Molecular Integrative Study on Inhibitory Effects of Pentapeptides on Polymerization and Cell Toxicity of Amyloid-β Peptide (1–42)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript proposed by the authors presents the results of a study on the potential of cationic arginine-rich peptides as inhibitors of amyloidogenesis. The authors identify, through a molecular docking procedure that explores the space of possible Arg-rich pentapeptides, two particularly promising peptides, which are then tested to assess their effects on amyloidogenesis. The manuscript is not always easy to read, and the English could be improved. For example, in the abstract, the phrase "energy was reduced due to the formation of three hydrophobic regions" is unclear.

 

For the manuscript to be accepted for publication, several points need to be clarified and improved. Given my expertise, I will focus my review on the computational aspects of the manuscript. Specifically:

1.      The introduction in the Materials and Methods section regarding MD (Molecular Dynamics) and its successful applications in the field of biomolecules is rather generic and could certainly be revised and perhaps shortened.

2.      The aspect that I believe needs significant improvement in this manuscript is the quality of the MD simulations. How can the authors claim that the MD simulation is converged after only 5 ns of simulation? I understand that it requires considerable additional work, but such short MD simulations are insufficient to reliably assess the binding of the peptide to Ab1-42. In Int. J. Mol. Sci. 2023, 24, 7673, the same authors consider t least 50 ns MD simulation. This is a minimum requirement that must be addressed in order for the paper to be considered for publication in this journal.

3.      The authors should include a table with the best docking scores for the selected peptides and the free energies obtained from the umbrella sampling calculations.

4.      The conclusions section is quite generic and should be written more precisely, highlighting the most significant results obtained from both the calculations and the experimental work.

5.      What could be the mechanism behind the reduced cytotoxicity of Ab-42 when interacting with the pentapeptide? Could the binding inhibit fibril formation? Additionally, how could the reduced propensity for ROS formation be explained?

In conclusion, I emphasize that for the paper to be accepted for publication, the MD simulations section needs to be enhanced and made more robust in terms of simulation times.

Comments on the Quality of English Language

English must be improved

Author Response

The manuscript proposed by the authors presents the results of a study on the potential of cationic arginine-rich peptides as inhibitors of amyloidogenesis. The authors identify, through a molecular docking procedure that explores the space of possible Arg-rich pentapeptides, two particularly promising peptides, which are then tested to assess their effects on amyloidogenesis. The manuscript is not always easy to read, and the English could be improved. For example, in the abstract, the phrase "energy was reduced due to the formation of three hydrophobic regions" is unclear.

For the manuscript to be accepted for publication, several points need to be clarified and improved. Given my expertise, I will focus my review on the computational aspects of the manuscript. Specifically:

 

Comment 1. The introduction in the Materials and Methods section regarding MD (Molecular Dynamics) and its successful applications in the field of biomolecules is rather generic and could certainly be revised and perhaps shortened.

 

Response: The ‘Abstract’ and the ‘Materials and Methods’ sections have been revised accordingly. All revision made in the manuscript has been highlighted by yellow color.

 

Comment 2. The aspect that I believe needs significant improvement in this manuscript is the quality of the MD simulations. How can the authors claim that the MD simulation is converged after only 5 ns of simulation? I understand that it requires considerable additional work, but such short MD simulations are insufficient to reliably assess the binding of the peptide to Ab1-42. In Int. J. Mol. Sci. 2023, 24, 7673, the same authors consider t least 50 ns MD simulation. This is a minimum requirement that must be addressed in order for the paper to be considered for publication in this journal.

 

Response: The 20 L-α-amino acids yield 3,200,000 pentapeptides, far beyond the 200 dipeptides. In this investigation on the impact of pentapeptides on Abeta polymerization, we first looked at the RMSD and IE (interaction energy) in 5 ns, followed by the assay of binding free energy which was around 250 ns in every simulation. In a system, the binding free energy is vital because it dictates whether a reaction will occur spontaneously or the direction of a reaction. The IE does not account for the effects of water and hydrophobic interactions, which are essential for molecular interactions. The binding free energy results are depicted in Figure 5.

 

Comments 3. The authors should include a table with the best docking scores for the selected peptides and the free energies obtained from the umbrella sampling calculations.

