Plant-Based Protein Bioinks with Transglutaminase Crosslinking: 3D Printability and Molecular Insights from NMR and Synchrotron-FTIR

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript investigates the application of TGase to improve the printability and textural properties of four plant protein isolates (soy, pea, mung bean, fava bean) for 3D food printing. The study integrates SDS-PAGE, 1H NMR, rheological characterization, printing performance evaluation, texture profile analysis (TPA), and SEM images to provide a comprehensive assessment. The research addresses a timely topic in sustainable food technology and presents valuable comparative data across multiple protein sources. While the dataset is promising and the work is methodologically sound in principle, significant issues with experimental clarity, unit consistency, and data presentation must be addressed to meet the standards.

• Line 219-220. The linear viscoelastic region (LVR) of the samples should first be determined. The authors are advised to provide the full amplitude sweep data in either the main text or supplementary materials to support the claims presented in Lines 219–
• Line 304-306. To fully eliminate the confounding effect of casein, the authors are advised to explicitly report its exact concentration in the experimental system.
• Line 316-318. The formatting of P<0.05 should be kept consistent throughout the manuscript.
• The sample names should be clearly labeled adjacent to the ¹H NMR spectra in both Figure 1A and Figure 2.
• Line 361-362. Is the notation “(B)” in the caption of Figure 3 a typographical error? Additionally, please clarify why the data corresponding to FP are not provided in Figure 3B.
• It is recommended that the enzyme units in the captions of Figures 5 and 6 be standardized, “U/g protein” or “U/ g plant-based protein”. Additional, the expression of enzyme concentration in Line 511 is inconsistent with that used throughout the rest of the manuscript, which may lead to misinterpretation by readers.

Author Response

Comment 1: This manuscript investigates the application of TGase to improve the printability and textural properties of four plant protein isolates (soy, pea, mung bean, fava bean) for 3D food printing. The study integrates SDS-PAGE, 1H NMR, rheological characterization, printing performance evaluation, texture profile analysis (TPA), and SEM images to provide a comprehensive assessment. The research addresses a timely topic in sustainable food technology and presents valuable comparative data across multiple protein sources. While the dataset is promising and the work is methodologically sound in principle, significant issues with experimental clarity, unit consistency, and data presentation must be addressed to meet the standards.

Response 1: We sincerely thank the reviewer for the careful evaluation of our manuscript. All comments have been carefully addressed, and the manuscript has been revised accordingly. Our detailed responses are provided below, with corresponding revisions indicated in the revised manuscript and have been highlighted in yellow.

Comment 2: Line 219-220. The linear viscoelastic region (LVR) of the samples should first be determined. The authors are advised to provide the full amplitude sweep data in either the main text or supplementary materials to support the claims presented in Lines 219–

Response 2: We agree with the reviewer. Prior to frequency sweep measurements, amplitude sweep tests were conducted for all protein systems to determine the LVR. In the revised manuscript, we have clarified this procedure in the rheological methods section and explicitly stated the strain value selected within the LVR. See line 227.

Comment 3: Line 304-306. To fully eliminate the confounding effect of casein, the authors are advised to explicitly report its exact concentration in the experimental system.

Response 3: Thank you for this important point. The exact concentration of casein used in the experimental system has now been explicitly stated in the Materials and Methods section. This clarification ensures transparency and confirms that the casein level was constant across all treatments, thereby minimizing any confounding effects on the observed TGase-induced crosslinking behavior. See line 312.

Comment 4: Line 316-318. The formatting of P<0.05 should be kept consistent throughout the manuscript.

Response 4:  This has been corrected throughout the entire manuscript.

Comment 5: The sample names should be clearly labeled adjacent to the ¹H NMR spectra in both Figure 1A and Figure 2.

Response 5:  Figures 1A and 2 have been revised. See line 381 and 386.

Comment 6: Line 361-362. Is the notation “(B)” in the caption of Figure 3 a typographical error? Additionally, please clarify why the data corresponding to FP are not provided in Figure 3B.

Response 6:  Corrected. See Fig 3.

