Nano-Biomechanical Analysis of a Corticosteroid Drug for Targeted Delivery into the Alveolar Air—Water Interface Using Molecular Dynamics Simulation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Probing the interaction of drugs with the lung surfactant (LS) is useful for improving drug delivery processes. To obtain the corresponding information, the authors adopted a coarse-grained molecular dynamics (MD) simulations, the MARTINI force field and the Gibbs free energy gradient calculations to perform the clinical test to reduce the experimental cost. Their work provides the helpful information for corticosteroid drug development and delivery. Overall, their work is vital for understanding the drug interacts with lung surfactant monolayer or bilayer research. Recommending that their work can be accepted after minor revision.

1. In their introduction, the authors had better richen their details to clarify why they perform this study.
2. The authors should simply compare the performance of coarse-grained molecular dynamics (MD) simulations and all atomic molecular dynamics simulation. They should simply explain why they choose coarse-grained molecular dynamics (MD) simulations.
3. In their coarse-grained molecular dynamics (MD) simulations, what is the time step？The authors should simply explain this issue.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Islam et al. employed coarse-grained molecular dynamics simulations with the MARTINI force field to parameterize the corticosteroid drug mometasone furoate and study its interaction with lung surfactant membranes. Their simulations demonstrated that drug incorporation into the lung surfactant monolayer reduces surface tension in a concentration-dependent manner, although diffusion is hindered at higher drug loadings. Overall, the findings highlight how drug concentration influences monolayer morphology and provide insights relevant to corticosteroid design and pulmonary drug delivery. While the study is interesting, several major issues must be addressed before the manuscript can be considered for publication.

1. The authors employed a coarse-grained (CG) model; however, for studying drug–lipid interactions in solution, all-atom simulations typically provide higher resolution and more reliable insights. It is unclear why a CG approach was chosen, given that all-atom simulations are now routinely achievable on the microsecond timescale using GPU-accelerated platforms. The authors should clarify their rationale for using a CG model, particularly since the primary focus of the work is on detailed drug–lipid interactions.

2. The methodology for drug parametrization is described vaguely. On page 4, the authors state: “Then, the parameters were set up by guessing for MD simulation and updating the parameters until a stable structure of the drug molecule was obtained.” It is unclear what criteria were used to generate the initial guesses and how structural stability was assessed and validated. A more systematic explanation, supported by clear validation metrics, is needed to ensure reproducibility and confirm that the derived parameters are physically meaningful.

3. The authors report: “The trajectories were collected to evaluate PMF over a total of 35 windows separated by 0.2 nm.” It is not clear whether sufficient overlap between adjacent windows was achieved. Overlap is critical for accurate reconstruction of the potential of mean force using WHAM or MBAR. Without adequate overlap, the PMF may exhibit discontinuities, large uncertainties, or artificial barriers. The authors should clarify whether overlap was monitored and provide evidence that the chosen window spacing and force constants ensured statistically reliable sampling.

4. In Figure 4b, the method for calculating error bars on the converged PMF is not described. Since error estimation is essential for assessing the reliability of free energy profiles, the authors should clarify the methodology used (e.g., block averaging, bootstrapping, or statistical uncertainty from WHAM/MBAR).

5. The manuscript would benefit from clearer and more precise writing. Several methodological descriptions, such as the parameterization strategy, umbrella sampling setup, and error analysis, are currently vague and require elaboration. Additionally, certain terms and abbreviations (e.g., APL) are introduced without definition, which may limit accessibility for readers outside the immediate field.

6. In the Drug Distribution by Clustering Analysis section, it is unclear how the authors defined a “clustered” drug. It appears that clustering was determined simply by counting the number of drug molecules within a cutoff distance. A more rigorous approach would involve calculating the center of mass (CoM) of each drug molecule at every frame and applying established clustering algorithms (e.g., DBSCAN or hierarchical clustering) to identify clusters and determine their sizes. The authors should clarify their methodology and justify their chosen approach to ensure the analysis is robust and reproducible.

7. Figure 8b shows the water density profile along the z-axis of the simulation box. In the bulk regions (|z| < 2 nm), the reported water density is below 900 kg/m³, whereas the expected bulk density for Martini water is approximately 990 kg/m³. This discrepancy may indicate issues with system equilibration, boundary effects, or data normalization. The authors should clarify the cause of this deviation and justify the reported values. Additionally, it would be helpful to indicate the z-axis in Figure 7 for clarity.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have thoroughly addressed all of my previous comments and concerns. The revisions have significantly improved the clarity and quality of the manuscript. I find the current version suitable, and I recommend it for publication in its present form.