 

Response: The table below shows the refined docking scores of the pentapeptides, and the table has been added to the supplementary files of the paper.

Ligand	△G*
TRRRR	-49.1802216
ARRGR	-41.3892555
RRRWR	-60.6516304
RRRDS	-46.2803879
TRRAR	-54.5261192

*△G was calculated using the generalized Born volume integral/weighted surface area method.

 

Comments 4. The conclusions section is quite generic and should be written more precisely, highlighting the most significant results obtained from both the calculations and the experimental work.

 

Response: The conclusion section has been revised accordingly on pages 20 to 21, lines 638 to 642, please check.

 

Comments 5: What could be the mechanism behind the reduced cytotoxicity of Ab-42 when interacting with the pentapeptide? Could the binding inhibit fibril formation? Additionally, how could the reduced propensity for ROS formation be explained?

 

Response: The solvent-solute, solute-solute, and solvent-solvent interactions determine molecular interaction, protein folding, and solubility in a solution, e.g. the hydrophobic interactions are especially significant in protein-protein interaction, polymerization, and folding. Our earlier publication identified three hydrophobic clusters in the Aβ42 polymerization core, each with a unique contribution to lowering the free energy of Aβ42 solution. The pentapeptide binds to the hydrophobic cluster of the Aβ42 polymerization core, preventing it from elongating into fibrils. Aβ42 fibrils, like prion protein, can harm neurons and brain tissue. As reported, aggregated Aβ42 not only causes inflammation but also increases ROS generation. Inhibiting Aβ42 aggregation reduces the detrimental effects.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors demonstrated the effect of two pentapeptides, TRRRR and ARRGR, in reducing the Aβ₄₂ aggregation and toxic effect in SH-SY5Y cells. The first part of this study, using many different approaches, the author showed that these two peptides can indeed decrease the aggregation of Aβ. The significance of the work is high, and the data in this part is solid. In the second part, the authors showed the effect of these two pentapeptides in reducing cell toxicity induced by Aβ. However, there are some major flaws in this second part that the authors must solve before this work is published.

 

1. The authors need to justify the time points chose for this experiment. In Figure 6, Aβ aggregated and plateaued out in between 1 hour and 2 hours (~100 min) without any addition of the pentapeptides. Both TRRRR and ARRGR showed inhibitory effect very early. Around 50min, both peptides reached their maximum effect. In Figure 9, however, the timing is not stated clearly to match the authors’ finding in Figure 6. Namely:

a. When did the author add the two peptides? Was it before the cells were transfected or 4 hours after transfection?

b. How long was the treatment of these two pentapeptides in total?

c. How many hours were the cells fixed/stained after transfection?

d. In the methods section, the authors claimed that the cells were transfected 4 hours total. How long did it take for the plasmid to express Aβ? And to fit into the time chart in Figure 6, when is the peptides supposed to take effect in cultured cells?

 

None of these above-mentioned time points were stated clearly in result section 3.5 nor the figure legend of Figure 9. The authors need to specify them and explain the logic behind choosing these timepoints.

 

2. In Figure 8, Result section 3.4, the ELISA experiment showed ‘the effect of TRRRR and ARRGR on Aβ₄₂ expression’. Is this experiment done in vitro? Did the authors treat the cells with the two peptides, or did they treat Aβ₄₂ with the two peptides? When the authors say ‘expression’ do they mean protein expression from the cells through mRNA? Please specify.

To follow up on this point, do SH-SY5Y cells generate Aβ endogenously? With ELISA the light absorption is all normalized to a control, in this study the control is cells not transfected with Aβ plasmid (there is no specification in the figure legend nor Result section 3.4. The whole paragraph is very vague and only stated ‘control group’ or ‘in groups which received doses of…’. The reviewer had to speculate the authors were referring to the cells that were not transfected was the ‘control group’. Please fix this), whereas the real control should be the medium to establish a baseline in which does not have any Aβ expression.

 

3. The authors used secreted Aβ plasmid to mimic the addition of Aβ peptide in vitro. However, there is no reporter shown in this study. The authors need to use either co-transfection of a fluorescent reporter (e.g.: GFP) or other methods to first show that the transfection works. Looking at the dead cells is very indirect. The ELISA data in Figure 8 showed the increase of Aβ after transfection (again, the samples used for the ELISA experiment are very ambiguous in text, see second comment.) but the readers still need to know the transfection rate. 