Comment 7: It is recommended that the enzyme units in the captions of Figures 5 and 6 be standardized, “U/g protein” or “U/ g plant-based protein”. Additional, the expression of enzyme concentration in Line 511 is inconsistent with that used throughout the rest of the manuscript, which may lead to misinterpretation by readers.

Response 7:  Corrected. See Fig 5.

Reviewer 2 Report

Comments and Suggestions for Authors

1. In the Introduction, where the limitations of plant proteins are discussed, the authors should also consider and explicitly include relevant processing conditions. In addition, several statements prematurely anticipate conclusions and should be revised or relocated to the Results and Discussion section.             

2. The information related to NMR presented in the Introduction does not appear to be directly relevant to the scope or methodology of the current study.
3. The scientific objective of the study is currently formulated in an unclear manner. The authors should explicitly identify the specific gap in the existing literature and clearly articulate the novel contribution of the present study relative to prior work.
4. In Section 2.2, the rationale for analyzing hydrated samples should be clearly justified. Additionally, the activity unit must be clarified and consistently reported (U/mL solution versus U/g protein).
5. In Section 2.3, it should be clearly stated that the NBS assay quantifies primary amines and does not directly measure crosslinking.
6. In Section 2.4, the use of mushroom water extract is questionable for the objectives of this study, and the described metabolomic analysis does not appear to be directly relevant to protein crosslinking. This methodology should be reconsidered or better justified.
7. In Section 2.5, an exceptionally large number of spectra (180 per treatment) is reported without a clear experimental or statistical justification.
8. In Section 2.8, key printing parameters are not specified, including how they were selected and on what basis. Moreover, paprika oleoresin is introduced without justification and is not presented or discussed in the Introduction.
9. In Section 3.1.1, should correct the interpretation of the TNBS assay: TNBS measures the decrease in accessible primary amines rather than network formation, and it should be treated as a chemical proxy rather than a direct metric of printability.
10. In Section 3.1.1, the manuscript lacks experimental data demonstrating the actual Gln/Lys content of each protein isolate, as well as information on solubility, controlled denaturation, or DSC analysis. Corresponding claims should therefore be revised or qualified.                                                                                         

11. In Section 3.1.2, calibration, linearity, and reproducibility are not demonstrated, despite the statement that spectral intensities can be used to assess crosslinking extent. Moreover, the analysis resembles a metabolomic profiling rather than a protein crosslinking assessment, and several claims are made without supporting binding experiments or proteolysis/hydrolysis measurements.
12. In Section 3.1.2, there is an inconsistency: the manuscript first states that markers (1.9–4.1 ppm) can be used for quantification, and subsequently claims that quantification is not possible.
13. In section 3.1.2, activities of 10, 20, and 50 U/g are discussed, whereas Figure 2 presents data only for 10 U/g.
14. In Section 3.2.1, TGase activity is inconsistently reported: the text refers to 6.25 U/g (2.5% TGase powder), while Figure 6 denotes Low (2.5 U/g) and High (12.5 U/g).
15. The manuscript states "frequency-dependent transitions from solid-like to liquid-like behavior"; however, tan δ values ​​remain well below 1 (≈0.08–0.20) across the measured range, indicating predominantly solid-like or gel-like behavior rather than a transition towards liquid-like behavior.
16. In Section 3.2.2, the discussion should be logically reorganized (introduction, heating trend, mechanistic explanation, cooling trend, and comparison between proteins and TGase). The respective contributions of TGase-induced crosslinking and thermal aggregation should be clearly distinguished. Furthermore, the discussion of tan δ trends should be internally consistent, explicitly addressing frequency dependence and specifying the relevant frequency ranges (e.g., 1–10 Hz versus 50–100 Hz).
17. In the title or caption of Figure 6, it is stated that "line styles indicate plant protein type," whereas the figure actually uses colors and symbols.
18. Please justify why shear-thinning behavior and printability are inferred based on oscillatory frequency sweep tests rather than steady shear or extrusion-relevant measurements.
19. In Section 3.3, the manuscript states that pea protein (PP) exhibits the lowest response, likely due to limited accessibility, while FTIR results simultaneously suggest that PP undergoes the highest β-sheet reorganization.
20. In the SEM analysis, images at 50× and 100× magnification do not sufficiently support claims regarding protective or network structures.
21. The Conclusions section should be entirely rewritten, as it currently overestimates the role of TGase activity without adequately considering the contribution of formulation additives and carrier proteins present in the commercial enzyme preparation.

