Novel Epigallocatechin Gallate (EGCG) Analogs with Improved Biochemical Properties for Targeting Extracellular and Intracellular Staphylococcus aureus

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article requires the following improvements before it can be published：

1. The methods of ADME properties needs to be described in detail. And the ADME properties of MCC-2 must be investigated.

2. The MIC values of tested compounds should be summarized in a table.

3. The syneristic effects of tested compounds with antibiotics should be used checkboard assay. And the results should be diagrammatic (e.g. Front. Microbiol. 15:1361508.)

4. Section 3.4 "MCC-1 and MCC-2 were tested at ¼ x to 4x MIC and were shown to delay or inhibit growth of USA300 MRSA in a dose-dependent manner over 24 h ". But Figure 3 A&B only showed  ¼ x to MIC. Please describe Figure 3 in detail.

5. In discussion section, the potetial targets of MCC analogs should be discussed.

 

Author Response

1. The methods of ADME properties needs to be described in detail. And the ADME properties of MCC-2 must be investigated.

 

Line 267: Data for MCC-2 has been added to table 2

 

Line 144: Details and citations about ADME tests are added

 

2. The MIC values of tested compounds should be summarized in a table.

 

Line 279: Table 3 has been added.

 

3. The syneristic effects of tested compounds with antibiotics should be used checkboard assay. And the results should be diagrammatic (e.g.  Microbiol. 15:1361508.)

 

Line 319: Figure 3 has been added.

 

4. Section 3.4 "MCC-1 and MCC-2 were tested at ¼ x to 4x MIC and were shown to delay or inhibit growth of USA300 MRSA in a dose-dependent manner over 24 h". But Figure 3 A&B only showed ¼ x to MIC. Please describe Figure 3 in detail.

 

 Line 298: Figure 2 has been revised.

 

5. In discussion section, the potential targets of MCC analogs should be discussed.

 

We hypothesized that the potential targets of EGCG are also the potential targets of MCC-1 and MCC-2 as the MCC analogs were designed and synthesized to retain the active structure of EGCG except for modifications to improve its biochemical properties. We reasoned that the improved activity of MCC-1 and MCC-2 over EGCG is due to their improved biochemical properties allowing them to reach higher concentrations at the site of action where the target is present or have more efficient binding to the target in question rather than an altered target. The discussion section has been revised to more clearly state this (lines 369-392).

 

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

 

Thank you for the opportunity to review your manuscript. Based on my expertise, I have some suggestions that I believe can enhance the understanding of your work.

 

1.The authors mention that the bacteria used in the study have different resistance profiles. I suggest including specific information about which antibacterials or classes of antibacterials these microorganisms are resistant to. Additionally, please provide a reference or describe how the isolates were previously identified.

 

2. When describing the methodology for evaluating antimicrobial activity, it is essential to clarify who the controls are in the experiments. For instance, you state that DMSO was used as a solvent for the solubilization of MCC-1 and MCC-2 at a concentration of 70%. Was the potential effect of DMSO on bacterial growth considered? What is the final concentration of DMSO in the assays?

 

3. I recommend that the authors specify all concentrations used in each test to ensure the methodology section contains sufficient information for readers to replicate the results.

 

4. In the results section, it appears that only two out of the ten different molecules were used in assays following the antimicrobial activity tests. This information should be clearly presented in the methodology section for better clarity.

 

5. Lastly, please enhance the figures in terms of size for improved visibility.

 

Thank you again for the opportunity to review your work. I look forward to seeing the revised manuscript.

 

Best regards.

Author Response

Responses to Reviewer 2:

 

1. The authors mention that the bacteria used in the study have different resistance profiles. I suggest including specific information about which antibacterials or classes of antibacterials these microorganisms are resistant to. Additionally, please provide a reference or describe how the isolates were previously identified.

 

Line 170: See references for two papers previously published from our laboratory from which the isolates were previously identified

 

Information about the isolates used have been attached as supplemental data

 

 

2. When describing the methodology for evaluating antimicrobial activity, it is essential to clarify who the controls are in the experiments. For instance, you state that DMSO was used as a solvent for the solubilization of MCC-1 and MCC-2 at a concentration of 70%. Was the potential effect of DMSO on bacterial growth considered? What is the final concentration of DMSO in the assays?

