Quality-by-Design Optimization of Mucoadhesive Trimethyl Chitosan-Coated Alginate/Dextran Sulfate Nanoparticles for Oral Insulin Delivery
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
Abstract
The abstract is overly descriptive and should be strengthened by including for example the release study data and other data obtained into the work.
Figure Captions
The figure resolution should be improved and standardized, particularly with respect to font size and lettering style.
Novelty
The manuscript claims novelty in the QbD optimization of ADS-TMC systems; however, similar alginate/chitosan/TMC nanoparticle systems for oral insulin delivery are already widely reported. The authors should better justify the novelty and clearly define the specific contribution of this work.
Conclusion
The conclusion mainly focuses on nanoparticle characterization and does not adequately summarize the key findings. The authors should briefly highlight the main results, similar to the abstract but in a more general manner. In addition, specific numerical values are typically not required in the conclusion section.
Statistical Analysis
The authors should report the statistical parameters for each model, including R² values and lack-of-fit analysis. Furthermore, the selected model for each response should be clearly justified.
Design of Experiments (DoE)
The manuscript should clearly state the targets of the DoE during the optimization process, including the criteria used for selecting the optimal formulation. In addition, standard deviation values should be reported to demonstrate the reliability of the optimized results.
Data Presentation
Some results are presented without standard deviation (SD). All experimental data should be expressed as mean ± SD to ensure consistency.
Author Response
Abstract
The abstract is overly descriptive and should be strengthened by including for example the release study data and other data obtained into the work.
Response: We thank the Reviewer for this helpful suggestion. The Abstract was revised to make it more quantitative and informative. In addition to the optimization and validation results, we now include a concise statement on the in vitro release behavior in simulated gastric and intestinal media.
Changes made in the manuscript:
The Abstract now includes the following sentence:
“In simulated gastric medium, partial insulin leakage was observed during the first 120 min, whereas cumulative insulin release reached 54% after 5 h in simulated intestinal medium.”
Location: Abstract, page 1, lines 28-29
Figure Captions
The figure resolution should be improved and standardized, particularly with respect to font size and lettering style.
Response: We agree. All figures were reformatted to improve readability and consistency. Figure captions were revised, and font style, font size, axis labels, and symbols were standardized throughout the manuscript and supplementary material.
Changes made in the manuscript:
All figures and figure captions were revised for consistency and readability.
Novelty
The manuscript claims novelty in the QbD optimization of ADS-TMC systems; however, similar alginate/chitosan/TMC nanoparticle systems for oral insulin delivery are already widely reported. The authors should better justify the novelty and clearly define the specific contribution of this work.
Response: We appreciate this comment. The Introduction was revised to clarify the novelty of the work. We now explicitly state that the novelty lies in the QbD-guided optimization of ADS/TMC nanoparticles, including the role of TMC degree of quaternization and mucoadhesion as a formulation-driving quality attribute, rather than in the general use of TMC- or alginate-based insulin carriers.
Changes made in the manuscript:
The Introduction now includes the following paragraph:
“Although TMC-containing insulin nanocarriers ……….have shown promising preclinical………………..nanoparticle performance”
Location: Introduction, page 3, lines 93-101
Conclusion
The conclusion mainly focuses on nanoparticle characterization and does not adequately summarize the key findings. The authors should briefly highlight the main results, similar to the abstract but in a more general manner. In addition, specific numerical values are typically not required in the conclusion section.
Response: Authors agree. The Conclusions section was fully revised to summarize the main results more clearly, including the outcomes of the screening design, the Box-Behnken optimization, the desirability-selected optimal formulation, and the successful experimental validation.
Changes made in the manuscript:
The Conclusions section now emphasizes:
-the key factors identified in the screening design,
-the variables optimized in the Box-Behnken design,
-the optimized formulation selected by desirability analysis,
-the experimental confirmation with prediction errors below 5%.
Location: Conclusions, page 18, lines 635-650
Statistical Analysis
The authors should report the statistical parameters for each model, including R² values and lack-of-fit analysis. Furthermore, the selected model for each response should be clearly justified.