Author Response

Comment 1: 1. In the Introduction, where the limitations of plant proteins are discussed, the authors should also consider and explicitly include relevant processing conditions. In addition, several statements prematurely anticipate conclusions and should be revised or relocated to the Results and Discussion section.             

Response 1:  The introduction has been revised to explicitly include processing-related limitations of plant proteins relevant to 3D food printing, and highlighted in green. See line 63-68, 83-88, 103.

Comment 2: 2. The information related to NMR presented in the Introduction does not appear to be directly relevant to the scope or methodology of the current study.
Response 2: We acknowledge this point. This section has been substantially reduced and refocused, retaining only information necessary to justify NMR as a supportive structural probe rather than a primary analytical endpoint. See line 114.

Comment 3: 3. The scientific objective of the study is currently formulated in an unclear manner. The authors should explicitly identify the specific gap in the existing literature and clearly articulate the novel contribution of the present study relative to prior work.
Response 3: We have rewritten. See line 119.

Comment 4: 4. In Section 2.2, the rationale for analyzing hydrated samples should be clearly justified. Additionally, the activity unit must be clarified and consistently reported (U/mL solution versus U/g protein).
Response 4:  Enzyme activity units have been fully standardized to U/g plant protein throughout the manuscript, and any previous ambiguity (U/mL vs. U/g) has been corrected. See line 144.

Comment 5: 5. In Section 2.3, it should be clearly stated that the NBS assay quantifies primary amines and does not directly measure crosslinking.
Response 5:  We agree that the TNBS assay does not directly measure protein network formation or crosslinking, but rather quantifies the availability of accessible primary amine groups. In the revised manuscript, Section 2.3 has been explicitly revised to state that the NBS assay is used as a chemical proxy for TGase-mediated crosslinking by monitoring reductions in free ε-amino groups, and that the results should be interpreted as indirect evidence of enzymatic modification rather than a direct measurement of crosslink density.

Comment 6: 6. In Section 2.4, the use of mushroom water extract is questionable for the objectives of this study, and the described metabolomic analysis does not appear to be directly relevant to protein crosslinking. This methodology should be reconsidered or better justified.
Response 6:  The use of mushroom water extract and the associated ¹H NMR conditions was not intended to investigate metabolomics or crosslinking mechanisms directly. Instead, this methodology was adopted solely as a validated sample preparation and NMR acquisition framework, based on our previous work (Sai-Ut et al., 2024), where aqueous protein-rich extracts were successfully prepared and analyzed by ¹H NMR with high spectral resolution and reproducibility.

Comment 7: 7. In Section 2.5, an exceptionally large number of spectra (180 per treatment) is reported without a clear experimental or statistical justification.
Response 7:  The reported number of spectra reflects technical spectral mapping rather than independent experimental replication. Each treatment consisted of six independently prepared biological replicates, and for each replicate, spectra were collected from approximately 30 randomly selected micro-areas to account for the inherent microscale heterogeneity of powdered protein samples when analyzed by synchrotron radiation–based FTIR. This approach resulted in 180 raw spectra per treatment (30 spectra × 6 replicates).

Comment 8: 8. In Section 2.8, key printing parameters are not specified, including how they were selected and on what basis. Moreover, paprika oleoresin is introduced without justification and is not presented or discussed in the Introduction.
Response 8:  We have revised Section 2.8 to explicitly specify all key 3D printing parameters and to clarify the rationale for their selection. Printing conditions (nozzle diameter, printing speed, layer height, and infill density) were selected based on preliminary optimization trials and literature precedents to ensure stable extrusion, adequate layer adhesion, and dimensional fidelity across all bioink formulations. This justification has now been clearly stated in the revised manuscript.  In addition, the role of paprika oleoresin has been clarified. Paprika oleoresin was incorporated at a low concentration solely as a visual tracer to enhance color contrast during printing and facilitate accurate post-printing dimensional assessment. It was not intended to participate in protein crosslinking or network formation. This clarification has been explicitly added to Section 2.8 to avoid ambiguity regarding its functional role. See line 250, 260.