 

Line 190 and 205: More information on DMSO controls was added.

 

We ran DMSO controls within all our assays. Although these controls are not listed, DMSO was not found to have a significant effect on bacterial growth when used at concentrations equivalent to the DMSO concentrations within MCC-1 and MCC-2 treatments. Additionally, to rule out any synergy with DMSO and EGCG, antimicrobial assays were also run with EGCG dissolved in equivalent concentrations of DMSO, but no differences between water dissolved EGCG and DMSO dissolved EGCG was observed.

 

The final concentrations of DMSO are as followed:

 

200ug/ml MCC-1/MCC-2= 10% DMSO

100ug/ml MCC-1/MCC-2= 5% DMSO

50ug/ml MCC-1/MCC-2= 2.5% DMSO

25ug/ml MCC-1/MCC-2= 1.25% DMSO

12.5ug/ml MCC-1/MCC-2= 0.625% DMSO

6.25ug/ml MCC-1/MCC-2= 0.3125% DMSO

3.13ug/ml MCC-1/MCC-2= 0.15625% DMSO

1.57ug/ml MCC-1/MCC-2= 0.078125% DMSO

 

 

 

3. I recommend that the authors specify all concentrations used in each test to ensure the methodology section contains sufficient information for readers to replicate the results.

 

Line 179, 185, 298, 321: Figures and methods have been revised to better show the concentrations that were used in each test.

 

4. In the results section, it appears that only two out of the ten different molecules were used in assays following the antimicrobial activity tests. This information should be clearly presented in the methodology section for better clarity.

 

Line 230: Section 3.1 states why MCC-1 and MCC-2 were the only compounds to move forward in our assays.

 

Line 176-225: The methodology section has been revised to state EGCG analogs when all analogs in general were used, whereas MCC-1 and MCC-2 are explicitly stated, when only they were in use.

 

5. Lastly, please enhance the figures in terms of size for improved visibility.

 

The quality of the figures and size of the figures have been revised accordingly.

Reviewer 3 Report

Comments and Suggestions for Authors

Grosso et al. manuscript aims to evaluate the effects of structural modifications of EGCG on the drug-like properties and on the antistaphylococcal activity. The study is well designed, the presentation and discussion of the results are elegant. The discussion on the possible mechanism of action of MCC-1 and MCC-2 is quite interesting. The M&M section should be improved including more details on the clinical isolates. Some strains codes such as HH35 appear in the Result section but were not described in the M&M section. Also, two points should be discussed by the authors: the lack of a methicillin susceptible reference strain and the toxicity of MCC-1 and MCC-2 to mammal cells.

Specific comments:

Please, include more information on clinical strains (susceptibility to antibacterials, i.e. methicillin).

Include more information on MRSA strain HH35 antimicrobial susceptibility and include references.

The authors reported that they used a methicillin resistant strain, why a methicillin susceptible strain was not included in the study.

What were the MRSA and MSSA strains used for checkerboard assays?

In line 225, the authors cite ‘MRSA strain HH35’, is this strain a clinical isolate? Please include the identification of the strains in the M&M section. Same comment for HH36, HH70 and LAC 164 cited in Figure 2. What is the meaning of HH and LAC?

Line 269 – “Kup5 cells infected with our reference MRSA strain or our clinical MSSA strain (HH70)…”, consider deleting “our” in both periods.

How was the subinhibitory concentration 6.25μg/ml selected? Include in the M&M section.

In line 418, the authors suggest that EGCG analogs were not potentially toxic to Kupfer cells. Is there any further information on MCC-1 and MCC-2 toxicity to mammal cells? Please, discuss this point.

Author Response

1. Please, include more information on clinical strains (susceptibility to antibacterials, i.e. methicillin).

 

A supplementary table has been added that includes more information on the used clinical strains.

 

2. Include more information on MRSA strain HH35 antimicrobial susceptibility and include references.

 

More information on the MRSA strain HH35 can be found in the added supplementary table. See below reference previously published on this strain.

 

https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2020.609280/full

 

The MIC for HH35 is 1ug/ml for both vancomycin and daptomycin as tested by our lab.

 

 

 

3. The authors reported that they used a methicillin resistant strain, why a methicillin susceptible strain was not included in the study.