 

4. The authors argued that transfection of secretive Aβ in SH-SY5Y cells can cause cell death. A critical control is missing here. The authors need to include a control with a plasmid that does not overexpress Aβ to prove the transfection do not kill the cells.

 

5. In result 3.6, the authors showed the protective effect of the two peptides on ROS produced by Aβ, the incubation time of the two peptides were 24 hours. In Figure 6 the authors already showed at approximately 50min (~1 hour) the inhibition of both peptides peaked. In Figure 6, the X-axis also stopped at 300min (5 hours). If the incubation time is extended to 24 hours and beyond, the authors need to extend incubation time in Figure 6 as well. Especially it is necessary to rule out the possibility that this inhibition effect on Aβ aggregation does not go away after 24 hours.

To continue on this point, why does the authors do not examine the effect of the two pentapeptides at a shorter time point?

 

6. The authors observed cell apoptosis after transfection of Aβ plasmid. What is the aggregation stage of Aβ? Different aggregation stage of Aβ have different effects on cells through different pathways. The authors need to at least recognize, discuss, and distinguish their Aβ in soluble, oligomerized or fibrillar form. 

 

 

 

Below are some minor issues:

 

1. Figure 5F is missing whereas the figure legend saying there is an overlay.

 

2. In Line 483-486, the author referred to Figure 6C, D and E, whereas it should be Figure 7. Whereas the manuscript has not referred to the real Figure 6 at all in-text.

 

3. Figure 9 is missing a scale bar.

 

4. The author stated ‘The absence of successful therapeutic targets for AD…’ in Line 836 and ‘…lead to uncovering new therapeutic targets for AD.’ In Line 852. The core argument in this manuscript is to use Aβ (aggregation) as a current therapeutic target, while these statements are saying the field needs new target. Please change the text in discussion and conclusion accordingly.

 

5. In methods section, 2.8, the authors need to give the information of the sequence of the secretive form of Aβ plasmid they used. 

 

6. The authors should give more information about how the transfection was done. It seems like Lipofectamine 2000 was used but not mentioned in 2.8 ‘Transfection and expression of secreted Aβ₄₂’. The only information the authors included in this section is the timing of the transfection. The authors need to include more detailed information, e.g.: what is the amount of DNA transfected in each group? What reagents were used and how much?

7. In the result section 3.5, the authors are mixing the figures. ‘Figure 9 G, H’ is supposed to be Figure 9A, g and h. ‘Figure 9 E and F’ should be Figure 9 A (or B) e and f. This figure has A, B and C panels only.

 

8. The authors have exciting and solid data of these two pentapeptides inhibiting the aggregation of Aβ₄₂. Would the authors consider combining the two peptides to see if it makes the inhibition stronger?

 

9. The introduction needs to provide more information about different stage of Aβ aggregation, the currently known pathways of Aβ toxicity to neurons, i.e.: how does Aβ kill cells?

Author Response

Comment 1. The authors need to justify the time points chose for this experiment. In Figure 6, Aβ aggregated and plateaued out in between 1 hour and 2 hours (~100 min) without any addition of the pentapeptides. Both TRRRR and ARRGR showed inhibitory effect very early. Around 50min, both peptides reached their maximum effect. In Figure 9, however, the timing is not stated clearly to match the authors’ finding in Figure 6. Namely:

1. When did the author add the two peptides? Was it before the cells were transfected or 4 hours after transfection?
2. How long was the treatment of these two pentapeptides in total?
3. How many hours were the cells fixed/stained after transfection?
4. In the methods section, the authors claimed that the cells were transfected 4 hours total. How long did it take for the plasmid to express Aβ? And to fit into the time chart in Figure 6, when is the peptides supposed to take effect in cultured cells?

 

None of these above-mentioned time points were stated clearly in result section 3.5 nor the figure legend of Figure 9. The authors need to specify them and explain the logic behind choosing these timepoints.

