Discovery of the Potentiator of the Pore-Forming Ability of Lantibiotic Nisin: Perspectives for Anticancer Therapy

Round 1

Reviewer 1 Report

The manuscript is very well written, and the experiments and conclusions are well presented. Only as a minor comment, in Figs.2-3 I'd clarify how the macroscopic current was estimated; was it the actual average of the fluctuating current ? Please specify.

Also, authors state that: 'the compounds diminishing the membrane boundary or dipole potential (in particular, phloretin and capsaicin) might be considered as agonists of anticancer action of nisin.'. I'd go down a note an this matter, as it isn't actually clear how such dipole potential modifying compounds would be delivered to a patient, or what are the side affects on other, healthy cells. At most, more in vivo experiments with such formulation are required before going to the conclusion suggested by authors.

Author Response

Reviewer 1 comment #1.

The manuscript is very well written, and the experiments and conclusions are well presented. Only as a minor comment, in Figs.2-3 I'd clarify how the macroscopic current was estimated; was it the actual average of the fluctuating current? Please specify.

Answer #1. We highly appreciate the Reviewer for the careful reading. Figure 2 and 3 represents the typical experiments made. The experiments were done in triplicates to tenplicates (3÷10 independent experiments), and the average value of the ratio between steady-state currents before and after small molecule addition was presented as mean ± s.d. A criterion for reaching a steady-state current level was dI/dt≤0. According to the Reviewer's comment, the Materials and Methods section has been supplemented (page 3, lines 129-130).

 

Comment #2. Also, authors state that: 'the compounds diminishing the membrane boundary or dipole potential (in particular, phloretin and capsaicin) might be considered as agonists of anticancer action of nisin.'. I'd go down a note an this matter, as it isn't actually clear how such dipole potential modifying compounds would be delivered to a patient, or what are the side affects on other, healthy cells.

Answer #2. Phloretin has been described as an antitumor agent, which can act using different mechanisms. For instance, it can serve as a glucose transporter (GLUT2) inhibitor (Wu et al., Int J Cancer. 2009 doi: 10.1002/ijc.24189), which can be linked to the apoptosis, including mitochondrial death pathway (Moley & Mueckler, Apoptosis. 2000 doi: 10.1023/a:1009697908332). Moreover, phloretin can induce apoptosis through suppression the expression of Bcl-2 and increasing the protein expression of cleaved caspase-3 and -9, which was shown on non-small cell lung cancer cell lines (Ma et al., Int J Oncol. 2016 doi: 10.3892/ijo.2015.3304).

The possible drug formulations of phloretin are extensively developed. Recently the anti-inflammatory potential of phloretin microemulsion for treatment of vaginitis has been shown (Abu-Azzam and Nasr, Pharm Dev Technol. 2020 doi: 10.1080/10837450.2020.1764032). To enhance the oral bioavailability and bioefficacy of phloretin mixed polymeric modified self-nanoemulsions has been explored (Wang et al., Food Sci Nutr. 2020 doi: 10.1002/fsn3.1637). The study by (Ranjanamala et al., Biointer Res Appl Chem. 2022, doi: 10.33263/BRIAC123.30763089) elucidated that encapsulated PLGA phloretin nanoparticles demonstrate high antioxidant and antimicrobial activity.

 

Comment #3. At most, more in vivo experiments with such formulation are required before going to the conclusion suggested by authors.

Answer #3. According to the Reviewer's suggestion, we have performed additional flow cytometry experiments to examine the assumption that phloretin can be considered as potential facilitator of anticancer action of nisin via depolarization of mitochondrial membrane. Using MitoTracker we have showed that combination of nisin and phloretin has been able to dramatically change the mitochondrial membrane potential. The appropriate paragraph has been included into the Materials and Methods and the Results and Discussion section (page 3, lines 106-113; page 4, lines 174-190; page 9-10, lines 372-380), and the Supplementary Materials have been added. 