 

As stated in the methods, both MRSA and MSSA strains were used in this study. For MIC assays, both USA300 MRSA and 4 clinical strains (two of which that are MSSA strains) were used. For the intracellular killing assay, both USA300 MRSA and HH70 MSSA were used. The assays which did not include an MSSA strain were the growth curve assay as the intention of this was to analyze the timeline and kinetics of the EGCG analogs activity and not to test its efficacy across multiple strains. The other assay which did not include an MSSA strain was the synergy assay, this is because previous papers (Zhao et al., 2001) states that EGCG’s synergy comes from its ability to make MRSA strains susceptible to beta lactam antibiotics, which is the main activity in which we wanted to test. Additionally, USA300 MRSA is used as the primary strain for this manuscript as it is one of the most prevalent strains in the USA and also most widely studied.

 

4. What were the MRSA and MSSA strains used for checkerboard assays?

 

Line 197: The methods section was corrected as only one strain was used for the checkerboard assay per the reasoning in comment 3 above. The methods section was also edited to include “(USA300)”

 

5. In line 225, the authors cite ‘MRSA strain HH35’, is this strain a clinical isolate? Please include the identification of the strains in the M&M section. Same comment for HH36, HH70 and LAC 164 cited in Figure 2. What is the meaning of HH and LAC?

 

Information about the clinical strains have been added to a supplemental table in supplemental information. HH and LAC are designation representing for strains based on study site.

 

6. Line 269 – “Kup5 cells infected with our reference MRSA strain or our clinical MSSA strain (HH70)…”, consider deleting “our” in both periods.

 

Line 324: The sentence has been edited to say “Kup5 cells infected with reference MRSA strain USA300 or clinical MSSA strain HH70”

 

7. How was the subinhibitory concentration 6.25μg/ml selected? Include in the M&M section.

 

Line 222: Information regarding the selection of 6.25ug/ml was added to the methodology.

 

8. In line 418, the authors suggest that EGCG analogs were not potentially toxic to Kupfer cells. Is there any further information on MCC-1 and MCC-2 toxicity to mammal cells? Please, discuss this point.

 

Line 479: In preliminary cell viability testing with Kup5 cells, it was shown that MCC-1 and MCC-2 dissolved in DMSO showed decreases in cell viability at concentrations 12.5ug/ml and above, whereas 6.25ug/ml did not have an effect on cell viability. However, this matches the cell viability of DMSO alone which decreased the cell viability of Kup5 cells to an equal percentage when present at the same concentration within the MCC-1 and MCC-2 treatments (12.5ug/ml of MCC-1/MCC-2 contains 0.625% DMSO). Since the cell viability information between MCC-1/MCC-2 treatments dissolved in DMSO and the DMSO control alone are not different from each other and since the parent compound EGCG is not found to reduce cell viability at concentrations relevant to this manuscript, we have hypothesized that the reduction in cell viability seen is due to DMSO rather than the analogs themselves

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript by Riley Grosso et al. entitled ‘Novel epigallocatechin gallate (EGCG) analogs against extracellular and intracellular Staphylococcus aureus’ depicted a novel design of the EGCG analogs against the SA intracellular and extracellular. The authors substituted the OH group with different groups, aimed to increase the lipophilicity of the compounds. The novel-designed lead analogs MCC-1 and MCC-2 perform well in the SA inhibition tests under various criteria. Additionally, the lead analogs act synergistically with antibiotics against SA. 

 

However, I am very curious whether there have been any studies on the EGCG analogs in anti-bacterial, as far as I know, some modifications on the EGCG were used for anti-cancer. But I have no idea if there are some trials in the anti-SA. Could you please add some related background to the manuscript? 

 

In addition, in Figure 5, I cannot find out the significant effects between antibiotics only versus antibiotics-EGCG or analogs, if the statistical analysis can be included in this part, it would be better.

 

 

I wonder how EGCG and analogs function. I found one paper titled ‘Epigallocatechin Gallate Induces Upregulation of the Two-Component VraSR System by Evoking a Cell Wall Stress Response in Staphylococcus aureus’, the paper indicates that the two-component VeaSR system is involved in modulating the cell response to EGCG, so do you believe the analogs you designed would take same mechanism to work? The conclusion would be interesting if qRT-PCR could be used to compare the expression level of some specific genes, such as vraR.  Could you provide a reliable hypothesis in the discussion section? 

