrobial resistance of bacteria and sheds the light on the potential association between bacterial metabolic progress and their non-metabolic phenotypes. The experiments were well designed and the manuscript is of interests to the researchers in microbiology field. However, there are some major concerns to be addressed before consideration for acceptance. Major: 1. The authors provided convincing but only phenotypical evidences to elucidate the linkage between periplasmic GSH and drug resistance. How will be the potent machinery of bacterial GSH enhance the antibiotic tolerance via SOS-efflux pathway? As we know soxS is highly-related to the intracellular ROS level, which generally quenched by GSH. Will the ROS be a checkpoint between periplasmic GSH and upregulated expression of efflux? The authors better perform experiments to see whether ROS involved in this mechanism.
2. The authors indicate that it is the periplasmic not cytoplasmic GSH functionalized the antibiotic resistance. It will be of great significance to decipher this site-dependent effect of GSH.
3. I noticed that there are the slight differences in resistant phenotypes among mutant from different E. coli strains. Could author explain why choose W3100 as the model strain throughout the manuscript. (1) Original comment: The authors need to notice that the MIC value of TMP resistance of Enterobacteriaceae should be greater than or equal to 16 μg/mL according to CLSI 2020. The standard strain ATCC 25922 was not used as a reference for the susceptibility testing. The agar used in the drug sensitivity test should be MHA, according to the standard methods given by CLSI. Therefore, the statement of TMP resistance (MIC=1.28 μg/mL) in current study needs to be further verified.

Reasons and modifications:
Thank you for your comments. We obtained the standard strain ATCC 25922 and also the MHA medium. Unfortunately, we noticed that the instruction for performing the TMP susceptibility using the MHA medium (CLSI2020) says that the medium may contain antagonists of the drug. Our data also showed that the TMP MIC of E. coli is much higher in the MHA medium than that in the LB medium (Table 1). Therefore, we constructed four single gene deletion mutants based on ATCC 25922 (ΔgshA, ΔgshB, ΔgrxA, and ΔcydD), and found that all those four mutants were more resistant to TMP (Table 2). (2) Original comment: The standard curve method or comparative Ct value method for RT-PCR was not mentioned.

Reasons and modifications:
Thank you for the reminder. The manuscript has been revised according to your suggestion (line 279 of the revised MS).
(3) Original comment: The authors claimed that after knocking out the cydD gene, the susceptibility to other antibiotics is reduced. Does that determine the sensitivity of other antibiotics in gene complement strains? been revised according to your suggestions (lines 124-129 of the revised MS). We also tested the sensitivities to other antibiotics of the complemented strain along with the ΔcydD mutant. The results showed that (Table S2 of the revised manuscript), complementing ΔcydD with an intact cydD gene could completely restore the susceptibility to TMP, largely restore the susceptibility to kanamycin and neomycin, but only partially restore the susceptibility to gentamycin. Susceptibility to chloramphenicol, the complemented strain was not tested since the pCA24N vector contains a chloramphenicol resistance cassette (Table S2). (4) Original comment: Efflux pump is a general antimicrobial mechanism against many kinds of antibiotics. So, the resistance of drugs that are the substrate of this mechanism would be all influenced by interrupting the GSH metabolism?

Reasons and modifications:
Reasons and modifications: Thank you for your suggestion. Based on the obtained data, it seems that resistance of drugs that are substrates of the AcrAB-TolC efflux pump would be all affected when the periplasmic content of GSH in E. coli is decreased. But GSH also play an important role in maintaining redox state of the cytoplasm, and this also needs to be considered.   (1). In brief, bacterial cells were grown with an initial OD 600 of 0.1 at 37 ℃ in LB medium to mid-log phase (OD 600 ~0.8) , harvested by centrifugation (3,000 g for 10 min at 4℃), washed twice with phosphate buffered saline (PBS, pH 7.0), resuspended in 1 mL 10 µm/L H 2 DCFDA solution (dissolved in PBS buffer). After incubation at 37℃ for 30 min, bacterial cells were harvested by centrifugation (3,000 g for 10 min), washed twice with PBS buffer, then resuspended in 0.5 mL PBS buffer. After that, the cell density (OD 600 ) was determined, and the fluorescence intensity was measured by a SyergyH1 Hybrid reader (BioTek, USA) (excitation, 488 nm; emission, 525 nm). The results showed that although the intracellular ROS level significantly increased upon TMP treatment, no statistical difference could be observed between the WT and ΔcydD, as shown in the below figure.