Reviewer 2 Report

This work addresses important problem of pore formation by nisin, which is crucial for its not only antibacterial, but also anticancer action. Although, authors seem to be completely out of the structural context of this peptide action and not considering all the immense body of data on the antimicrobial action of nisin (which is primary, specific, obvious). In my opinion, this work should be completely rethought, redesigned and resubmitted.

1.      Authors seem to have no idea about primary (antibacterial) action of nisin and its structural features. At the same time, this is a priority issue, especially in the light of how many data has been obtained for it in recent 20 years. Authors beyond any doubt must start this journey from studying the structure of nisin complex with lipid II (Hsu et al., NSMB 2004) and other structural aspects, reviewed, e.g., in Panina et al., 2019 (10.1134/S1068162019010126). Moreover, most references in this work are >15 y.o., whassup? Authors stopped reading scientific literature years ago? There are very well elaborated concepts of how nisin acts on bacteria, including membrane recognition, pore formation, stoichiometry, etc.; and in that comparison authors’ attempts to say something about “mechanisms” of anticancer action are miserable.

2.      In connection with the previous, what is real evidence that nisin has anticancer potential? Too many compounds are referred to as “anticancer”, but just a handful really are. Authors do not provide concrete confirmation of these fact. Can nisin be of real use if the acting concentration is millimoles? Authors postulate that nisin is an excellent pore-former for membranes without lipid II, referring to the paper which describes lipid II – mediated action, which also seems misleading and maybe even wrong.

3.      The investigation aims and choice of the objects are confusing. Why measure nisin action on mere anionic bilayers? Authors state that nisin forms pores “predominantly” in membranes of such composition. No! Nisin “predominantly” forms pores in lipid II – containing bilayers, and 1000-fold more intensively as compared to “common” lipids. Action on the anionic lipids is feature of almost all AMPs, and nisin is not the special case — I would say, conversely, because its all specialness is due to its ability to specifically capture lipid II in bacterial membranes.

4.      I have little idea why talk so much about cardiolipin (CL) and mitochondrial membranes. Is CL an affine ligand for nisin? I suppose not. CL is mitochondrial, nisin is extracellular and will not cross the barrier of the healthy eucaryotic cell (unlike bacterial, which is disturbed by AMP and specifically nisin) — why authors see any link with its apoptotic action? Anyway, should nisin act at mitochondria of the cancer cells and just leave normal cells alone? I don’t think so. By the way, even the abbreviation CL can hardly be deciphered throughout the text.

5.      The “agonist” word in the title and throughout the work is misleading. “Agonist” assumes affine and selective action alike ligand—receptor, which is absolutely not the case of nisin and capsaicin. In this case, word “facilitator” may suite better. Next, what’s the idea behind pouring capsaicin — real agonist of TRPV1 receptors — on the membrane and measuring nisin action at the same time? This idea seems ridiculous and incompatible with real medical use, or at least not explained.

6.      “Boundary” potential is misleading. This term comes from physical chemistry, but for molecular biology more important is transmembrane (TM) potential, and the reader will probably mix them up, assuming “compounds that diminish boundary potential” as ionophores that reset TM-potential to zero and therefore are toxic to any cell types.

7.      Experiments show that application of capsaicin decreased the “boundary” — or transmembrane? — potential by 100 mV. Which was the initial potential? Native TM-potential is −70 mV, what is the meaning of the applied potential in this case? Why use 400 μM of capsaicin — a really “hot” solution?

8.      From the primary experiment (Fig. 1) we may see that nisin forms pores when cardiolipin is involved (or when bi-anionic membrane is treated). But why? Explanation of this fact from structural perspective would be interesting, but authors do not seem to have tried to do this.

9.      English is very poor, and many signs of Russian appear (all thar «ы»’s and «зщку»).

Author Response

Reviewer 2 general comment.

