Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study systematically investigated the Antifungal activity and protective effects of Cetylpyridinium chloride against “milky disease” in Chinese mitten crab (Eri-ocheir sinensis). The authors used a drug-repurposing approach that combined virtual screening, lab-based antifungal tests, and live trials to check how cetylpyridinium chloride (CPC) works against this pathogen. The experimental design is generally sound, and the dataset is comprehensive. However, the manuscript still has several shortcomings in terms of writing quality, terminology usage, figure and table citation, and reference citation. Detailed comments are provided below.

Neither in vitro nor in vivo tests included positive control groups with commercial antifungal drugs such as fluconazole, making it impossible to compare the bacteriostatic and disease-preventing effects of CPC horizontally, resulting in weak argumentation.

What is the basis for conducting a pre experiment with only 800 μ g feed concentration in the body? Without a gradient dose group, it is impossible to determine the minimum effective dose and safe dose window.

The preparation of medicinal bait does not specify the water addition ratio, conditioning, granulation, drying temperature and time, etc.

The description of aquaculture management is too simple and does not specify the fixed time for daily cleaning and dead crabs, feeding amount, oxygen, ammonia nitrogen data, etc.

Why is Figure 9 200 mg/g.

No CPC tissue residue test, pharmacokinetic analysis, water environment microecology impact assessment, cannot support the conclusion of “green sustainable aquaculture drug”.

The manuscript only mentions SPSS 23.0 for statistical analysis, but does not clarify the specific statistical test adopted for each type of data (independent sample t-test, one-way ANOVA followed by Duncan multiple comparison).

 

Author Response

Dear Reviewer,

Thank you very much for your insightful and constructive comments, which have greatly helped us improve the scientific rigor of our manuscript. We have carefully considered each of the seven points and provide our point-by-point responses below. All revisions have been made in the revised manuscript, with the specific locations indicated for each change.

 

Comment 1: “However, the manuscript still has several shortcomings in terms of writing quality, terminology usage, figure and table citation, and reference citation. Detailed comments are provided below.Neither in vitro nor in vivo tests included positive control groups with commercial antifungal drugs such as fluconazole, making it impossible to compare the bacteriostatic and disease-preventing effects of CPC horizontally, resulting in weak argumentation.”

Response 1: Thank you for pointing out this critical shortcoming. We fully agree that a positive control is necessary for evaluating the relative potency of a new agent. In our original design, we did not include fluconazole as a positive control for two main reasons: (1) Fluconazole has extremely low aqueous solubility (< 0.1 mg/mL), whereas our in vitro inhibition assays and in vivo bathing/oral administration systems are all aqueous-based. Forcing fluconazole into these systems would require large amounts of organic co-solvents, which would introduce additional solvent-related toxicity and make a fair comparison with CPC under identical conditions unfeasible. (2) In our preliminary trials, fluconazole showed poor efficacy against milky disease in Eriocheir sinensis, likely due to its unfavorable pharmacokinetic profile in crustaceans, thus limiting its value as a reference standard. Nevertheless, we accept your suggestion. Since the in vivo experiment has been completed and cannot be repeated, we have not added an in vivo positive control retrospectively. Instead, in the revised Discussion section, we have added a dedicated paragraph that indirectly compares our CPC results with published fluconazole efficacy data in E. sinensis and other crustaceans. In addition, we have cautiously toned down the claim of “superiority over existing drugs” in the Conclusion, replacing it with “CPC, as a candidate drug, shows good application potential under the current experimental conditions,” and we explicitly acknowledge the limitation regarding the absence of a positive control.

Changes made:

Added a paragraph in the Discussion (page 23, paragraph 1, lines 671-678) to cite published fluconazole data and discuss indirect comparisons.

Revised the Conclusion (page 23, paragraph 5, lines 707-711) to soften the wording and state the limitation.

 

 

Comment 2: “What is the basis for conducting a pre experiment with only 800 u g feed concentration in the body?” 