 

Response: The two peptides were added 4 h after transfection, and the expression of Aβ was tested 24 h after the addition of the pentapeptides. The cells were fixed at room temperature for 15 min. For detection of the expression of Aβ by ELISA method, the overall expression time was 28 h (4 h transfection + 24 h). The detailed information has been added to Section 2.9 and 2.10 of the Experimental Methods accordingly. In the present manuscript, the information related to fixation and staining has been added to the section 2.5. In the Figure 6, Aβ, the pentapeptides and ThT were mixed first, then ThT fluorescence was detected every 5 min. For the concerns related to the section 3.5 or the Figure 9, the detailed information has been added to the section 2.10 of the Materials and Methods.

 

Comment 2. In Figure 8, Result section 3.4, the ELISA experiment showed ‘the effect of TRRRR and ARRGR on Aβ42 expression’. Is this experiment done in vitro? Did the authors treat the cells with the two peptides, or did they treat Aβ42 with the two peptides? When the authors say ‘expression’ do they mean protein expression from the cells through mRNA? Please specify.

 

To follow up on this point, do SH-SY5Y cells generate Aβ endogenously? With ELISA the light absorption is all normalized to a control, in this study the control is cells not transfected with Aβ plasmid (there is no specification in the figure legend nor Result section 3.4. The whole paragraph is very vague and only stated ‘control group’ or ‘in groups which received doses of…’. The reviewer had to speculate the authors were referring to the cells that were not transfected was the ‘control group’. Please fix this), whereas the real control should be the medium to establish a baseline in which does not have any Aβ expression.

 

Response: First, in the experiment shown in the Figure 8, we tested the effects of the pentapeptides on Aβ protein levels expressed from the cells transfected with the plasmid encoding secretable Aβ. The cells, not Aβ, were treated with the pentapeptides. Second, the SH-SY5Y cells express Aβ endogenously. The third, according to your concern, we have modified the labels of the figures to identify the groups clearly, including Figure 8, 9, 10 and 11.

 

Comment 3. The authors used secreted Aβ plasmid to mimic the addition of Aβ peptide in vitro. However, there is no reporter shown in this study. The authors need to use either co-transfection of a fluorescent reporter (e.g.: GFP) or other methods to first show that the transfection works. Looking at the dead cells is very indirect. The ELISA data in Figure 8 showed the increase of Aβ after transfection (again, the samples used for the ELISA experiment are very ambiguous in text, see second comment.) but the readers still need to know the transfection rate.

 

Response: In the study, we used the pcDNA3.1-Aβ plasmid containing the signal peptide to express secretable Aβ to mimic naturally secreted Aβ. The plasmid has the positive selection gene marker to enable the cells to be resistant to the antibiotic, G418, which was applied in the study. We have relabeled Figure 8 to avoid misunderstanding. In Figure 8, the increase of Aβ levels were resulted from the transfection of the pcDNA3.1-Aβ plasmid.

 

Comment 4. The authors argued that transfection of secretive Aβ in SH-SY5Y cells can cause cell death. A critical control is missing here. The authors need to include a control with a plasmid that does not overexpress Aβ to prove the transfection do not kill the cells.

 

Response: In the old figures, the ‘con’ indicates that the cells were transfected with blank plasmids. We have changed the labels in the Figures 8 to 11 to make it clear. In these figures, ‘Aβ⁻’ represents the cells transfected with pcDNA3.1 blank plasmids.

 

Comment 5. In result 3.6, the authors showed the protective effect of the two peptides on ROS produced by Aβ, the incubation time of the two peptides were 24 hours. In Figure 6 the authors already showed at approximately 50min (~1 hour) the inhibition of both peptides peaked. In Figure 6, the X-axis also stopped at 300min (5 hours). If the incubation time is extended to 24 hours and beyond, the authors need to extend incubation time in Figure 6 as well. Especially it is necessary to rule out the possibility that this inhibition effect on Aβ aggregation does not go away after 24 hours.

 

To continue on this point, why does the authors do not examine the effect of the two pentapeptides at a shorter time point?

 

Response: In the experiment shown in the section 3.6, after the Opti-MEM was replaced with DMEM medium containing 5% FBS and antibiotics (penicillin and streptomycin), the SH-SY5Y cells expressing Aβ42 were incubated with TRRRR or ARRGR at final concentration of 10 μM and 50 μM for 24 h. The information has been added to the section 2.11 of the Materials and Methods section. This experimental procedure ensured both the adequate expression level of Aβ and adequate time for the peptides to exert effects.