Response: We thank the Reviewer for this useful suggestion. We clarified in the manuscript that model adequacy was assessed by ANOVA and regression statistics including R², adjusted R², and predicted R². We also explicitly indicated that the detailed outputs are available in the Supplementary Data.
Changes made in the manuscript:
The following sentence was added:
“Detailed values of R², adjusted R², predicted R², and coefficient estimates are provided in Table S1.”
Location: Table S1, page 21, lines 694-696
Design of Experiments (DoE)
The manuscript should clearly state the targets of the DoE during the optimization process, including the criteria used for selecting the optimal formulation. In addition, standard deviation values should be reported to demonstrate the reliability of the optimized results.
Response: We agree. The optimization criteria were rewritten more explicitly. The manuscript now clearly states that the optimization goals were minimization of particle size, attainment of a sufficiently positive zeta potential compatible with aqueous colloidal stability, and maximization of in vitro mucoadhesion.
Changes made in the manuscript:
The relevant section now states:
“Numerical optimization was performed using the desirability function, with the goals of minimizing particle size, obtaining a sufficiently positive zeta potential compatible with aqueous colloidal stability, and maximizing in vitro mucoadhesion.”
Location: Response Surface Optimization of the Formulation by Box-Behnken Design Optimization, page 17, lines 592-594
Data Presentation
Some results are presented without standard deviation (SD). All experimental data should be expressed as mean ± SD to ensure consistency.
Response: We thank the Reviewer for identifying this point. The manuscript was reviewed to improve consistency in statistical reporting, and the presentation of experimental values as mean ± standard deviation was clarified where appropriate.
Changes made in the manuscript:
Clarifying sentences were added to the Results and figure captions where needed, for example:
“The mean particle size of the TMC-coated nanoparticles was 387 ± 31 nm, calculated from measurements obtained for three independently prepared batches, with a PDI of 0.40 ± 0.04…..........nanoparticles.”
Location: Results, page 5, lines 194-196
Reviewer 2 Report
Comments and Suggestions for Authors
This article examines trimethyl chitosan (TMC)-coated alginate/dextran sulfate (ADS) anoparticles, which were developed as mucoadhesive nanocarriers for oral insulin delivery using a "Quality by Design" strategy. The article is relevant and contains some novel elements, but there are some concerns regarding the methodology.
1. The Ishikawa diagram in Figure 1 is too small, and some branches are not labeled but are shown. It is necessary to remove unnecessary branches and provide a more detailed explanation of each branch of the diagram and its impact on the quality of TMC-coated nanoparticles. Otherwise, it is simply a meaningless image.
2. Insert a list of abbreviations at the end of the article (or at the beginning). Its absence confuses the reader when reading the text.
3. Move Section 3.2.1.3. Intrinsic Viscosity and Viscosities Average Molecular Weight (Mv) to Section 3.2.3. Physicochemical characterization (specify).
4. In section 3.2.3.4. Insulin Release Studies, clarify the pH of the solution for the study.
5. Section 3.2.3.7. Fourier-Transform Infrared Analysis does not fully present the obtained results. It is necessary to clarify why the fingerprint zone is missing in the range from 800 to 400 cm-1.
6. The FTIR spectra of all studied substances and components should be uniform, fingerprint zones should be shown, and their description should be provided.
7. Why is the Insulin Releasing Profile shown only for the nanoparticle formulation coated with TMC at 0.3? Where are the other results?
8. The results of the Response Surface Optimization and Desirability Analysis raise many questions. The reviewer is unsure whether the authors were able to select the factors that influence insulin release from nanoparticles. The response surfaces demonstrate that optimal values for the parameters being modified were not achieved. The authors' conclusions do not correspond to the results.
Author Response
This article examines trimethyl chitosan (TMC)-coated alginate/dextran sulfate (ADS) anoparticles, which were developed as mucoadhesive nanocarriers for oral insulin delivery using a "Quality by Design" strategy. The article is relevant and contains some novel elements, but there are some concerns regarding the methodology.