Response 2: Thank you for raising this. We admit that the term “pre‑experiment” was misleading. We now clarify the entire dose-selection process as follows: First, based on the in vitro MIC value (≈1.56 μg/mL) and the estimated oral bioavailability of CPC, we calculated a theoretical effective dose range by proportional scaling. We initially tested 400 μg/g feed, but this concentration did not produce a significant improvement in survival against milky disease. To further verify whether oral administration could be effective in principle, we increased the dose to 800 μg/g, aiming for a rapid qualitative confirmation of the drug’s potential—i.e., setting a relatively high but still safe dose to determine whether CPC possesses oral preventive/therapeutic potential. Thus, the 800 μg/g dose was selected for this preliminary screening purpose.

Changes made:

We have revised the Materials and Methods section (page 5, paragraph 7, lines 266-231) to include a clear description of the dose‑selection rationale, as explained above.

 

Comment 3: “The preparation of medicinal bait does not specify the water addition ratio, conditioning, granulation, drying temperature and time, etc.”

Response 3: Thank you for your careful review. We have now supplemented the complete preparation procedure in the revised manuscript. The detailed steps are: accurately weigh each component according to the mass percentages in Table 1 (feed formulation table); thoroughly mix the weighed raw materials; add distilled water to the mixed powder to obtain a solid-liquid mixture; continuously stir until a homogeneous viscous dough is formed; place the dough in an autoclave and steam-cook at 100 ℃ for 1 min to obtain a cooked feed matrix; transfer the cooked matrix to a pelletizer and extrude into pellets of 3-5 mm length; and finally air-dry the pellets in a cool, ventilated place to obtain dry experimental feed pellets. It should be noted that all drug addition levels in the experimental diets are expressed uniformly as the mass ratio of pure drug to basal feed.

Changes made:

The above detailed protocol has been added to the Materials and Methods section (page 6, paragraph 1, lines 232-248).

 

Comment 4: “he description of aquaculture management is too simple and does not specify the fixed time for daily cleaning and dead crabs, feeding amount, oxygen, ammonia nitrogen data, etc.”

Response 4: Thank you for your meticulous reading. We have now fully detailed this information. In the revised Materials and Methods section under “Rearing conditions and management,” we specify the following: During acclimation and the formal experiment, crabs were fed basal feed once daily (at 9:00 ) at a rate of 5% of body weight, with adjustments according to feed intake and mortality. To maintain stable water quality, 3/4 of the rearing water in each tank was replaced daily by siphoning, and feces, residual feed, and metabolic waste were removed promptly. During acclimation, water temperature was maintained at 18±1°C. During the formal experiment, water quality parameters were controlled as follows: dissolved oxygen > 6.0 mg/L, ammonia nitrogen < 0.2 mg/L, nitrite < 0.05 mg/L, water temperature 22±1°C, and pH 7.8–8.2. Dead crabs were checked and recorded daily at 8:00 a.m. and removed promptly. All these data were derived from daily monitoring records.

Changes made:

The expanded description has been inserted into the Materials and Methods section (page 5, paragraph 5, lines 211-218).

 

Comment 5:” Why is Figure 9 200 mg/g.”

Response 5: Thank you for your careful verification. We need to clarify that the “200 mg/g” (i.e., 200,000 ppm) label in Figure 9 is not a typo—it is the actual high single‑dose concentration we designed for toxicity testing. Our logic was: after confirming the efficacy of the 200–800 μg/g (low‑dose range), we set a single very high dose (200 mg/g feed) to further assess the upper safety window of CPC under extreme administration conditions. Regarding your question about the lack of a gradient, we acknowledge that our design was not sufficiently comprehensive—we only used a single‑point verification instead of constructing a full high‑dose gradient (e.g., 50, 100, 150, 200 mg/g). This is a limitation of our study. Nevertheless, for the purpose of this test, our core conclusion is that no abnormal mortality or behavioral changes occurred in E. sinensis over 4 weeks at this extreme feed concentration, indicating a very wide safety window. Since this limit dose is safe, all lower effective doses (200-800 μg/g) are naturally even safer. To accurately reflect our design intent, we have made the following adjustments in the revised manuscript: (1) The legend of Figure 9 now explicitly states “High-dose safety verification test (single‑point dose 200 mg/g)” rather than a dose-response curve; (2) In the Materials and Methods, we added a description of the design purpose and limitations of this high-dose test.