Author Response

1. However, I am very curious whether there have been any studies on the EGCG analogs in anti-bacterial, as far as I know, some modifications on the EGCG were used for anti-cancer. But I have no idea if there are some trials in the anti-SA. Could you please add some related background to the manuscript? 

 

Line 67: New introduction section has been added detailing previous methods that have been used to modify EGCG for antibacterial action.

 

2. In addition, in Figure 5, I cannot find out the significant effects between antibiotics only versus antibiotics-EGCG or analogs, if the statistical analysis can be included in this part, it would be better.

 

Line 341: Figure 4 has been revised with added statistical analysis

 

3. I wonder how EGCG and analogs function. I found one paper titled ‘Epigallocatechin Gallate Induces Upregulation of the Two-Component VraSR System by Evoking a Cell Wall Stress Response in Staphylococcus aureus’, the paper indicates that the two-component VeaSR system is involved in modulating the cell response to EGCG, so do you believe the analogs you designed would take same mechanism to work? The conclusion would be interesting if qRT-PCR could be used to compare the expression level of some specific genes, such as vraR.  Could you provide a reliable hypothesis in the discussion section? 

 

The EGCG analogs are based on the active structure of EGCG with structural modifications to improve EGCG’s biochemical properties, EGCG’s major downfall as a potential therapeutic compound. The EGCG analogs were designed to maintain EGCG’s action and thus it is hypothesized that the potential targets of EGCG are also the potential targets of MCC-1 and MCC-2, as well as future analogs that will be synthesized. The improved activity of MCC-1 and MCC-2 over EGCG is not hypothesized to come from a change in target but rather improved biochemical properties that may allow it to reach higher concentrations in site where the target is present, and/or have more efficient binding to the target in question. The discussion section has been revised to more clearly state this. (lines 394-417) We have future studies planned to more thoroughly research the mechanism of action of EGCG and its analogs, however this study is a proof-of-concept study showing that our means of structural modification has improved on the antistaphylococcal activity EGCG.

 

Reviewer 5 Report

Comments and Suggestions for Authors

Here, Grosso et al. try to describe potential anti-infective EGCG analogs for combating S. aureus during infections. I went point-to-point throughout this paper and the following points need to be addressed:

Points of major concerns:

1.    The missing point in this manuscript is the lack of being bound together in one story. I do not understand how do the analogs inhibit the growth of the used strain. If the analogs really inhabited the growth, what was the significance of the downstream killing experiments?

2.    The findings in the paper are all a description of a quantitative analysis. I wondering, why did the authors use the USA300 strains that harbor a GFP promoter? What was the significance of using such a reporter strain? Did the author preserve the GFP expression using antibiotics in the study? How could the authors ascertain that such genetic manipulation did not affect the phenotypic characteristic of this strain? It could be that was the only available strain; however, describing the strain like that (even it is not needed to mention this) means a little proofreading of this manuscript.

3.    In the intracellular killing assay, and according to the describtion of methodology, there is no proof, that the used macrophages were simultaneously infected with the bacterial cells. How did the authors ascertain this? Additionally, after 10 and 24 hrs of incubating Kupfer cells engulfing bacterial cells, the intracellular bacteria cells will try to escape and proliferate outside as long as there is no gentamycin or lysostaphin in the medium. This means what the author performed and described as intracellular killing strategy could be misleading. How could the authors explain this?

Points of minor concerns:

1.    The title is somehow wide. It needs to be more specific.

2.    General comment: please indicate in all experiments how many biological and/or technical replicates used.

3.    Sentence 162, please indicate what the abbreviation “CAMHB” stands for?

4.    Is there a reason of using CAMHB specifically and not for example the BHI medium?

5.    Sentence 204-205, please indicate that this finding was only tested in vitro.

6.    Sentence 207-208 looks complicated especially using the two words “were facilitated”.

7.    In table 2, please make the titles in the tables (In-Vitro ADME-Assay, Target Value, EGCG and MCC-1) bold.

8.    Figure 2 is visualized in a difficult way. Please change the colors or make the bars distinguishable. Describtion of the figure should also be adjusted.

9.    Figure 3, the quality of graphs looks bad. Please replace them using a better format with a higher resolution. Are these values the original OD₆₀₀ values or the log10 values of the OD₆₀₀ measurement?