 

Comment 6. The authors observed cell apoptosis after transfection of Aβ plasmid. What is the aggregation stage of Aβ? Different aggregation stage of Aβ have different effects on cells through different pathways. The authors need to at least recognize, discuss, and distinguish their Aβ in soluble, oligomerized or fibrillar form.

 

Response: In the study, we have observed that the pentapeptides can inhibit the aggregation of Aβ. Since Aβ protein aggregates rapidly, we do not ascertain that the apoptosis was induced totally by aggregated Aβ, however, Aβ did aggregate within 24 h as shown in the study.

 

Below are some minor issues:

 

1. Figure 5F is missing whereas the figure legend saying there is an overlay.

 

Response: Figure 5F has been removed, and the corresponding figure legend will also be eliminated, as indicated in the manuscript.

 

2. In Line 483-486, the author referred to Figure 6C, D and E, whereas it should be Figure 7. Whereas the manuscript has not referred to the real Figure 6 at all in-text.

 

Response: We have corrected the error.

 

3. Figure 9 is missing a scale bar.

 

Response: The scale bars have been added, as shown in Figure 9A(a)/B(a) of the revised manuscript (line 478).

 

4. The author stated ‘The absence of successful therapeutic targets for AD…’ in Line 836 and ‘…lead to uncovering new therapeutic targets for AD.’ In Line 852. The core argument in this manuscript is to use Aβ (aggregation) as a current therapeutic target, while these statements are saying the field needs new target. Please change the text in discussion and conclusion accordingly.

 

Response: We have removed the paragraph discussing the therapeutic targets from the discussion section.

 

5. In methods section, 2.8, the authors need to give the information of the sequence of the secretive form of Aβ plasmid they used.

 

Response: The nucleotide sequence of Aβ42 including the sequence encoding the signal peptide has been added to the revised manuscript in Methods Section 2.8, as detailed in lines 198 to 202.

 

6. The authors should give more information about how the transfection was done. It seems like Lipofectamine 2000 was used but not mentioned in 2.8 ‘Transfection and expression of secreted Aβ42’. The only information the authors included in this section is the timing of the transfection. The authors need to include more detailed information, e.g.: what is the amount of DNA transfected in each group? What reagents were used and how much?

 

Response: We have added related information to the section 2.8 accordingly on page 5, lines 193 to 207.

 

7. In the result section 3.5, the authors are mixing the figures. ‘Figure 9 G, H’ is supposed to be Figure 9A, g and h. ‘Figure 9 E and F’ should be Figure 9 A (or B) e and f. This figure has A, B and C panels only.

 

Response: Changes have been made (see Results Section 3.5, lines 460 to 465, in the revised manuscript).

 

8. The authors have exciting and solid data of these two pentapeptides inhibiting the aggregation of Aβ42. Would the authors consider combining the two peptides to see if it makes the inhibition stronger?

 

Response: Thank you for your suggestions. A compound drug containing both pentapeptides may merit to the dose of each peptide, but it may not change the maximum effects.

 

9. The introduction needs to provide more information about different stage of Aβ aggregation, the currently known pathways of Aβ toxicity to neurons, i.e.: how does Aβ kill cells?

 

Response: We have added information to the introduction section of the manuscript on page 2, lines 53 to 64. The present study is part of our serial studies focusing on the aggregation of Aβ42 and screening of its inhibitors, and the main purpose of the study was to screen pentapeptides that inhibit the aggregation of Aβ42. Further studies on different aggregated forms of Aβ42 would be done in future studies.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

• The table proposed is fine but the units of measure are not mentioned. Moreover 7 digits is too much, reduce to 2.

The paper can be published.

 

Comments on the Quality of English Language

English must be improved

Author Response

Comment 1. The table proposed is fine but the units of measure are not mentioned.  Moreover 7 digits is too much, reduce to 2.

Response 1: Thank you for your careful review. We have added the unit, kcal·mol^-1 to the supplementary table, and reduced the digits to 2.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors fixed most of the critical information that were previously completely missing. However, with the supplemented detail, it raises a couple of concerns that the authors need to address.