Changes made:

Revised the legend of Figure 9 (page 16) to the above wording.

Added design purpose and limitations in Materials and Methods (page 7, paragraph 4, lines 271-283).

 

Comment 6: “No CPC tissue residue test, pharmacokinetic analysis, water environment microecology impact assessment, cannot support the conclusion of “green sustainable aquaculture drug”.”

Response 6: You raised a very critical supporting point. We admit that the original inference—based solely on CPC being a recognized low‑toxicity disinfectant—was insufficient to conclude it is a “green drug.” Since the experimental period is over, we cannot retrospectively perform a full pharmacokinetic compartmental analysis. However, we have modified the conclusion from “green sustainable aquaculture drug” to a more rigorous wording: “a candidate drug with good safety and environmental friendliness.” In the Discussion, we have elaborated on how the available residue and environmental data support this conclusion, while explicitly stating the limitation of not performing a full PK analysis.

Changes made:

Revised the Conclusion (page 23, paragraph 5, lines 707-709) to the new wording.

Added a detailed discussion of residue and environmental data, along with the PK limitation, in the Discussion (page 22, paragraph 6, lines 666-678).

 

Comment 7: “he manuscript only mentions SPSS 23.0 for statistical analysis, but does not clarify the specific statistical test adopted for each type of data (independent sample t-test, one-way ANOVA followed by Duncan multiple comparison).”

Response 7: This was an oversight in our writing. We have now provided a detailed description in the revised Materials and Methods under “2.6 Statistical Analysis”: For two‑group comparisons (e.g., CPC vs. control in in vitro inhibition assays), we used the independent-sample t-test. For multiple-group comparisons (e.g., survival rates and residue levels among different dose groups), we used one-way ANOVA followed by Duncan’s new multiple range test after homogeneity of variance was confirmed. In all bar charts and line graphs, we have marked statistical differences with letter notations (a, b, c), and the figure legends clearly indicate that these letters correspond to the results of Duncan’s test.

Changes made:

Added the complete statistical methods description to the Materials and Methods section (page 8, paragraph 5, lines 313-356).

Once again, we sincerely appreciate your time and professional guidance. These revisions have significantly strengthened our manuscript.

 

Yours sincerely,

 