10. Why were the growth curve assays performed only to the USA300 MRSA strain and not to the MSSA strain (HH70)?

11. Line 236, the author sated that they the analogs were dissolved in 70% DMSO, therefore, they performed growth curves using DMSO; however, they used a range of 1.25-0.078125%. Please indicate what was the percentage used for each bacterial culture. What was the initial percentage of DMSO used upon adding the analogs dissolved in 70% DMSO?

12. Sentences 249-250, please rephrase this sentence. It looks like these data were used before and not the original findings of this paper.

13. Designing the intracellular killing experiments is not correct. What do you know about the growth kinetics of the clinical MSSA strain (HH70)? The findings shown here could be due to a defective growth phenotype of this strain.

14. What were the parameters of interpreting the data of the checkboard synergy assays?

 

 

Comments for author File: Comments.pdf

Author Response

1. The missing point in this manuscript is the lack of being bound together in one story. I do not understand how do the analogs inhibit the growth of the used strain. If the analogs really inhabited the growth, what was the significance of the downstream killing experiments?

 

The antibacterial activity of EGCG has been widely studied in many previous papers, with many papers showing that EGCG can inhibit growth of S. aureus, amongst other bacterial pathogens. The proposed mechanism of action is through cell wall and cell membrane damage, possibly through the intercalation of EGCG within the cell membrane of S. aureus (as hypothesized by Kitichalermkiat et al., 2020) and also the binding of EGCG to peptidoglycan within the S. aureus cell wall, as suggested in Zhao et al, 2001. The EGCG analogs are designed to improving the biochemical properties of EGCG while preserving the active moeity of EGCG for antibacterial action. Thus, the EGCG analogs are hypothesized to share the same molecular targets as EGCG. Future studies are planned to further investigate the mechanism of action of EGCG and EGCG analogs but this paper serves as a proof of concept study demonstrating that structural changes can overcome the unfavorable biochemical properties of EGCG to make it a better candidate for targeting intracellular S. aureus. We have thus revised the discussion section to reflect the above.

 

Downstream killing experiments were performed to specifically evaluate the ability for EGCG analogs to kill intracellular S. aureus within Kupffer cells, a niche shown by others to harbor surviving bacteria that contribute to persistent bacteremia and dissemination despite treatment with standard antibiotic therapy. We first performed experiments to determine MIC, Growth Curve and Synergy assays to screen our analogs to ensure that the structural modifications made preserved the antibacterial activity of EGCG against extracellular S. aureus. Thereafter, intracellular killing experiments were performed to address our primary focus which showed that MCC-1 and MCC-2 were able to potentiate the killing activity of relevant antibiotics against intracellular S. aureus, whereas EGCG could not. Thus, we have shown here that the chemical modifications we made to EGCG, successfully allowed EGCG analogs to enter the cell in sufficient concentrations to enhance the intracellular clearance of S. aureus when combined with antibiotics. Our findings show promise for further development of the EGCG analogs in that prior papers have found that S. aureus persists inside Kupffer cells due in part to the bacterial virulence-mediated dampening of Kupffer cell function as well as reduced intracellular accumulation of antibiotics. The aim of this study was to synthesize an EGCG analog which could function as an adjunctive agent for use with an antistaphylococcal antibiotic to effectively eradicate intracellular bacteria to eliminate the risk of persistence or systemic dissemination of any surviving bacteria.

 

 

2. The findings in the paper are all a description of a quantitative analysis. I wondering, why did the authors use the USA300 strains that harbor a GFP promoter? What was the significance of using such a reporter strain? Did the author preserve the GFP expression using antibiotics in the study? How could the authors ascertain that such genetic manipulation did not affect the phenotypic characteristic of this strain? It could be that was the only available strain; however, describing the strain like that (even it is not needed to mention this) means a little proofreading of this manuscript.

 

The GFP promoter strain was used both as validation for our experimental design and to inform future planned experiments. In previous experiments to validate our intracellular killing protocol, the GFP strain was used to show successful infection of our Kupffer cells and was used to optimize infection timing, as the GFP allowed us to visualize the success entry of the GFP bacteria into the Kupffer cells during the 1.5-hour infection timeframe. Additionally, when tested at higher concentrations of EGCG (not used in this paper), EGCG’s coloration makes it difficult to determine EGCG MIC visually per CLSI guidelines (observation by eye); thus, GFP afforded us the option for quantifying growth by spectrophotometric measurement. Additionally, we plan to perform confocal microscopy as well as quantitative fluorescent lifetime imaging microscopy for future mechanistic studies in which we will relate our results and timepoints from the imaging to that of our intracellular killing assay.