 

1. This is the authors’ response: The two peptides were added 4 h after transfection, and the expression of Aβ was tested 24 h after the addition of the pentapeptide. The authors have data to support that 4 h after transfection, secreted Aβ level is high in culture. In Figure 6 the authors also showed both peptides inhibit the polymerization of Aβ starting from ~50 or 60min, and this effect persist afterwards. In this case, what is the reason the peptide stayed in the cell culture medium 24 hours after Aβ starts to express? I assume the authors want to see the long-term effect on the cells. If this is true, in Figure 6 the Aβ treatment should have at least one more time point to show if the effect persist after 24 hours, instead of stopping at 300min (6hr).

 

2. Regarding the authors’ reply: Second, the SH-SY5Y cells express Aβ endogenously. Do the cells directly express β-amyloid peptide endogenously? Or do they express amyloid precursor protein? For most neuroblastoma cells, they express APP and undergo cleavages by secretases to generate Aβ. Whereas the authors later used a plasmid to directly overexpress β-amyloid peptide (a secretive form). The authors have to make very clear distinction between these two types of Aβ in the manuscript.

 

3. The authors provided information regarding the Aβ construct they used in this revised version of their manuscript. G418 concentration and source is still missing and needs to be provided in the main manuscript. This type of information was largely missing in the first draft. The authors should be obligated and more attentive to provide them so that their results could be potentially repeated by other groups.

Author Response

Comment 1. This is the authors’ response: The two peptides were added 4 h after transfection, and the expression of Aβ was tested 24 h after the addition of the pentapeptide. The authors have data to support that 4 h after transfection, secreted Aβ level is high in culture. In Figure 6 the authors also showed both peptides inhibit the polymerization of Aβ starting from ~50 or 60 min, and this effect persist afterwards. In this case, what is the reason the peptide stayed in the cell culture medium 24 hours after Aβ starts to express? I assume the authors want to see the long-term effect on the cells. If this is true, in Figure 6 the Aβ treatment should have at least one more time point to show if the effect persist after 24 hours, instead of stopping at 300 min (6hr).

Response 1: In section 2.8 of the Materials and Methods, “4 hours” refers to the duration of the transfection procedure. In section 2.9, “4 hours after transfection” refers to the time point when the pentapeptides were added, but not to say that 4 h after transfection, secreted Aβ level is high in culture.

In Results 3.2 (Figure 6), the Aβ42 used is commercial Aβ42 monomer protein (see Materials and Methods 2.6), which demonstrated that both peptides started to inhibit Aβ aggregation after approximately 50 or 60 minutes of incubation, with the inhibition effect persisting for up to 300 minutes (5 hours). Results 3.4 (Figure 8), 3.5 (Figure 9), 3.6 (Figure 10), and 3.7 (Figure 11) involve cell experiments (secreted Aβ42, see Materials and Methods 2.8) and were conducted 28 hours after transfection (as visible under the microscope, the pentapeptides reduced cell toxicity by inhibiting Aβ42 aggregation, as shown in Figure 9A). The different Aβ42 proteins used in the experiments and the treatment time points are not related.

 

Comment 2. Regarding the authors’ reply: Second, the SH-SY5Y cells express Aβ endogenously. Do the cells directly express β-amyloid peptide endogenously? Or do they express amyloid precursor protein? For most neuroblastoma cells, they express APP and undergo cleavages by secretases to generate Aβ. Whereas the authors later used a plasmid to directly overexpress β-amyloid peptide (a secretive form). The authors have to make very clear distinction between these two types of Aβ in the manuscript.

Response 2: In the study, SH-SY5Y cells endogenously express Aβ42 itself, not APP. The cDNA sequence encoding Aβ42 with the signaling peptide at its N terminus has been added to the section 2.8. The Aβ42 levels in the Aβ42 group (SH-SY5Y cells secreting Aβ42) were significantly higher than those in the control group. Additionally, Figure 8 and its legend have been updated to include the relevant information.

Comment 3. The authors provided information regarding the Aβ construct they used in this revised version of their manuscript. G418 concentration and source is still missing and needs to be provided in the main manuscript. This type of information was largely missing in the first draft. The authors should be obligated and more attentive to provide them so that their results could be potentially repeated by other groups.

Response 3: Changes have been made accordingly, please refer to lines 198 to 199 of the manuscript.