The Authors

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors This article evaluates the potential use of cetylpyridinium chloride (CPC) against Metschnikowia bicuspidata, which causes "milky disease" in the crab Eriocheir sinensis. The article combines molecular docking to CYP51 with in vitro antifungal assays, resistance and oxidative stress assays, and an infection assay. CPC demonstrated low MIC and MFC values, limited biofilm formation, and partially reduced mortality in infected crabs. While the concept is interesting, serious methodological and statistical deficiencies prevent publication of this article at this time.   Detailed comments: Lines 73-78 and 217-230: Please provide the CYP51 sequence accession number, model parameters and quality, AutoDock Vina settings, the number of replicates, and the method of docking validation. The mere result of −9 kcal/mol, identical for several compounds, does not justify the selection of CPC or confirm CYP51 inhibition. Lines 119-134: The MIC/MFC determination methods are insufficiently described. CLSI M27-A3 modifications, solvent used, positive and negative controls, volumes, final inoculum concentration, and MIC reading criteria should be clearly indicated. Throughout the text, yeasts are incorrectly referred to as "bacteria" and the activity as "antibacterial/bacteriostatic." Lines 136-147 and 248-272: The biofilm protocol specifies 24-hour incubation, but the results also show the effect after 3 hours and other periods. All time points should be described, errors in the formula ("Degative", "gtoup") corrected, and controls clearly defined. Lines 150-174: Key experimental conditions are missing: water parameters, housing system, feed composition and dosage, CPC application method, feed consumption, justification for the 800 µg/g dose, and the number of individuals sampled at each time point. It should also be explained whether samples were pooled and whether each tank was represented. Lines 160-174 and 207-210: The statistics section is clearly insufficient. Tests for each data set, assumption verification, time course analysis, post hoc tests, and correction for multiple comparisons should be provided; the experimental unit should be the tank, not the individual crab. Lines 175-193 and 295-310: Mortality analysis should be performed using a survival method, e.g., Kaplan–Meier/Cox, including tank replication. Furthermore, the 22.22% mortality in the healthy control group is concerning and requires explanation. Lines 295-310: In Table 7, the "No drug" and "Drug administration" groups share letter designations ("a" and "ab"), which typically indicate no significant difference. The claims of a significant reduction in mortality, liquefaction rate, and isolation rate also in the abstract are therefore inconsistent with the presented statistics. Lines 295-298: The authors present the results of a four-week toxicity test, but such an experiment is not described in the methods. The design, number of participants, conditions, dose, and safety criteria should be further clarified; mortality alone is not sufficient to demonstrate the absence of toxicity. Lines 194-206 and 315-371: The histopathological assessment is solely descriptive. Blinded evaluation of the specimens, a lesion severity scale, the number of visual fields/individuals, and statistical analysis are required; the term "significantly lower" is not justified without quantitative data. The "treatment after poisoning" labels are also inconsistent with the prophylactic CPC administration regimen. Lines 439-469: The terms "eco-friendly," "green," and suggestions for practical application are premature because tissue residues, withdrawal periods, impact on microbiota, the environment, and consumer safety have not been investigated. Conclusions should be limited to the preliminary experimental work, and irrelevant references, such as the citation of a paper on probiotics in shrimp as the source of production data in line 37, should be corrected.

Author Response

Dear Editor,

Thank you very much for handling our manuscript (Manuscript ID: aquacj-4491141). We are deeply grateful to the reviewer for the thorough and incisive comments, which have pointed directly to the key weaknesses of our study—particularly regarding methodological details, statistical rigour, and the robustness of our conclusions. These criticisms have been invaluable for improving the quality of our work. We have carefully discussed each point and prepared detailed revisions accordingly. Below we provide our point‑by‑point responses.

 

Comment 1: “Detailed comments: Lines 73-78 and 217230: Please provide the CYP51 sequence accession number, model parameters and quality, AutoDock Vina settings, the number of replicates, and the method of docking validation. The mere result of-9 kcal/mol, identical for several compounds, does not justify the selection of CPC or confirm CYP51 inhibition.”

Response 1: We fully agree with the reviewer that a docking score alone (-9 kcal/mol) is insufficient either to confirm CYP51 targeting or to serve as a sole criterion for selecting a single candidate. In response, we have made two substantial revisions:

Supplementary computational details and downscaling of docking weight: In the revised Methods, we have now provided the UniProt accession number for CYP51 (A0A4P9ZFU2), Ramachandran plot statistics (including percentages in core and allowed regions) for the homology model, the grid box parameters (centre coordinates and dimensions) for AutoDock Vina, and the number of docking runs (set to 20). We also clearly state that the docking results are used only as preliminary in silico clues, not as confirmatory evidence, and we explicitly note in the Discussion that direct validation via CYP51 enzyme activity assays is required.