1. The Ishikawa diagram in Figure 1 is too small, and some branches are not labeled but are shown. It is necessary to remove unnecessary branches and provide a more detailed explanation of each branch of the diagram and its impact on the quality of TMC-coated nanoparticles. Otherwise, it is simply a meaningless image.
Response: We agree. Figure 1 was improved, and the text was expanded to explain the meaning of each main branch and its impact on nanoparticle quality.
Changes made in the manuscript:
The explanatory paragraph after Figure 1 has been rewritten as follows:
“In the Ishikawa diagram, the Material branch includes the chemical attributes of the polymers, drug, crosslinker, and stabilizer, which directly affect electrostatic complexation, particle formation, and insulin retention. The Method branch reflects critical preparation steps such as ionotropic gelation and polyelectrolyte complexation, which govern core formation and coating efficiency. The Equipment branch includes stirring and pumping conditions that may influence mixing and particle formation, whereas the Environment branch includes pH and temperature, which can alter polymer ioni-zation and colloidal stability. Finally, the Measurement and Operator branches reflect analytical robustness and handling variability, both of which may affect the reproducibility of the final CQAs. Among these variables, the polymer, drug, and surfactant concentrations were selected as critical factors for the present study based on literature data [19, 20] and preliminary process knowledge.“
Location: Introduction, page 3, lines 104-115
2. Insert a list of abbreviations at the end of the article (or at the beginning). Its absence confuses the reader when reading the text.
Response: We agree and added an Abbreviations section to improve readability.
Changes made in the manuscript:
An abbreviations list was added.
Location: Introduction, page 19, lines 659-675
3. Move Section 3.2.1.3. Intrinsic Viscosity and Viscosities Average Molecular Weight (Mv) to Section 3.2.3. Physicochemical characterization (specify).
Response: We agree. This subsection was moved to the characterization section in the revised manuscript.
Changes made in the manuscript:
The subsection on intrinsic viscosity and viscosity-average molecular weight was relocated accordingly.
4. In section 3.2.3.4. Insulin Release Studies, clarify the pH of the solution for the study.
Response: We agree and have revised the section to specify the pH values of the gastric and intestinal media.
Changes made in the manuscript:
The release-method description was updated to state the pH values explicitly.
Location: Methods page 15, lines 506-507
5. Section 3.2.3.7. Fourier-Transform Infrared Analysis does not fully present the obtained results. It is necessary to clarify why the fingerprint zone is missing in the range from 800 to 400 cm-1.
Response: We thank the Reviewer for this observation. We clarified that FTIR spectra were recorded over the range 750-3750 cm⁻¹, and that the region below 750 cm⁻¹ was therefore not accessible under the selected acquisition conditions.
Changes made in the manuscript:
The following clarification was added:
“Although the full 400-750 cm⁻¹ fingerprint region was not recorded, the lower-wavenumber portion that was experimentally accessible was retained for all samples and discussed only in relation to the most informative polysaccharide bands.”
Location: Results, page 6, lines 218-220
6. The FTIR spectra of all studied substances and components should be uniform, fingerprint zones should be shown, and their description should be provided.
Response: We agree. The FTIR spectra were reformatted so that all samples are shown over the same accessible range, and the discussion was expanded accordingly.
Changes made in the manuscript:
The Results section now includes:
“FTIR spectra of the formulations are shown in Figure 5. All spectra were plotted over the same experimentally accessible range of 750–3750 cm⁻¹.The broad band observed in all FTIR spectra between 2360 and 3720 cm-1 was assigned to the characteristic C-H, N-H, and O-H stretches. The band at 1482 cm−1 (spectrum C) was attributed to the expected angular deformation of C-H bonds of the N-methyl group-exhibiting TMC structure [37], whereas the band at 1642 cm−1 observed for TMC-coated nano-particles (spectrum B) was attributed to C=O bonds of secondary amide groups, arising from acetylated residues that remain in the TMC structure and from axial asymmetric deformation of carboxylates. The band observed at approximately 1420 cm⁻¹ is attributed to the symmetric stretching of carboxylate groups [38]. In the nanoparticle spectrum, this band appears with lower intensity than in the ADS spectrum (spectrum a), which can be explained by partial protonation of the alginate carboxylate groups and by their participation, together with dextran sulfate groups, in electrostatic complexation with the positively charged sites of TMC. Although the full 400-750 cm⁻¹ fingerprint region was not recorded, the lower-wavenumber portion that was experimentally accessible was retained for all samples and discussed only in relation to the most informative polysaccharide bands.”