Comment 9: 9. In Section 3.1.1, should correct the interpretation of the TNBS assay: TNBS measures the decrease in accessible primary amines rather than network formation, and it should be treated as a chemical proxy rather than a direct metric of printability.
Response 9: The interpretation has been revised to consistently describe TNBS results as reflecting reductions in accessible primary amines. See line 304.

Comment 10: 10. In Section 3.1.1, the manuscript lacks experimental data demonstrating the actual Gln/Lys content of each protein isolate, as well as information on solubility, controlled denaturation, or DSC analysis. Corresponding claims should therefore be revised or qualified.                                                                                         

Response 10: We acknowledge that the present study did not experimentally determine the glutamine/lysine content, solubility, or thermal denaturation behavior (e.g., DSC) of the individual protein isolates. Accordingly, Section 3.1.1 has been carefully revised to qualify all mechanistic interpretations, ensuring that they are presented as literature-supported inferences rather than direct experimental evidence from this work. See line 316.

Comment 11: 11. In Section 3.1.2, calibration, linearity, and reproducibility are not demonstrated, despite the statement that spectral intensities can be used to assess crosslinking extent. Moreover, the analysis resembles a metabolomic profiling rather than a protein crosslinking assessment, and several claims are made without supporting binding experiments or proteolysis/hydrolysis measurements.
Response 11: Accordingly, Section 3.1.2 has been revised to remove statements implying quantitative assessment of crosslinking extent based on spectral intensities. The role of ¹H NMR in this study is now clearly defined as a qualitative and comparative analytical tool, employed to detect relative changes in spectral regions associated with TGase-mediated modification under identical experimental conditions, rather than to provide absolute or calibrated measures of crosslinking. All interpretations have been reframed to emphasize pattern recognition and comparative spectral trends, consistent with established applications of NMR for monitoring enzymatic modification of proteins. See line 363,364.

Comment 12: 12. In Section 3.1.2, there is an inconsistency: the manuscript first states that markers (1.9–4.1 ppm) can be used for quantification, and subsequently claims that quantification is not possible.
Response 12: We thank the reviewer for pointing out this inconsistency. In the revised manuscript, statements suggesting that the 1.9–4.1 ppm resonances can be used for quantitative assessment have been removed. The text now clearly states that these resonances serve solely as qualitative indicators of TGase-mediated modification and provide comparative insight into relative molecular mobility and structural changes among plant protein systems under identical treatment conditions. We have explicitly acknowledged that direct quantification is not possible due to spectral overlap, signal congestion, and the lack of calibration, linearity, and absolute standards. See line 369, 378.

Comment 13: 13. In section 3.1.2, activities of 10, 20, and 50 U/g are discussed, whereas Figure 2 presents data only for 10 U/g.
Response 13: We appreciate the reviewer’s comment. Figure 2 has been revised and is now titled “¹H NMR spectra of plant-based protein isolates after treatment with TGase” to clarify that it represents the spectral profiles of the proteins following enzymatic treatment. Although spectra were recorded at 10, 20, and 50 U/g TGase, the 10 U/g data are presented as representative examples, as all enzyme levels produced similar spectral patterns. This approach reduces redundancy while effectively illustrating protein-specific structural changes induced by TGase.

Comment 14: 14. In Section 3.2.1, TGase activity is inconsistently reported: the text refers to 6.25 U/g (2.5% TGase powder), while Figure 6 denotes Low (2.5 U/g) and High (12.5 U/g).
Response 14: Corrected. See Fig. 6.

Comment 15: 15. The manuscript states "frequency-dependent transitions from solid-like to liquid-like behavior"; however, tan δ values ​​remain well below 1 (≈0.08–0.20) across the measured range, indicating predominantly solid-like or gel-like behavior rather than a transition towards liquid-like behavior.
Response 15: We agree that the tan δ values (≈0.08–0.20) indicate predominantly solid-like or gel-like behavior across the tested frequency range. The manuscript has been revised to clarify that the protein bioinks exhibit frequency-dependent viscoelastic behavior within a solid-like regime, without implying a full transition to liquid-like behavior. See line 534.