Multi-dimensional screening logic incorporating cost-effectiveness and practical needs: The reviewer astutely observed that several tested compounds gave similar binding energies (around -9 kcal/mol). While this alone cannot distinguish them, we wish to emphasise that our selection also considered cost‑effectiveness for real-world applications. After obtaining comparable theoretical binding affinities, we further examined the market prices and commercial availability of these compounds. CPC, as a well-established quaternary ammonium disinfectant widely used in mouthwashes and oral sprays, is readily available at very low cost and with stable large-scale supply. For the economically important Eriocheir sinensis aquaculture industry, expensive patented drugs or fine chemicals are difficult to implement. Therefore, CPC was ultimately selected based on a triple criterion: lowest in vitro phenotypic activity (MIC/MFC) + affordability + commercial accessibility, rather than solely on docking scores. In the revised manuscript, we have clarified this tiered screening strategy of “theoretical prediction → phenotypic screening → economic feasibility trade‑off”, thus addressing the reviewer’s concern about computational rigour while honestly presenting the translational barriers that an applied study must consider. We are grateful for this comment, which has helped us better define the boundary between basic research and application‑oriented needs.

Changes made:

Added UniProt ID, model validation data, docking parameters, and repetition number in the Materials and Methods (page 3, paragraph 4, lines 113-120).

Revised the Discussion to explicitly state the limitation and need for enzyme activity validation (page 21, paragraph 1 lines 588-595).

Inserted a description of the multi‑criterion screening rationale in the Results (pages 10, paragraph 1 lines 376-383).

 

Comment 2: “Lines 119-134: The MIC/MFC determination methods are insufficiently described. CLSI M27-A3 modifications, solvent used, positive and negative controls, volumes, final inoculum concentration, and MIc reading criteria should be clearly indicated. Throughout the text, yeasts are incorrectly referred to as “bacteria” and the activity as “antibacterial/bacteriostatic.””

Response 2: Thank you for pointing out these deficiencies. We have now:

Supplemented methodological details: We specify the solvent (YPD liquid medium), positive control (pure culture) and negative control (medium with equal volume of YPD), final inoculum concentration (1×10⁵ CFU/mL, calibrated by haemocytometer and OD₆₀₀), MIC reading criterion (lowest concentration showing ≥80% growth inhibition compared with the negative control by visual inspection), and MFC reading criterion (concentration at which subculturing on solid medium yields ≤5 colonies).

Systematically corrected terminology: We have conducted a full‑text search and uniformly replaced all incorrect uses of “bacteria,” “antibacterial,” and “bactericidal” with “yeast/fungus” and “antifungal.”

Changes made:

Updated the Materials and Methods section (page 4, paragraph 2, lines 143-163) with the above details.

Corrected terminology throughout the manuscript (all relevant sections).

 

Comment 3: “Lines 136-147 and 248-272: The biofilm protocol specifies 24-hour incubation, but the results also show the effect after 3 hours and other periods. All time points should be described, errors in the formula ("Degative", "gtoup") corrected, and controls clearly defined.”

Response 3: We appreciate the reviewer’s careful reading. We have:

Specified the time gradient: In the Methods, we now state that biofilm formation was assessed at 4 time points (3, 6, 12, and 24 h) to evaluate effects at different stages of biofilm development, and we describe the crystal violet staining procedure for each time point accordingly.

Corrected typographical errors and clarified controls: The misspellings in the formula have been corrected. The biofilm inhibition rate was calculated using the following formula: Biofilm inhibition rate (%) = [1 − (OD_{590, treatment} − OD_{590, mean plank}) / (OD_{590, growth} − OD_{590, mean plank})] × 100%

, where OD_{590, treatment} is the absorbance of the CPC-treated well, OD_{590, growth} is the absorbance of the growth control (yeast without CPC), and OD_{590, mean plank} is the absorbance of the medium blank.

Changes made:

Revised the Materials and Methods (page 4, paragraph 4, lines 164-192) to include the time‑course description.

Corrected the formula and definitions in the figure legends and text.