Location: Results page 6, lines 205-206 and 218-220
7. Why is the Insulin Releasing Profile shown only for the nanoparticle formulation coated with TMC at 0.3? Where are the other results?
Response: We thank the Reviewer for this important point. The insulin release study was not included as a response in the DoE models and was instead performed as a proof-of-concept functional assay using one representative formulation. We clarified this rationale in the manuscript.
Changes made in the manuscript:
The following clarification was added:
“Although insulin release was not included as a CQA in the optimization workflow, it was evaluated as a proof-of-concept functional test using a representative TMC-coated formulation selected from the initial screening stage. This formulation was chosen because it displayed submicron size and no detectable micrometric population by laser diffraction, making it suitable for an initial assessment of pH-dependent insulin release.“
Location: Results, page 7, lines 226-229
8. The results of the Response Surface Optimization and Desirability Analysis raise many questions. The reviewer is unsure whether the authors were able to select the factors that influence insulin release from nanoparticles. The response surfaces demonstrate that optimal values for the parameters being modified were not achieved. The authors' conclusions do not correspond to the results.
Response: We appreciate this comment and revised the manuscript to clarify the scope of the optimization. The response-surface analysis was performed only for the selected CQAs, namely particle size, zeta potential, and in vitro mucoadhesion. Insulin release was not included as a modeled response. We also clarified that the optimal formulation identified by desirability analysis represents the best compromise among the modeled CQAs within the investigated experimental domain.
Changes made in the manuscript:
The optimization section now states:
“The response-surface optimization was performed only for the CQAs selected in the Box-Behnken design, namely particle size, zeta potential, and in vitro mucoadhesion. Insulin release was not included as a modeled response and was instead evaluated separately as a complementary proof-of-concept functional parameter.”
Location: Results, page 11, lines 315-318
and:
“The optimized formulation identified by desirability analysis should therefore be interpreted as the best compromise among the three modeled CQAs within the experimental factor space investigated in this study, rather than as a universal optimum for all possible formulation properties.”
Location: Results, page 12, lines 375-378
The Conclusions were revised accordingly.
Reviewer 3 Report
Comments and Suggestions for Authors
Diabetes is a serious worldiwide health issue. Insulin delivery efficiency is of vast importance. the authors suggest use of polysaccharide-based composite particles to do it. Polysaccharides are very well known for their abundancy, biocompatibility and safety. However, their interaction with human tissiues and stability under different pH levels is of question. The manuscript fits aims and scope of Marine Drugs well. The design of the study is good. However, there are some critical points that need to be revised prior to publication. Therefore I believe that the manuscript after major revision.
First and foremost, unfortunately, the paper lacks evidence of nanoparticles formation and their structure, which can be proved by microscopic techniques.
In Introduction Section authors should emphasize why exactly the oral delivery systems of insulin are needed. Injections of insulin quite popular and effective.
Fig. 2. Y-axis caption. You should either add percentage values or avoid using % and use relative units instead.
Line 177. "387±31 nm " According to Fig. 2 the error is much bigger. How did you calculate this error?
Unfortunately, DLS cannot give evidence of absence of micrometric contamination. If particles sediment, they will not be seen in the graph.