Comment 16: 16. In Section 3.2.2, the discussion should be logically reorganized (introduction, heating trend, mechanistic explanation, cooling trend, and comparison between proteins and TGase). The respective contributions of TGase-induced crosslinking and thermal aggregation should be clearly distinguished. Furthermore, the discussion of tan δ trends should be internally consistent, explicitly addressing frequency dependence and specifying the relevant frequency ranges (e.g., 1–10 Hz versus 50–100 Hz).
Response 16: The discussion in Section 3.2.2 has been reorganized to follow a logical sequence: introduction, heating trend, mechanistic explanation, cooling trend, and comparison among different proteins and TGase levels. The contributions of TGase-mediated crosslinking and thermal aggregation have been clearly distinguished. Specifically, the rise in G′ during heating (40–60 °C) is attributed to thermal unfolding and aggregation of protein networks, whereas the further enhancement in network cohesion during cooling reflects cold-set gelation and TGase-mediated crosslinks. See line 576, 589, 595.

Comment 17: 17. In the title or caption of Figure 6, it is stated that "line styles indicate plant protein type," whereas the figure actually uses colors and symbols.
Response 17:  The figure 6 caption has been corrected to accurately reflect the presentation in the figure. See Fig. 6.

Comment 18: 18. Please justify why shear-thinning behavior and printability are inferred based on oscillatory frequency sweep tests rather than steady shear or extrusion-relevant measurements.
Response 18:  Oscillatory frequency sweep tests were used to assess shear-thinning behavior as they provide insight into the viscoelastic response and network recovery of protein bioinks under small deformations, which is relevant for predicting structural stability post-extrusion. While steady-shear or extrusion tests directly measure flow under large deformations, oscillatory measurements allow comparison of gel strength, elasticity, and viscosity changes across formulations under controlled conditions. The results, combined with observed extrusion performance during 3D printing trials, confirm that the bioinks exhibit appropriate shear-thinning behavior for smooth extrusion and shape fidelity. See line 534.

Comment 19: 19. In Section 3.3, the manuscript states that pea protein (PP) exhibits the lowest response, likely due to limited accessibility, while FTIR results simultaneously suggest that PP undergoes the highest β-sheet reorganization.
Response 19: The lower TNBS response of PP reflects limited accessibility of lysine residues in its native conformation, which constrains ε-(γ-glutamyl)lysine bond formation detectable by the TNBS assay. In contrast, FTIR measures secondary structure rearrangements, including β-sheet formation, which can occur through non-covalent interactions and is not necessarily proportional to covalent crosslinking. Therefore, the high β-sheet reorganization observed in PP by FTIR indicates extensive structural rearrangements during TGase treatment, even if covalent crosslinking at accessible lysine residues remains limited. We have revised the manuscript to clarify this distinction, emphasizing that TNBS reflects reactive primary amine accessibility, whereas FTIR captures broader secondary structural modifications.

Comment 20: 20. In the SEM analysis, images at 50× and 100× magnification do not sufficiently support claims regarding protective or network structures.

Response 20: We acknowledge that SEM images at 50× and 100× magnification primarily provide an overview of the sample surface and are insufficient to resolve fine protein network structures or protective matrices. Higher-magnification imaging (e.g., 500×–2000×) would be required to directly visualize detailed fibrillar networks formed by TGase-induced crosslinking. In the manuscript, we have revised the discussion to clarify that the SEM images at the presented magnifications are intended to show general surface morphology and gross structural differences between samples, rather than detailed network architecture. We also emphasize that the claims regarding network formation are supported by complementary analyses, including TNBS, NMR, and SR-FTIR data, rather than solely by SEM images. See line 684, 705.
Comment 21: 21. The Conclusions section should be entirely rewritten, as it currently overestimates the role of TGase activity without adequately considering the contribution of formulation additives and carrier proteins present in the commercial enzyme preparation.

Response 21: The conclusions section has been fully revised to provide a more balanced interpretation of the results. See line 714.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript addresses a timely and relevant topic in the field of improving the 3D printability of plant-based proteins. The wide range of analyses performed, combining techniques from different areas of knowledge, adds interest and depth to the study.