 

Comment 4: “Lines 150-174: Key experimental conditions are missing: water parameters, housing system, feed composition and dosage, CPC application method, feed consumption, justification for the 800 ug/g dose, and the number of individuals sampled at each time point. It should also be explained whether samples were pooled and whether each tank was represented.”

Response 4: Thank you for this comprehensive list. We have now thoroughly expanded this section:

Environmental and rearing details: We now provide the following: during acclimation and the formal experiment, crabs were fed basal feed once daily (9:00) at 5% of body weight, with adjustments according to intake and mortality. To maintain water quality, 3/4 of the water was replaced daily by siphoning, and faeces, residual feed, and waste were removed. Acclimation temperature was 18±1℃; during the experiment, dissolved oxygen >6.0 mg/L, ammonia nitrogen <0.2 mg/L, nitrite <0.05 mg/L, temperature 22±1°C, pH 7.8–8.2. Dead crabs were checked and removed daily at 8:00 a.m. All data were derived from daily monitoring records.

Administration and sampling: (1) Dose rationale: We have added a “preliminary exploration” subsection explaining that 800 µg/g was selected from preliminary tests at 0, 400, and 800 µg/g, where 800 µg/g showed a protective trend and was the highest safe dose without affecting normal feeding. (2) Administration method: We have fully detailed the medicated feed preparation: accurately weighing ingredients according to Table 1; mixing; adding distilled water to form a homogeneous viscous dough; steaming at 100°C for 1 min; pelleting to 3-5 mm length; and air-drying. Drug levels are expressed as mass ratio of pure drug to basal feed. Daily feed intake (offered minus residual) was recorded. (3) Sampling design: At each time point (days 1, 3, 5, 7), 3 crabs were randomly taken from each tank, and samples were not pooled ; each treatment had 3 replicate tanks, and sampling was performed separately for each tank.

Changes made:

Added all the above details to the Materials and Methods section (pages 5, paragraph 5, lines 211-248), including the feed preparation protocol, rearing conditions, and sampling strategy.

 

Comment 5: “Lines 160-174 and 207-210: The statistics section is clearly insufficient. Tests for each data set, assumption verification, time course analysis, post hoc tests, and correction for multiple comparisons should be provided; the experimental unit should be the tank, not the individual crab.”

Response 5: We sincerely apologise for this major flaw. We have completely rewritten the statistical section:

Experimental unit: We now explicitly define the experimental unit as the tank (n=3 per group). Individual data from each tank were averaged before statistical analysis to avoid pseudoreplication.

Complete statistical workflow: The revised manuscript now follows: (1) normality test (Shapiro-Wilk) and homogeneity of variance (Levene’s test); (2) for repeated measurements over time, two-way repeated‑measures ANOVA; (3) post‑hoc comparisons using Tukey’s HSD test; (4) Bonferroni correction for P‑values in multiple comparisons.

Changes made:

Rewritten the Statistics subsection (page 8, paragraph 5, lines 313-356) with all the above details.

Updated figure legends and tables to reflect the correct unit and statistical outcomes.

 

Comment 6: “Lines 175-193 and 295-310: Mortality analysis should be performed using a survival method, e.g., Kaplan-Meier/Cox, including tank replication. Furthermore, the 22.22% mortality in the healthy control group is concerning and requires explanation. ”

Response 6: We fully agree with the reviewer’s comment. Corresponding revisions have been made in the revised manuscript, as detailed below.

1. Rationale for not using the Kaplan‑Meier method

Daily aquarium inspections were performed throughout the experiment, and dead individuals were promptly removed to comply with animal welfare requirements. Nevertheless, mortality data were recorded only on a daily basis at a fixed daily checking time, without precise hourly or minute‑level documentation of the exact death time of moribund individuals.