This work addresses important problem of pore formation by nisin, which is crucial for its not only antibacterial, but also anticancer action. Although, authors seem to be completely out of the structural context of this peptide action and not considering all the immense body of data on the antimicrobial action of nisin (which is primary, specific, obvious). In my opinion, this work should be completely rethought, redesigned and resubmitted.

1. Authors seem to have no idea about primary (antibacterial) action of nisin and its structural features. At the same time, this is a priority issue, especially in the light of how many data has been obtained for it in recent 20 years. Authors beyond any doubt must start this journey from studying the structure of nisin complex with lipid II (Hsu et al., NSMB 2004) and other structural aspects, reviewed, e.g., in Panina et al., 2019 (10.1134/S1068162019010126). Moreover, most references in this work are >15 y.o., whassup? Authors stopped reading scientific literature years ago? There are very well elaborated concepts of how nisin acts on bacteria, including membrane recognition, pore formation, stoichiometry, etc.; and in that comparison authors’ attempts to say something about “mechanisms” of anticancer action are miserable.
2. In connection with the previous, what is real evidence that nisin has anticancer potential? Too many compounds are referred to as “anticancer”, but just a handful really are. Authors do not provide concrete confirmation of these fact. Can nisin be of real use if the acting concentration is millimoles? Authors postulate that nisin is an excellent pore-former for membranes without lipid II, referring to the paper which describes lipid II – mediated action, which also seems misleading and maybe even wrong.
3. The investigation aims and choice of the objects are confusing. Why measure nisin action on mere anionic bilayers? Authors state that nisin forms pores “predominantly” in membranes of such composition. No! Nisin “predominantly” forms pores in lipid II – containing bilayers, and 1000-fold more intensively as compared to “common” lipids. Action on the anionic lipids is feature of almost all AMPs, and nisin is not the special case — I would say, conversely, because its all specialness is due to its ability to specifically capture lipid II in bacterial membranes.
4.  I have little idea why talk so much about cardiolipin (CL) and mitochondrial membranes. Is CL an affine ligand for nisin? I suppose not. CL is mitochondrial, nisin is extracellular and will not cross the barrier of the healthy eucaryotic cell (unlike bacterial, which is disturbed by AMP and specifically nisin) — why authors see any link with its apoptotic action? Anyway, should nisin act at mitochondria of the cancer cells and just leave normal cells alone? I don’t think so. By the way, even the abbreviation CL can hardly be deciphered throughout the text.
5.  The “agonist” word in the title and throughout the work is misleading. “Agonist” assumes affine and selective action alike ligand—receptor, which is absolutely not the case of nisin and capsaicin. In this case, word “facilitator” may suite better. Next, what’s the idea behind pouring capsaicin — real agonist of TRPV1 receptors — on the membrane and measuring nisin action at the same time? This idea seems ridiculous and incompatible with real medical use, or at least not explained.
6.  “Boundary” potential is misleading. This term comes from physical chemistry, but for molecular biology more important is transmembrane (TM) potential, and the reader will probably mix them up, assuming “compounds that diminish boundary potential” as ionophores that reset TM-potential to zero and therefore are toxic to any cell types.
7. Experiments show that application of capsaicin decreased the “boundary” — or transmembrane? — potential by 100 mV. Which was the initial potential? Native TM-potential is −70 mV, what is the meaning of the applied potential in this case? Why use 400 μM of capsaicin — a really “hot” solution?
8. From the primary experiment (Fig. 1) we may see that nisin forms pores when cardiolipin is involved (or when bi-anionic membrane is treated). But why? Explanation of this fact from structural perspective would be interesting, but authors do not seem to have tried to do this.
9. English is very poor, and many signs of Russian appear (all thar «ы»’s and «зщку»).

 

General answer. We are disappointed and extremely surprised by the non-academic style of Reviewer’s comments. However, we are very pleased that our work caused such a strong emotional response.

We absolutely disagree with the Reviewer's comments.