Nevertheless, although the experimental design is appropriate and the results are generally coherent, the manuscript presents conceptual and methodological weaknesses that should be addressed before the work can be considered.

1. Improvement of the discussion

It is recommended to improve the discussion of the results presented in the manuscript, particularly with respect to the relationships between β-sheet content and the rheological, textural, and printing properties. In its current form, this relationship is presented in a rather general manner, without sufficiently discussing its possible limitations or its dependence on protein type and formulation.

2. Insufficiently defined printability analysis

The analysis of printability represents one of the weakest aspects of the manuscript. Although images of printed structures and dimensional accuracy values (Table 3) are provided, it is not clearly described how printability was evaluated, nor which quantitative criteria were applied beyond visual inspection and final dimensions.

In particular:

The printing parameters (pressure, extrusion speed, nozzle diameter, layer height) are not clearly specified.

The exact time point at which dimensions were measured (immediately after printing, after resting, or after incubation) is not described.

The relationship between the data presented in Table 3 and the concept of “good printability” is not sufficiently justified.

It is recommended to:

Explicitly describe the procedure used to evaluate printability.

Include scale bars in the printed images instead of relying solely on graphical annotations.

Clarify the limitations of the analysis, acknowledging that printability is a multifactorial property.

3. Rheological results

It would be appropriate to clarify whether an amplitude sweep was performed to determine the linear viscoelastic region (LVR). In the current version, it is not experimentally demonstrated that the deformation applied during oscillatory measurements lies within the linear viscoelastic regime for all formulations.

In addition, it would be advisable to discuss shear recovery or thixotropic behavior, as these parameters are particularly relevant for extrusion-based processes. The discussion also does not sufficiently address the optimal balance between flowability and structural rigidity.

It is recommended to clarify that oscillatory rheology provides indirect indications of printability, but does not by itself constitute a complete evaluation of this property.

4. Delimitation of the application scope

Since the study does not evaluate biological aspects, it is recommended to explicitly delimit the proposed applications as exclusively food-related, avoiding any ambiguity regarding potential biomedical uses.

5. Quality and clarity of the figures

Figures 1 and 2: improve resolution, font size, and axis clarity; figure captions should be more descriptive and self-contained.

Figure 8: improve legibility, as part of the text is not clearly visible.

6. Additional comments

Review terminological consistency and avoid expressions that may lead to extrapolations beyond the food science context.

Consider including a brief discussion on the reproducibility of the printing process.

Author Response

Comment 1: The manuscript addresses a timely and relevant topic in the field of improving the 3D printability of plant-based proteins. The wide range of analyses performed, combining techniques from different areas of knowledge, adds interest and depth to the study.

Nevertheless, although the experimental design is appropriate and the results are generally coherent, the manuscript presents conceptual and methodological weaknesses that should be addressed before the work can be considered.

Response 1:  We thank the reviewer for the constructive comments. Please note that all revisions addressing your suggestions have been carefully incorporated into the manuscript, and the revised sections have been highlighted in blue.

Comment 2: 1. Improvement of the discussion

It is recommended to improve the discussion of the results presented in the manuscript, particularly with respect to the relationships between β-sheet content and the rheological, textural, and printing properties. In its current form, this relationship is presented in a rather general manner, without sufficiently discussing its possible limitations or its dependence on protein type and formulation.

Response 2: The discussion has been revised to provide a more detailed analysis of the relationship between β-sheet content and the observed rheological, textural, and 3D printing properties. Specifically, we now clarify that the impact of β-sheet enrichment on gel strength, viscoelasticity, and shape fidelity is protein-dependent, reflecting differences in amino acid composition, tertiary structure, and susceptibility to TGase-mediated crosslinking. We also acknowledge the potential limitations of correlating β-sheet content directly with functional performance, noting that formulation additives and carrier proteins can modulate network formation and rheological behavior. These revisions provide a more nuanced interpretation of how structural changes contribute to the macroscopic properties of plant-based bioinks. See line 424, 429, 439.