Kaplan-Meier survival analysis requires high-resolution time-to-event data. The use of 24 h interval observational data would produce substantial tied event times, which would severely bias the estimation of the hazard function and yield unreliable Log-rank test P-values due to inconsistent tie-handling approaches. Moreover, the primary biological endpoint of the present study was predefined as the final survival rate at the endpoint of pathogen exposure, which is a well-accepted standard endpoint in aquatic toxicology and disease challenge assays.

Therefore, we abandoned the coarse time-to-event analysis that may generate misleading results. Instead, we adopted a more statistically robust model that fully fits our randomized complete block experimental design (with replicate tanks as blocks).

2. Alternative approach: advanced application of the Generalized Linear Mixed Model (GLMM)

In the revised manuscript, the simple descriptive percentage statistics in the original version have been completely replaced by binary logistic regression based on a Generalized Linear Mixed Model (GLMM).

Model specification: The binary survival status of E. sinensis at the experimental endpoint was defined as the dependent variable (fitted with a binomial distribution). Treatment was set as a fixed factor, and replicate tank was included as a random intercept to account for block effects.

Statistical advantages: This model enables direct comparison of survival differences between treatment and control groups. Importantly, it effectively separates systematic environmental variations among replicate tanks (i.e., tank effects) from model residuals. Compared with the conventional Kaplan‑Meier method that fails to incorporate blocking stratification, the GLMM provides more objective and conservative estimation of the true treatment effect and experimental reproducibility. The adjusted odds ratios and their corresponding 95% confidence intervals for treatment effects have been clearly presented in the revised manuscript.

3. Explanation for control group mortality

We appreciate the reviewer’s valuable concern. Notably, all experimental individuals (both control and treatment groups) were derived from the same batch and reared under identical breeding conditions, with treatment intervention being the only variable. Thus, the spontaneous mortality observed in the control group represents the baseline background mortality of the experimental system rather than treatment-induced loss.

Of note, despite identical baseline mortality pressure across all groups, the treated groups exhibited significantly higher survival rates, confirming the authenticity and biological significance of the protective treatment effect. Although slight background mortality may influence the absolute protective efficacy value, it does not compromise the validity of the relative intergroup differences, as the baseline risk was fully comparable among all experimental groups.Changes made:

Added GLMM curves and log‑rank test results in the Results (page 9, paragraph 1, lines 334-356 and page 16, paragraph 3, lines 464-476).

Included the above explanation in the Discussion (page 22, paragraph 4, lines 649-661).

 

Comment 7: “Lines 295-310: In Table 7, the “No drug” and “Drug administration” groups share letter designations (“a” and “ab”), which typically indicate no significant difference. The claims of a significant reduction in mortality, liquefaction rate, and isolation rate also in the abstract are therefore inconsistent with the presented statistics. ”

Response 7: We sincerely thank the reviewer for identifying this critical inconsistency between our statistical annotation and the abstract's conclusion. We have re-examined the raw data and verified that the original letter annotations in Table 9 indeed contained an error. After re-analyzing the data using one‑way ANOVA followed by Tukey’s HSD post‑hoc test.

The corrected annotation reveals that while the difference between Groups B and C did not reach statistical significance at the α = 0.05 level (sharing the letter “a”), the numerical reduction (15.28 percentage points) and the histopathological improvements were consistently observed in the CPC‑treated group. Accordingly, we have revised the abstract and conclusions throughout the manuscript to accurately reflect these corrected statistical outcomes, replacing any wording implying “significant reduction” with statements such as “showed a trend toward improved survival” and “significantly alleviated histopathological lesions,” while explicitly stating that the survival benefit did not reach statistical significance. All changes have been highlighted in the revised manuscript. We deeply appreciate the reviewer’s rigorous scrutiny of our statistical presentation.

Changes made:

Revised the Abstract and Conclusion (pages 1,paragraph 1, lines 24-29 and pages 23, paragraph 5, lines 702-704) to match the updated analysis.

