Skip to Content
MDPIMDPI

Advanced

by
  • Yong Eun Jang 1,†,
  • Minjun Kwon 2,† and
  • Gwang Lee 2,4,*
  • et al.

Reviewer 1: Wenjin Li Reviewer 2: Anonymous Reviewer 3: Enrico Gallo Reviewer 4: Anonymous

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors reviewed works on peptide therapeutics covering chemical modification strategies, cell-penetrating peptide-based delivery systems, and artificial intelligence (AI)-assisted design. Although the review covers a broad range of topics, the description of each topic lacks depth if not being superficial. Compared with recent in-deep reviews [1–3], this paper offers readers with limited if not little useful information. In addition, the review has the following issues:

  1. Section 2.1 relies heavily on references 7 (cited eight times) and 32 (cited five times), suggesting that this part draws mainly from reference 7 and partly from reference 32. Similarly, Section 3.1.2 appears to depend largely on reference 64, and Section 3.1.4 on references 76 and 79.
  2. There exists considerable redundancy in the text. For example, the text on lines 144–148 on page 4 has already been mentioned in the Introduction, and the content on lines 203–204 on page 6 repeats information found on page 2, line 49.
  3. Several paragraphs provide little useful information. Examples include lines 184–191 on page 5, lines 199–200 on page 5, and lines 629–637 on page 17.
  4. Many key references are outdated. In Section 3.1.1, the main research examples cited (references 43, 57, and 61) are publications 10 to 20 years ago.
  5. References are required for some statements, such as those on page 9, lines 376–377, and on page 13, lines 528–529.
  6. "Future Perspectives and Development Trends" (Section 6) is supposed to provide valuable insights in a review article. However, the current version remains largely generic and fails to provide in-depth and informative perspectives.
  7. Cited references 214-217 in lines 585-587, on page 15, may be unnecessary.

 

[1] Peptide-based drug design using generative AI Chem. Commun., 2026, 62, 672-691. DOI: 10.1039/D5CC04998A

[2] Emerging landscape of bioinformatics and artificial intelligence applications in cell-penetrating peptide-based delivery Expert Opin Drug Deliv, 2026 Mar;23(3):495-515. doi: 0.1080/17425247.2025.2587940.

[3] Strategies for the design of biomimetic cell-penetrating peptides using AI-driven in silico tools for drug delivery. Biomaterials advances, Volume 169, 2025. PP 214153-214153. DOI: 10.1016/J.BIOADV.2024.214153

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript discusses strategies for developing peptide drugs with high therapeutic potential and for overcoming intrinsic limitations through cell-penetrating peptides and AI-based approaches. However, the intended readership is not clearly defined. The review devotes substantial space to well-established peptide-mimetic strategies, while the section on AI as a tool to overcome the issues is barely described with a lack of connection with them. In particular, the manuscript should provide a more clear and concrete discussion of how AI is applied in peptide drug design, rather than relying primarily on citations. Expanding this section would better align the content with the stated scope and strengthen the overall impact of the review.

Additional comments:

The title does not fully reflect the content and should be revised accordingly.

The abstract states that AI tools can accelerate peptide drug development through the design of highly active compounds with improved bioavailability. To better reflect the broader scope of the manuscript, CPPs should be described as one of the strategies.

“Nanocarrier” is mentioned in the text one time (line 467) and shouldn’t be included in the keywords.

Lines 73–74: The statement beginning with “As a result…” should be revised. Peptide modifications require systematic exploration of large sequence spaces and are not limited to CPPs.

Lines 88–92: It is unclear what the authors aim to convey in this part. Please clarify and refine the statement.

Lines 118–119: The statement is generally incorrect. Peptides often exhibit higher immunogenic potential than antibodies, especially when unmodified or foreign to the human proteome. The authors discussed the issue later in Section 2.2.

Section 2.2 title should be revised from “Challenges of Peptide Therapeutics” to “Challenges in Peptide Therapeutics.”

Figure 1 contains incorrect structures. Please revise panel (b), confirm whether the structure in panel (d) is L rather than K, and verify whether “NAc” or “N-ac” is intended.

Line 199 should be revised to “solid- or solution-phase.”

Section 3.1 mentions amide-bond alterations, but these are not included in the figure or cited in the text.

Line 231 mentions hydrocarbon stapling, but this approach is not shown in Figure 1a.

Lines 250–251: The phrase “potential for reduced binding affinity” should be removed or revised, since cyclization design is generally based on SAR studies and often aims to improve, rather than reduce, binding affinity.