3.2.3.1. Granulometric size distribution. And again, DLS is not evident of granulometric composition of samples.
Author Response
Diabetes is a serious worldiwide health issue. Insulin delivery efficiency is of vast importance. the authors suggest use of polysaccharide-based composite particles to do it. Polysaccharides are very well known for their abundancy, biocompatibility and safety. However, their interaction with human tissiues and stability under different pH levels is of question. The manuscript fits aims and scope of Marine Drugs well. The design of the study is good. However, there are some critical points that need to be revised prior to publication. Therefore I believe that the manuscript after major revision.
First and foremost, unfortunately, the paper lacks evidence of nanoparticles formation and their structure, which can be proved by microscopic techniques.
Response: We acknowledge this limitation. At this stage, nanoparticle formation was inferred from DLS, laser diffraction, zeta potential, FTIR, and mucoadhesion data, but no direct microscopy-based characterization was included. In response to the Reviewer’s concern, we revised the manuscript to avoid overstatement regarding nanoparticle morphology and added a limitation statement indicating that future work should include TEM or SEM.
Changes made in the manuscript:
A limitation statement was added:
“A limitation of the present study is the absence of direct microscopic visualization of the nanoparticles. Although DLS, laser diffraction, zeta potential, FTIR, and mucoadhesion results support the formation of nanoscale ADS/TMC polyelectrolyte complexes, future work should include microscopy-based characterization to confirm particle morphology and structural organization.”
Location: Conclusions, page 18, lines 644-648
In Introduction Section authors should emphasize why exactly the oral delivery systems of insulin are needed. Injections of insulin quite popular and effective.
The manuscript should better explain why oral insulin delivery is needed when subcutaneous injections are already effective.
Response: We appreciate this comment. The Introduction was revised to more clearly explain the interest of oral insulin delivery despite the effectiveness of subcutaneous administration.
Changes made in the manuscript:
The following sentence was added to the Introduction:
“Nevertheless, oral insulin remains attractive because it is non-invasive, may improve patient acceptance and adherence, and could partially reproduce the physiological portal route of endogenous insulin delivery before systemic distribution.”
Location: Introduction, pages 1-2, lines 40-42
Fig. 2. Y-axis caption. You should either add percentage values or avoid using % and use relative units instead.
Response: We thank the Reviewer for this observation. The corresponding figure was revised to ensure that the axis labeling accurately reflects the plotted data, and the caption was updated for clarity.
Changes made in the manuscript:
Figure labeling and caption were corrected.
Line 177. "387±31 nm " According to Fig. 2 the error is much bigger. How did you calculate this error?
Response: We appreciate this comment. The manuscript now explicitly states how the reported mean size was obtained.
Changes made in the manuscript:
The following sentence was added:
“The mean particle size of the TMC-coated nanoparticles was 387 ± 31 nm, calculated from measurements obtained for three independently prepared batches, with a PDI of 0.40 ± 0.04…......nanoparticles.”
Location: Results, page 5, lines 194-196
Unfortunately, DLS cannot give evidence of absence of micrometric contamination. If particles sediment, they will not be seen in the graph.
Response: We agree. The text was revised to clarify that DLS was used to estimate hydrodynamic size distribution, while laser diffraction was used as a complementary method to screen for possible larger particle populations.
Changes made in the manuscript:
The relevant sentence now reads:
“Nanoparticle size was assessed by dynamic light scattering, whereas laser diffraction was used as a complementary method to screen for the possible presence of larger particle populations.“
Location: Results, page 5, lines 192-194
3.2.3.1. Granulometric size distribution. And again, DLS is not evident of granulometric composition of samples.
Response: We thank the Reviewer for this important clarification. We revised the wording throughout the manuscript to avoid overinterpretation of DLS data and now refer to “hydrodynamic size distribution” or “size profile” where appropriate.
Round 2
Reviewer 2 Report
Comments and Suggestions for Authors
The authors have answered the questions satisfactorily and made the appropriate corrections to the article. The article is now ready for publication.
Reviewer 3 Report
Comments and Suggestions for Authors
The authors reasonably answered raised questions. Therefore I believe that the manuscript can be published in present form.