Answers to comments 1 and 3. The aim of our study was to investigate the molecular mechanisms of action of nisin on the phospholipid membranes without the lipid II and possible consequences for its anticancer action. Specific target for antimicrobial action of nisin is discussed in the Introduction, while its anticancer activity is not related to specific interaction with lipid II, as lipid II is a specific adjuvant of Gram-positive bacteria, and it is not found in mammalian cell membranes. The reference list includes the data until 2022.

Answer to comment 2. The current literature evidences to a shift in focus from the antimicrobial to anticancer activity of nisin. The high anticancer potential of nisin is widely discussed (Ahmadi et al., 2017 Microb Pathog. doi: 10.1016/j.micpath.2017.08.037; Lewies et al., Biochimie. 2018 doi: 10.1016/j.biochi.2017.10.009; Goudarzi et al., AAPS PharmSciTech. 2018 doi: 10.1208/s12249-018-0969-4; Zainodini et al., Asian Pac J Cancer Prev. 2018 doi: 10.22034/APJCP.2018.19.8.2217; Prince et al., Phys Chem Chem Phys. 2019 doi: 10.1039/c8cp06378h; El-Sayed Ibrahim et al., Rep Biochem Mol Biol. 2021 doi: 10.52547/rbmb.9.4.452; Sadri et al., Biochem Cell Biol. 2022 doi: 10.1139/bcb-2021-0225; Sadri et al., Biochem Cell Biol. 2022 doi: 10.1139/bcb-2021-0225; Sadri et al., Biochem Cell Biol. 2022 doi: 10.1139/bcb-2021-0225; Radaic et al., J Biomed Nanotechnol. 2022 doi: 10.1166/jbn.2022.3314; Balcik-Ercin and Sever, Chem Biol Interact. 2022 doi: 10.1016/j.cbi.2022.110152).

Answer to comment 4. In the plasma membranes of eukaryotic cells coincidental surface exposure of cardiolipin and phosphatidylserine is not unprecedented. The observed decrease in the threshold concentration of nisin in phosphatidylserine-containing bilayers compared to membrane composed of phosphatidylcholine might explain the antitumor action of nisin by lantibiotic-induced increase in the permeability of the plasma membrane of malignant cells at phosphatidylserine externalization. This may explain the entry of nisin into malignant cells and its «specific» interaction with their mitochondria.

Answer to comment 5. We have performed additional flow cytometry experiments to examine the assumption that phloretin can be considered as potential facilitator of anticancer action of nisin via depolarization of mitochondrial membrane. Using MitoTracker we have showed that combination of nisin and phloretin has been able to dramatically change the mitochondrial membrane potential. The appropriate paragraph has been included into the Materials and Methods and the Results and Discussion section (page 3, lines 106-113; page 4, lines 174-190; page 9-10, lines 372-380), and the Supplementary Materials have been added.

Capsaicin was used only to examine the effects of dipole potential alteration on the pore-forming activity of nisin.

Answer to comments 6 and 7. The Reviewer confuses the different terms (transmembrane and boundary potential). For details please see (Demchenko and Yesylevskyy, Chem Phys Lipids. 2009 doi: 10.1016/j.chemphyslip.2009.05.002 Cohen and Venkatachalam, Annu Rev Biophys. 2014 doi: 10.1146/annurev-biophys-051013-022717; Clarke, Adv Exp Med Biol. 2019 doi: 10.1007/978-3-030-04278-3_6 Galassi and Wilke, Membranes (Basel). 2021 doi: 10.3390/membranes11070478; Sarkar and Chattopadhyay, J Phys Chem B. 2022 doi: 10.1021/acs.jpcb.2c02476).

Capsaicin concentration was chosen according our previous study of its action on the properties of model lipid membranes (Efimova et al. Front Cell Dev Biol. 2020 doi: 10.3389/fcell.2020.00537).

Answer to comment 8. The structural determinants of the interaction between nisin and cardiolipin are of a key importance, but it is a separate study that remains to be done.