It is notable that in previous experiments we have performed MIC, growth curve and intracellular killing assays with the parent strain (USA300 strain) without the GFP reporter and the two have comparable results.

 

3. In the intracellular killing assay, and according to the description of methodology, there is no proof, that the used macrophages were simultaneously infected with the bacterial cells. How did the authors ascertain this? Additionally, after 10 and 24 hrs of incubating Kupfer cells engulfing bacterial cells, the intracellular bacteria cells will try to escape and proliferate outside as long as there is no gentamycin or lysostaphin in the medium. This means what the author performed and described as intracellular killing strategy could be misleading. How could the authors explain this?

 

As described above, the intracellular killing experiment has been optimized by use of the GFP strain and is a well-established and validated assay in the field. Utilizing fluorescent microscopy we have identified that the GFP USA300 is able to successfully inoculate the Kup5 cells. Additionally, T0 timepoints are taken for all replicates of this experiments and across all treatments, these T0 counts show how much bacteria have entered the cell before treatment and are all comparable across all replicates, showing consistent infection of the cells.

 

In regards to the extracellular proliferating bacteria, for treatments in which antibiotics or EGCG are in use, the concentrations tested are greater than the MIC for the USA300 and HH70 strains, thus meaning that growth of any escaped bacteria is effectively inhibited. Furthermore, before cells are lysed and before cell lysate is plated at each time point to enumerate the intracellular bacterial burden, the cells are washed gently 3 times with PBS to remove any extracellular bacteria, to avoid extracellular contamination. In the case of MCC-1 and MCC-2, sub-MICs were tested to avoid use of high DMSO concentrations for dissolution that could affect Kupffer cell viability. Despite the use of sub-MICs, both analogs were able to potentiate the intracellular clearance of bacteria when combined with antibiotics.

 

 

Reviewer 5 Minor Concerns:

 

1. The title is somehow wide. It needs to be more specific.

 

Line 1: Title has been changed.

 

2. General comment: please indicate in all experiments how many biological

and/or technical replicates used.

 

The materials and methods section has been edited to include information on replicates.

 

3. Sentence 162, please indicate what the abbreviation “CAMHB” stands for?

 

Line 187: The full name of CAMHB (Cation-adjusted Mueller-Hinton broth) has been added.

 

4. Is there a reason of using CAMHB specifically and not for example the BHI

medium?

 

CAMHB is the standard medium for minimum inhibitory concentration testing per CLSI methodology and S. aureus has been shown to grow well in CAMHB. Generally, BHI is used for fastidious organisms that do not grow well in CAMHB and require extra nutrients to grow; it is not as widely used and/or standardized for MIC testing as CAMHB.

 

5. Sentence 204-205, please indicate that this finding was only tested in vitro.

 

Line 246: “in-vitro” has been added to the header for section 3.2

 

6. Sentence 207-208 looks complicated especially using the two words “were

facilitated”.

 

Line 249: Sentence has been edited to be more clear

 

7. In table 2, please make the titles in the tables (In-Vitro ADME-Assay, Target

Value, EGCG and MCC-1) bold.

 

Line 267: the titles in Table 2 have been bolded

 

8. Figure 2 is visualized in a difficult way. Please change the colors or make the

bars distinguishable. Describtion of the figure should also be adjusted.

 

Line 298: Per another reviewer’s suggestion, Figure 2 has been changed into a table to make it more easily understandable.

 

9. Figure 3, the quality of graphs looks bad. Please replace them using a better

format with a higher resolution. Are these values the original OD600 values or

the log10 values of the OD600 measurement?

 

Line 298: Figure 2 (previously figure 3) has been remade and saved at a higher resolution. These values are the original OD600 values.

 

 

10. Why were the growth curve assays performed only to the USA300 MRSA strain

and not to the MSSA strain (HH70)?

 

The main goal for the growth curve assay was to show the concentration-dependent inhibition of bacterial growth relative to the measured MIC for MCC-1 and MCC-2. Additionally, we used the growth curve assay to show that bacterial growth was not affected at the corresponding concentrations of DMSO required for dissolution of MCC-1 and MCC-2. We believe that to categorize these findings, it is not necessary to perform them on different strains for the primary categorization of these compounds.