Figure 2: Abbreviations such as GNN, CNN, and ESM-2 should be defined.

Figure 3: Abbreviations should also be defined.

Section 6, line 641: Since AI includes machine learning, “AI and ML algorithms” should be revised to “AI, specifically ML algorithms.”

Lines 640–650: The discussion of wet-lab validation and clinical evaluation of targeted peptide delivery is unclear and should be revised.

Lines 651–652: Please clarify the intended meaning. If the authors mean that undruggable targets with delivery issues can become druggable once toxicity is reduced, this should be stated more clearly.

Section 6 should be revised to convey a clearer message.

The conclusion states the therapeutic potential and limitations of peptides, but it does not clearly address the role of AI, as the reviewer pointed out earlier. The manuscript should explain how AI contributes to peptide drug development clearly.

The discussion of chemical engineering and next-generation delivery platforms should be clarified.

CPP-based delivery strategies represent only one of several approaches to peptide delivery and should not be presented as the sole solution. Other strategies for improving peptide delivery should also be stated briefly.

Overall, the two main themes, peptide therapeutics and AI utilizations, are not sufficiently linked, which may confuse readers. The AI section should include a clear take-home message.

Abbreviations should be added consistently in tables and figures, and full terms and abbreviations should not be mixed inconsistently throughout the text.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The review by Jang et al. provides a comprehensive overview of peptide therapeutics, chemical modification strategies, cell-penetrating peptide (CPP)-based delivery systems, and AI-guided peptide drug design. The manuscript is informative and timely; however, several points require clarification or expansion to enhance rigor, interpretability, and translational relevance. The following comments are intended to guide the authors in strengthening the review:

  1. The authors should specify what practical criteria they propose for selecting between small molecules, peptides, and antibodies in drug development.
  2. How do the authors operationally define “peptide therapeutics” in this review.
  3. Could the authors provide references supporting the claim that peptides can achieve selectivity “comparable to antibodies”?
  4. Beyond the figure 1, could the authors provide also a comparative table summarizing the relative advantages and limitations of each chemical modification strategy for clarity?
  5. While the benefits of cyclization are described, could the authors provide quantitative or comparative data demonstrating improvements in proteolytic stability and cell permeability?
  6. Could the potential immunogenic implications of D-amino acids and non-canonical amino acids (NCAAs) be discussed in more detail, particularly in the context of translational development?
  7. Could the authors explicitly connect the chemical modification strategies to CPP-based delivery and other peptide-based delivery platforms, including peptide-derived nanoparticles, highlighting which modifications are compatible with or complementary to these approaches? In this context, the authors could cite the following paper concerning the development of peptide-based nanoparticles for delivery: https://doi.org/10.3390/biomedicines13061415 https://doi.org/10.3390/pharmaceutics17020263 https://doi.org/10.1038/s41598-022-23940-0Could the authors discuss the potential integration of peptide-based nanoparticles as a complementary delivery strategy to CPPs, particularly for enhancing intracellular targeting, drug stability, and controlled release?
  1. Could the authors clarify the criteria used to define CPPs versus other intracellular delivery systems (e.g., exosomes, nanoparticles, etc.)?
  2. Are there comparative data on the relative efficiency of different CPP classes for intracellular delivery in vitro and in vivo
  3. The section AI-guided peptide drug design lists multiple CPP datasets (MLCPP2.0, CPPsite1–3, CPP924, CPP1708). Could the authors discuss differences in dataset size, sequence diversity, and experimental validation quality?
  4. Please clarify terminology where “peptide activity” and “therapeutic efficacy” are used interchangeably.
  5. Some sentences in the review are long and could be split to improve readability. Please revise all text.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Comments are in the file beiow.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Athough the authors have make changes to many paragraphs, they did not resolve directly to a critical issue in the previous comments, that is, compared with recent in-deep reviews [1–3], what are the useful information that this paper offers to readers? This issue should have been addressed in the introduction part.  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have responded thoroughly to all of my comments. In my opinion, the manuscript has significantly improved in terms of quality and accuracy, and it is now more readable and useful for the reader. In its current form, I believe the manuscript can be accepted for publication without the need for further revisions.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors,

the revised manuscript shows, that you understood my remarks and inserted some sentences about the large complexity of delivery related properties of cell-penetration. These remarks in lines 716 to 726 and 740 to 744 are very similar. Furthermore, I could not find References, which exclusively describe this complexity of internalization processes. Your manuscript can be published after minor revision.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed the previous concern.