Answer to comment 9. We appreciate the Reviewer for the careful review of the text; some language edits have been made; the unfortunate mistakes have been corrected.

Reviewer 3 Report

  Nisin is cationic overall positively charged at pH below its isoelectric point due to the presence of positively charged residues and the absence of negatively charged equivalents. Thus nisin can preferentially binds to negatively-charged cell and model membranes and well  known by its ability to interact with biological and model membranes both electrostatically and via a lipid receptor (lipid II) -mediated ways. Accordingly nanomolar nisin concentrations were sufficient to form transient pores, while micromolar concentrations were necessary in absence of lipid II but in presence of negatively charged phospholipids. However nisin structure simultaneously possesses amphipathic properties.  In accordance with these properties nisin has at least a dual or even multiple mode of actions: apart from the detergent-like effect of nisin on plasma and model membranes lantibiotic nisin uses a lipid II or negatively charged phospholipids as docking molecules to forms defined but transient and nonselective, multi-state pores in  in vivo and in vitro (including planar lipid bilayers  utilized by authors of submitted MS). This approach allowed authors to discriminate either pore forming ability of nisin alone (Fig. 1) or combined with small molecules potentially acting as adjuvants (Table 2) in BLM consisting of single lipid constituent (zwitterionic (PE), net neutral (PC), monoanionic (PG and PS) and dianionic (CL) or used in combination with either PC or PE as bulk lipids.

 

 

In the accordance with calcein- or carboxyfluorescein-based leakage assays anionic lipids promote the nisin activity independently of the headgroup type, since DOPS , TOCL as DOPG were equivalent in promoting of the nisin activity.  However authors  of submitted MS

demonstrated that of the pore-forming ability of nisin depends on the type of negatively charged lipid. Dianionic CL was most advantageous, monoanionic DOPG was much less effective in the presence of neutral DOPC and zwitterionic DOPE which were not effective by themselves (Fig. 1 and Table 1). Surprisingly maximal threshold detergent-like concentration of nisin required to disrupt BLM consisting of binary mixture of DOPS and DOPC was dropped 3 times compared to DOPC alone (Table 1). This mode of action suggests an increase in intrinsic permeability of black bilayer to this lantibiotic. Authors extended this finding to hypothesize that non-apoptotic surface exposure of PS in cancer cells can contribute to this phenomenon in vivo.

Authors demonstrated that phloretin and capsaicin were able to potentiate the membrane activity of nisin (Table 2). The synergistic effect of these compounds is consistent with a decrease in the membrane boundary potential (Fig. 3)

 

 

This is a technically sound  and thoroughly executed experimentally study.

The manuscript is appropriately concise.

 However I have several concerns which should be addressed:

 

 

 

1.     Indeed several previous studies led to suggestion that nisin plays its role in apoptosis by increasing mitochondrial intrinsic pathway of apoptosis. However the ability of nisin to either up regulate pro-apoptotic molecules or down regulate anti-apoptotic molecules never been demonstrated.  I would like to suggest another mechanism which can be consistent with action of nisin on plasma membrane as primary target.

 

In the plasma membranes of eukaryotic cells coincidental surface exposure of CL and PS is not unprecedented. CL could also be missorted from mitochondria to the cell surface on the early stages of apoptosis which acts as an in vivo trigger for the production of anti CL antibodies In this case “penetration of the lantibiotic across the barrier to meet its intracellular targets” as suggested by authors is not necessary.

 

 M. Sorice, A. Circella, R. Misasi, V. Pittoni, T. Garofalo, A. Cirelli, A. Pavan, G. Pontieri, G. Valesini, Cardiolipin on the surface of apoptotic cells as a possible trigger for antiphospholipid antibodies. Clinical & Experimental Immunology 122, 277-284 (2000).

M. Sorice, A. Circella, I. Cristea, T. Garofalo, L. Di Renzo, C. Alessandri, G. Valesini, M. Degli Esposti, Cardiolipin and its metabolites move from mitochondria to other cellular membranes during death receptor-mediated apoptosis. Cell Death & Differentiation 11, 1133-1145 (2004).