 

11. Line 236, the author sated that they the analogs were dissolved in 70% DMSO,

therefore, they performed growth curves using DMSO; however, they used a range of 1.25-0.078125%. Please indicate what was the percentage used foreach bacterial culture. What was the initial percentage of DMSO used upon adding the analogs dissolved in 70% DMSO?

 

Line 291: The DMSO concentrations as tested in the growth curve assay for the analogs directly corresponds to the range of concentrations tested in the DMSO control growth curve. For example, 25ug/ml of MCC-1 and MCC-2 contains 1.25% DMSO. This is stated in line 267. 70% DMSO was added to 1mg of MCC-1 and MCC-2, then further diluted down by addition of CAMHB to achieve the range of concentrations from1.57ug/ml to 25 ug/ml

 

12. Sentences 249-250, please rephrase this sentence. It looks like these data were

used before and not the original findings of this paper.

 

Line 304: The sentence is rephrased to reflect the original findings of this paper “We have shown here that EGCG, MCC-1 and MCC-2 restored susceptibility of MRSA USA300 to oxacillin and cefazolin” (Figure 3).

 

13. Designing the intracellular killing experiments is not correct. What do you know

about the growth kinetics of the clinical MSSA strain (HH70)? The findings

shown here could be due to a defective growth phenotype of this strain.

 

HH70 is a strain that has been fully sequenced by our lab (Algorri et al., 2021). The bacteria shows consistent and regular intracellular growth as expected when compared to that of USA300 (see graph below). Furthermore, our intracellular killing assays are all compared to a no treatment control, if HH70 showed defective growth, you would see a decrease in growth/dysfunctional growth within the no treatment group. However, the no treatment group of HH70 follows the expected growth rate.

 

 

14. What were the parameters of interpreting the data of the checkboard synergy

assays?

 

The points where FIC was measured using the checkerboard synergy assay followed the following criteria:

1. The concentration of oxacillin and cefazolin when combined with EGCG or EGCG analogs should be ≤2 which is the CLSI breakpoint defining an aureus strain as “susceptible” to oxacillin and cefazolin. If this could not be achieved while following the two criteria below, the next closest concentration of oxacillin and cefazolin would be used for the FIC value, ie 8ug/ml for oxacillin with MCC-1 and MCC-2.
2. The concentration of EGCG, MCC-1 and MCC-2 should be as close as possible to cell viable concentrations for Kup5 cells (6.25ug/ml is 100% viable for MCC-1 and MCC-2)
3. The point where the lowest FIC value could be achieved.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Confirm publication.

Author Response

Thank you for confirming publication of our manuscript

Reviewer 5 Report

Comments and Suggestions for Authors

The reviewer thanks the authors for addressing the comments mentioned earlier. Based on the two versions submitted till the moment and the in the light of the presented answers from the authors, the following points should be clarified:

Responses to major points:

1.    Proposing that the EGCG analogs can overcome the unfavorable biochemical properties of EGCG is very interesting; however, in the paper there is no a proof of such assumption. Regarding the killing experiments, the question was if the EGCG analogs retarded the growth of bacterial cells, what did the authors expect while doing these experiments? The results are completely expected and they are due to this retarded growth phenotype. This will also open the door for another important question, what is the effect of the analogs on the Kupffer cells? The authors need to perform additional experiments excluding any detrimental effect of their analogs on macrophages or testing whether the analogs could stimulate or activate macrophages directly.

2.    The authors need to describe the genetic background of the used USA300. Was the expression of GFP constitutive or induced? In addition, please provide pictures of your validation results. It is then up to the authors to add them to the main text or supplementary files.

3.    Thanks for clarifying the point of the macrophage infection. If I understood well, the T0 timepoints will represent the initial internalization rate of Kup5 cells. Please indicate in a graph whether, there was a difference in this rate or not upon using your analogs. If you measured the cell viability in all of your experiments, please indicate the viability rates.

