Dimensional Changes in Lipid Rafts from Human Brain Cortex Associated to Development of Alzheimer’s Disease. Predictions from an Agent-Based Mathematical Model

Round 1

Reviewer 1 Report

The paper is well written and offers a new method of assessing the risk of developing neurodegenerative diseases. 
Further analysis is needed to obtain an adequate sample size.

Author Response

Point 1. The paper is well written and offers a new method of assessing the risk of developing neurodegenerative diseases. 
Further analysis is needed to obtain an adequate sample size.

Answer: Thanks you very much for your comments. We have included the sample size of the experimental data used to validate the mathematical model in the section 4.1:

“Sample sizes for control, AD I/II, AD III/IV and AD V/VI groups were 19, 12, 15 and 10, respectively”.

Reviewer 2 Report

In the manuscript is presented a computational approach to identify the dimensional changes in lipid rafts from human brain cortex 2
associated to development of Alzheimer’s disease. Undoubtedly, the manuscript presents some interesting hints. However, some revisions and clarification are needed to improve the quality of the manuscript.

-Authors presented a coarse-grained MD. A paragraph with the limitation of the approach should be introduced since, although coarse-grained MD provide the possibility to simulate large systems it suffers from several drawbacks. A clear discussion about considering coarse-grained instead of all atoms simulation should be reported (considering also to report the number of atoms of the systems considering all atoms and coarse-grained).

-Similar approaches of coarse-grained MD simulation should be reported in the introduction

-no mention about the time of simulation is reported, as well as information about the stability of the systems since no RMSD is calculated for the trajectories. Information about the water model (if considered) should be reported as well as the parameters adopted for the simulation and the software package used.

-a conclusion section should be introduced in which the main findings should be summarized and in which the authors should discuss the utility of the approach and how can be reflected in Alzheimer's diagnosis/therapies

-I highly recommend to revise the language since there are several errors through the text including the abstract

After performing these revisions the paper can be reevaluated

Author Response

Point 1. Authors presented a coarse-grained MD. A paragraph with the limitation of the approach should be introduced since, although coarse-grained MD provide the possibility to simulate large systems it suffers from several drawbacks. A clear discussion about considering coarse-grained instead of all atoms simulation should be reported (considering also to report the number of atoms of the systems considering all atoms and coarse-grained).

Answer: We agree with the reviewer on the importance of remarking the limitations of our methodology. We have added a dedicated paragraph at the end of the discussion section:

“We are aware of the limitations coming from the coarse-grained nature of our model. However, coarse-grain molecular simulation using agent-based approaches are extremely useful as they allow simulations of large molecular systems by simplifying the molecular detail of the elements in the model. In our case we deal with about 40000 different molec-ular species distributed within the different lipid classes, with more than 2000000 atoms and their corresponding degrees of freedom. Even considering this enormous complexity, the results of the model presented here exhibit a significant degree of reliability as they closely resemble the experimental data previously published in human frontal cortex [14,15]”.

Point 2. Similar approaches of coarse-grained MD simulation should be reported in the introduction.

Answer: Considering the suggestion of the reviewer, we have included a collection of articles following a coarse-graining strategy to make molecular dynamics simulations of cell membranes:

“Agent-based models represent a strategy to reduce the scale of molecular dynamics simulations through coarse-graining of the structure of lipid molecules. Similar approaches using this strategy with different scaling levels have been reported. For example, a coarse-grained simulation of the lipid membrane has been used to study the spontaneous separation of elements in a mix of cholesterol, saturates and unsaturated fatty acids [26]. Other studies have been performed to create molecular dynamics simulations of cell membranes reducing the atomic scale by coarse-graining the elements [27–30]. If the simulations require larger models or consideration of additional elements, like signal transduction, the process of coarse-graining can lead to a cellular automata kind of simulation [31]. In this sense, our current work may be ascribed to a cellular automata simulation”.

Point 3. no mention about the time of simulation is reported, as well as information about the stability of the systems since no RMSD is calculated for the trajectories. Information about the water model (if considered) should be reported as well as the parameters adopted for the simulation and the software package used.

Answer: We consider that the reviewer have pointed to very important issues in this comment that were lacking in our manuscript. We have included all the information addressing this comments in the manuscript.

-Information of the system and software used for the simulations, as well as, the time of simulation has been included in section 4.3:

“Simulations were run in an Intel Core i5-9400F CPU 32 GB RAM using Matlab 2020b software. As a common rule, the total number of solutions simulated for a single scenario was 10000”.

-We have included an estimation of the RMSD as the LVW forces that act deviating the positions of the elements from their stable situation. We showed the results of RMSD from the simulation in Figure 2AD in Figure 2E and we also included a brief description of the use of LVW forces as an estimation of the RMSD of the simulations in section 4.3:

“As indirect estimation of simulation noise, we used the root-mean-square deviation (RMSD) of LVW forces that deviate the element positions from their initial location”.

-We did not consider the effect of water molecules interaction ate the scale implemented in our model. We have provided with this information in a new sentence in the 4.2 section:

“At the scale considered in this coarse-grained model we have neglected the contribution of the water molecules”.

-Respecting the simulation parameters. All the parameters of the simulations in our model are determined by the dimension of the elements and the average distance between them. This information is included in Table 1.

Table 2. Model parameters estimated from mice frontal cortex membranes [25] to represent lipid classes as cylinders in the agent-based model.

Point 4. a conclusion section should be introduced in which the main findings should be summarized and in which the authors should discuss the utility of the approach and how can be reflected in Alzheimer's diagnosis/therapies

Answer: Thanks for this very convenient suggestion. Following the review’s recommendation we have added a conclusion section at the end of the manuscript:

“5.        Conclusions

In this work we have adapted our previous mathematical model for predicting lipid raft formation and evolution from the lipid composition of cell membranes of an AD animal model to sporadic AD in human. We validate the predictions from the model by comparing the results with the lipid compositions of lipid rafts reported for human frontal cortex. The model predicts that lipid rafts sizes increase in AD brains and that this increase parallels a reduction in their frequency. Both anomalies correlate with the severity of the disease. We also observed that membrane peroxidability decreases with disease severity, and that most of this reduction occurs in non-raft domains. These predictions might help to better understand the proven links be-tween nerve cell membrane domains destabilization and the early events leading to AD”.

Point 5. I highly recommend to revise the language since there are several errors through the text including the abstract.

Answer: We have reviewed the grammar and the typographic errors to improve the manuscript. Thanks for pointing out to this issue.

Round 2

Reviewer 2 Report

I really appreciated the efforts for revising the manuscript. The current version addressed my major concerns and the paper can be published