 

2.     Through the body of submitted manuscript: “dioleylphosphocholine and phosphoserine”, “phosphoserine externalization” “Phosphoserine is asymmetrically distributed”. “However, in cancer cells, phosphoserine” etc.  instead of phosphoserine should be phosphatidylserine.

 

3.     “Phosphoethanolamine is also found in bacterial membranes and its externalization 255 from the inner leaflet of plasma membrane is associated to the malignant transformation 256 of cells [50]. Instead Ref # 50 it is more appropriate to cite

 

Jason H Stafford , Philip E Thorpe

Increased exposure of phosphatidylethanolamine on the surface of tumor vascular endothelium Neoplasia, 2011 Apr;13(4):299-308.

 Broughton LJ, Crow C, Maraveyas A, Madden LA. Duramycin induced calcium release in cancer cells. Anticancer Drugs. 2016 Mar;27(3):173-82.

Author Response

Reviewer 3 general comment #1.

Nisin is cationic overall positively charged at pH below its isoelectric point due to the presence of positively charged residues and the absence of negatively charged equivalents. Thus nisin can preferentially binds to negatively-charged cell and model membranes and well  known by its ability to interact with biological and model membranes both electrostatically and via a lipid receptor (lipid II) -mediated ways. Accordingly nanomolar nisin concentrations were sufficient to form transient pores, while micromolar concentrations were necessary in absence of lipid II but in presence of negatively charged phospholipids. However nisin structure simultaneously possesses amphipathic properties.  In accordance with these properties nisin has at least a dual or even multiple mode of actions: apart from the detergent-like effect of nisin on plasma and model membranes lantibiotic nisin uses a lipid II or negatively charged phospholipids as docking molecules to forms defined but transient and nonselective, multi-state pores in  in vivo and in vitro (including planar lipid bilayers  utilized by authors of submitted MS). This approach allowed authors to discriminate either pore forming ability of nisin alone (Fig. 1) or combined with small molecules potentially acting as adjuvants (Table 2) in BLM consisting of single lipid constituent (zwitterionic (PE), net neutral (PC), monoanionic (PG and PS) and dianionic (CL) or used in combination with either PC or PE as bulk lipids.

 

In the accordance with calcein- or carboxyfluorescein-based leakage assays anionic lipids promote the nisin activity independently of the headgroup type, since DOPS , TOCL as DOPG were equivalent in promoting of the nisin activity.  However authors  of submitted MS

demonstrated that of the pore-forming ability of nisin depends on the type of negatively charged lipid. Dianionic CL was most advantageous, monoanionic DOPG was much less effective in the presence of neutral DOPC and zwitterionic DOPE which were not effective by themselves (Fig. 1 and Table 1). Surprisingly maximal threshold detergent-like concentration of nisin required to disrupt BLM consisting of binary mixture of DOPS and DOPC was dropped 3 times compared to DOPC alone (Table 1). This mode of action suggests an increase in intrinsic permeability of black bilayer to this lantibiotic. Authors extended this finding to hypothesize that non-apoptotic surface exposure of PS in cancer cells can contribute to this phenomenon in vivo.

Authors demonstrated that phloretin and capsaicin were able to potentiate the membrane activity of nisin (Table 2). The synergistic effect of these compounds is consistent with a decrease in the membrane boundary potential (Fig. 3)

This is a technically sound and thoroughly executed experimentally study.

The manuscript is appropriately concise.

 

Answer #1. We highly appreciate the Reviewer for the careful reading and analysis of the manuscript.

 

Comment #2. Indeed several previous studies led to suggestion that nisin plays its role in apoptosis by increasing mitochondrial intrinsic pathway of apoptosis. However the ability of nisin to either up regulate pro-apoptotic molecules or down regulate anti-apoptotic molecules never been demonstrated.  I would like to suggest another mechanism which can be consistent with action of nisin on plasma membrane as primary target.