Responses to minor points:  

All points were answered

Author Response

Round 2 Response to Reviewer 5:

 

1.Proposing that the EGCG analogs can overcome the unfavorable biochemical properties of EGCG is very interesting; however, in the paper there is no a proof of such assumption. Regarding the killing experiments, the question was if the EGCG analogs retarded the growth of bacterial cells, what did the authors expect while doing these experiments? The results are completely expected, and they are due to this retarded growth phenotype. This will also open the door for another important question, what is the effect of the analogs on the Kupffer cells? The authors need to perform additional experiments excluding any detrimental effect of their analogs on macrophages or testing whether the analogs could stimulate or activate macrophages directly.

 

It is well known that EGCG has poor bioavailability and stability, with the bioavailability of EGCG only reaching 0.1% (Warden et al., 2001) (Chow et al., 2001) (Chow et al., 2003) (Ullman et al., 2003). The main factor behind EGCG’s poor biochemical properties has been shown to lie within its 8 hydroxyl groups which make it highly polar, highly hydrophilic and have a high molecular weight, thus hindering its absorption and penetrate through the lipid bilayer to accumulate intracellularly. Generally, other studies try to avoid removal of EGCG hydroxyl groups which have been attributed to its antioxidant property for anti-cancer and neuroprotective action. However, in the context of antimicrobial action, we hypothesized that such antioxidant properties are not necessary and could even counteract the ability for macrophages to kill intracellular bacteria through reactive oxidant species (ROS) generation. Thus, we altered the chemical structure to remove the hydroxyl groups and modify the weak bonds in-between functional groups. These changes have led to the lead analogs exhibiting improved potency against extracellular and intracellular antimicrobial action as well as attaining the majority of ADME properties within the targeted range of drug-like properties, such as lipophilicity, cell permeability.   

 

While our MIC experiments indicate that the EGCG analogs show greater activity compared to EGCG against extracellular S. aureus, killing experiments on infected macrophages were performed to determine and compare efficacy of EGCG and its analogs against intracellular bacteria. Indeed, even when tested at below the MIC for the analogs, when combined with antistaphylococcal antibiotics, the analogs potentiated the intracellular clearance of bacteria compared to antibiotic treatment alone.  Importantly, we have shown that at the concentration tested, the EGCG analogs do not affect viability of Kupffer cells.

The reviewer raised an important follow up question regarding whether the EGCG analogs could stimulate or activate the macrophages directly. This is an active area of investigation given the promising data shown in this proof-of-concept study demonstrating that modifying the chemical structure of EGCG is a feasible approach to improving its intracellular antimicrobial action.

 

 

 

 

 

2. The authors need to describe the genetic background of the used USA300. Was the expression of GFP constitutive or induced? In addition, please provide pictures of your validation results. It is then up to the authors to add them to the main text or supplementary files.

 

 

Information on this GFP strain of USA300 can be found in a 2016 paper by Surewaard et al., who shared this isolate with us. Note that the GFP expression feature of our USA300 strain was only used for the preliminary experiments when testing the high end of EGCG concentrations. Data from those experiments are not relevant in this study.  At high concentrations, EGCG appeared opaque in green and thus difficult to read visually following conventional MIC testing methodology, however at the concentrations used in this paper, this was not an issue.

 

Line 163: Citation for Surewaard paper in which the GFP strain was obtained from was added.

 

 

3. Thanks for clarifying the point of the macrophage infection. If I understood well, the T0 timepoints will represent the initial internalization rate of Kup5 cells. Please indicate in a graph whether, there was a difference in this rate or not upon using your analogs. If you measured the cell viability in all your experiments, please indicate the viability rates

 

The reviewer is correct, the T0 points represent initial internalization of S. aureus within Kupffer cells after 1.5-hour infection and 2-hour gentamicin wash. The graph below shows the difference and raw Log CFU counts of intracellular S. aureus at T0 between HH70 and USA300. Since this study focuses on the potential clinical application of EGCG analogs as adjunct to antibiotic therapy, we performed only post-infection treatments in our intracellular killing assays, whereby treatments were added after Kupffer cells have already internalized the S. aureus mimicking the clinical scenario. Others have shown data suggesting that EGCG (and thus potentially our EGCG analogs) may increase macrophage phagocytic activity (Monobe et al., 2013) (Huang et al., 2013). This feature will be further tested in additionally studies but is beyond the scope of this paper which is to identify the effects of EGCG chemical structure modification on antimicrobial activity against both extracellular and intracellular S. aureus.