Comment 3: 2. Insufficiently defined printability analysis

The analysis of printability represents one of the weakest aspects of the manuscript. Although images of printed structures and dimensional accuracy values (Table 3) are provided, it is not clearly described how printability was evaluated, nor which quantitative criteria were applied beyond visual inspection and final dimensions.

In particular:

The printing parameters (pressure, extrusion speed, nozzle diameter, layer height) are not clearly specified.

The exact time point at which dimensions were measured (immediately after printing, after resting, or after incubation) is not described.

The relationship between the data presented in Table 3 and the concept of “good printability” is not sufficiently justified.

It is recommended to:

Explicitly describe the procedure used to evaluate printability.

Include scale bars in the printed images instead of relying solely on graphical annotations.

Clarify the limitations of the analysis, acknowledging that printability is a multifactorial property.

Response 3: We explicitly report all key parameters used for 3D printing, including extrusion pressure, printing speed, nozzle diameter, layer height, and environmental conditions (25 ± 2 °C, 45 ± 5% RH). Dimensions were measured at two critical time points: immediately after printing to assess initial shape fidelity, and after incubation at 55 °C for 2 h to allow TGase-mediated gelation, ensuring structural stability. This information is now included in Section 2.8. and 3.3.1.

Printability was quantified by calculating dimensional deviations from CAD specifications (length, width, height), and by comparing shape fidelity across multiple geometries (cubic, cylindrical, heart; 20 × 20 × 6 mm).

The printed objects were not placed directly on a scale bar; instead, we provide a dimension guide alongside the images to indicate the sizes of the constructs.

The rationale linking these measurements to “good printability” is now explicitly described, noting that lower deviations indicate higher fidelity and structural stability.

Finally, We added a statement acknowledging that printability is multifactorial, depending on rheological properties, protein formulation, extrusion parameters, and post-printing gelation. Therefore, dimensional fidelity provides a useful but partial assessment of overall printability.

Comment 4: 3. Rheological results

It would be appropriate to clarify whether an amplitude sweep was performed to determine the linear viscoelastic region (LVR). In the current version, it is not experimentally demonstrated that the deformation applied during oscillatory measurements lies within the linear viscoelastic regime for all formulations.

In addition, it would be advisable to discuss shear recovery or thixotropic behavior, as these parameters are particularly relevant for extrusion-based processes. The discussion also does not sufficiently address the optimal balance between flowability and structural rigidity.

It is recommended to clarify that oscillatory rheology provides indirect indications of printability, but does not by itself constitute a complete evaluation of this property.

Response 4: The amplitude sweep was performed during preliminary experiments to determine the linear viscoelastic region for all formulations, ensuring that the deformations applied in the oscillatory frequency and temperature sweeps remained within this regime. These details have now been clarified in the revised Methods section. While shear recovery and thixotropic behavior were not measured in this study, we acknowledge their importance for extrusion-based printing and have added a discussion highlighting that such properties, along with the balance between flowability and structural rigidity, are critical for printability. See line 538-541.

Comment 5: 4. Delimitation of the application scope

Since the study does not evaluate biological aspects, it is recommended to explicitly delimit the proposed applications as exclusively food-related, avoiding any ambiguity regarding potential biomedical uses.

Response 5: We have clarified in the manuscript that the applications of TGase-crosslinked plant protein bioinks are exclusively food-related.

Comment 6: 5. Quality and clarity of the figures

Figures 1 and 2: improve resolution, font size, and axis clarity; figure captions should be more descriptive and self-contained.

Figure 8: improve legibility, as part of the text is not clearly visible.

Response 6: The resolution has been improved, and the captions have been revised. See Fig. 1-2, 8.

Comment 7: 6. Additional comments

Review terminological consistency and avoid expressions that may lead to extrapolations beyond the food science context.

Consider including a brief discussion on the reproducibility of the printing process.

Response 7: Terminological consistency has been reviewed throughout the manuscript, and any expressions that could suggest applications beyond the food science context have been clarified or removed. See line 73, 91.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I have carefully reviewed the revised version of the review manuscript and confirm that all previously raised comments have been satisfactorily addressed.
 The revisions have improved the clarity, structure, and critical depth of the literature analysis, strengthening the overall coherence of the review.
The manuscript now provides a balanced, well-supported, and up-to-date overview of the topic.