In the plasma membranes of eukaryotic cells coincidental surface exposure of CL and PS is not unprecedented. CL could also be missorted from mitochondria to the cell surface on the early stages of apoptosis which acts as an in vivo trigger for the production of anti CL antibodies In this case “penetration of the lantibiotic across the barrier to meet its intracellular targets” as suggested by authors is not necessary.

1. Sorice, A. Circella, R. Misasi, V. Pittoni, T. Garofalo, A. Cirelli, A. Pavan, G. Pontieri, G. Valesini, Cardiolipin on the surface of apoptotic cells as a possible trigger for antiphospholipid antibodies. Clinical & Experimental Immunology 122, 277-284 (2000).
2. Sorice, A. Circella, I. Cristea, T. Garofalo, L. Di Renzo, C. Alessandri, G. Valesini, M. Degli Esposti, Cardiolipin and its metabolites move from mitochondria to other cellular membranes during death receptor-mediated apoptosis. Cell Death & Differentiation 11, 1133-1145 (2004).

Answer #2. We are very grateful to the Reviewer for the informal attitude to the manuscript, the desire to find other possible explanations of the results obtained and the suggestion of one more possible mechanism of nisin action. According to the Reviewer's suggestion, we have added the appropriate paragraph in the Discussion section (page 7, lines 280-283), the reference list has been updated accordingly (refs 51, 52).

 

Comment #3. Through the body of submitted manuscript: “dioleylphosphocholine and phosphoserine”, “phosphoserine externalization” “Phosphoserine is asymmetrically distributed”. “However, in cancer cells, phosphoserine” etc.  instead of phosphoserine should be phosphatidylserine.

Answer #3. We highly appreciate the Reviewer for the careful reading; the inaccuracies have been corrected throughout the text.

 

Comment #4. Phosphoethanolamine is also found in bacterial membranes and its externalization 255 from the inner leaflet of plasma membrane is associated to the malignant transformation 256 of cells [50]. Instead Ref # 50 it is more appropriate to cite

Jason H Stafford , Philip E Thorpe Increased exposure of phosphatidylethanolamine on the surface of tumor vascular endothelium Neoplasia, 2011 Apr;13(4):299-308.

Broughton LJ, Crow C, Maraveyas A, Madden LA. Duramycin induced calcium release in cancer cells. Anticancer Drugs. 2016 Mar;27(3):173-82.

Answer #4. The reference list has been updated accordingly Reviewer's suggestion (refs 53, 54).

Round 2

Reviewer 2 Report

While authors state disagreement with my comments, I completely disagree with the results delivery. All of my remarks have left "as is". Even the blatantly misleading term "agonist" is unchanged in the paper title. Check the definition, btw: https://en.wikipedia.org/wiki/Agonist. And there's still no structural perspective/discussion of the results.

Author Response

Reviewer 2 general comment.

While authors state disagreement with my comments, I completely disagree with the results delivery. All of my remarks have left "as is". Even the blatantly misleading term "agonist" is unchanged in the paper title. Check the definition, btw: https://en.wikipedia.org/wiki/Agonist. And there's still no structural perspective/discussion of the results.

 

Answer to Reviewer 2 comment.

We have addressed major Reviewer 2 comments in the first round (please see the previous answers).

According to the Reviewer 2 comment on terminological inaccuracy, we have revised the title of the manuscript.

According to the Reviewer 2 remarks about the absence of structural discussion of the results, we have supplemented the Results and Discussion section with the discussion of possible structural basis for nisin specificity toward phosphatidylglycerol-and cardiolipin-enriched bilayers, the reference list has been also updated (page 1, lines 2-3; page 4, lines 175-196; page 5, lines 242-245; page 8, lines 312-328; page 9, lines 361-379; page 13, lines 564-566; page 14, lines 608-623).