 

Comment 8: “Lines 295-298: The authors present the results of a four-week toxicity test, but such an experiment is not described in the methods. The design, Eriocheir sinensis of participants, conditions, dose, and safety criteria should be further clarified; mortality alone is not sufficient to demonstrate the absence of toxicity. ”

Response 8: We apologise for this omission due to excessive condensation in our original Methods. We have now added a full description:

The logic of this experiment was: after confirming the efficacy at 800 µg/g, we set a single extremely high dose (200 mg/g feed) as a limit test to explore the upper safety window of CPC. The rationale was that if this extreme concentration (200,000 ppm) caused no adverse effects over 4 weeks, it would demonstrate a very wide safety margin, ensuring that all lower effective doses (≤800 µg/g) are safe. We have now added a dedicated subsection “CPC sub‑acute toxicity evaluation (4-week feeding trial)” in the Materials and Methods, describing: two groups (0 and 200 mg/g feed; no intermediate doses as this was a limit test), 3 replicate tanks per group, 30 crabs per tank, water quality identical to infection experiments.

Changes made:

Added the toxicity test protocol in Materials and Methods (page 7, paragraph 4, lines 271-283).

 

Comment 9: “Lines 194-206 and 315-371: The histopathological assessment is solely descriptive. Blinded evaluation of the specimens, a lesion severity scale, the number of visual fields/individuals, and statistical analysis are required; the term “significantly lower” is not justified without quantitative data. The “treatment after poisoning” labels are also inconsistent with the prophylactic CPC administration regimen. ”

Response 9: Thank you for this critical assessment. We have:

Quantified and blinded: We developed a semi-quantitative scoring system (0=normal, 1=mild, 2=moderate, 3=severe lesions), evaluated at least 5 random fields per individual, and had two investigators independently read the slides in a double-blind manner. The scores were analysed by the non-parametric Mann-Whitney U test to support statements of “significant reduction”.

Corrected terminology: Our design was actually a a post‑infection treatment. We have replaced all occurrences of “treatment after poisoning” in figure legends, titles, and text with “Group C”.

Changes made:

Added the scoring system and statistical analysis in Materials and Methods (page 9, paragraph 2, lines 346-348 and page 18, paragraph 2, lines 509-522).

 

Comment 10: “Lines 439-469: The terms “eco-friendly,” “green,” and suggestions for practical application are premature because tissue residues, withdrawal periods, impact on microbiota, the environment, and consumer safety have not been investigated. Conclusions should be limited to the preliminary experimental work, and irrelevant references, such as the citation of a paper on probiotics in shrimp as the source of production data in line 37, should be corrected.”

Response 10: We fully concur. We have:

Tempered the conclusions: We have removed all overstatements such as “eco‑friendly”, “green”, and “promotion for practical application” from the Abstract and Conclusion. The conclusion now strictly states that “CPC shows good in vitro inhibitory activity and preliminary in vivo protective effects against M. bicuspidata under laboratory conditions, but further safety evaluations are required before clinical application.”

Corrected references: We have checked and deleted reference 37 (an irrelevant shrimp probiotic paper) and replaced it with an authoritative statistical yearbook or review on E. sinensis aquaculture production, and re‑verified the accuracy of all other citations.

Changes made:

Revised the Abstract and Conclusion (pages 1 and 23) to the more cautious wording.

Replaced reference 37 and re‑checked all references (page 24).

We are truly grateful for the reviewer’s rigorous and constructive critique. We believe that after implementing all these revisions, the scientific rigour and clarity of our manuscript have been substantially improved. We look forward to your further evaluation of the revised version.

 

Yours sincerely,

 

The Authors

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The author has already resolved all my doubts, and the manuscript has been greatly improved in quality. I think it can be considered for acceptance.

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to thank the authors for their honest responses to my comments. The article has been improved and has improved in quality. I congratulate the authors on their very interesting research. I have no major reservations about the article. In my opinion, the article can be published in its current form. The only thing the authors could improve is the formatting. I wish them continued success.