Review Reports - The Effect of Calsporin<sup>®</sup> (<i>Bacillus subtilis C-3102</i>) on Laying Performance, Follicular Development, and Microorganisms of Breeder Geese

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study addresses a relevant and timely topic, as the poultry industry seeks viable antibiotic alternatives to promote green and healthy production. The experiment is generally well-designed, and the findings provide a valuable contribution to the field. However, several points require improvement before the manuscript can be considered for publication as below;

The abstract and introduction use the term "green breeding" and "green and healthy development" without a precise operational definition within the context of the study. While the link to reducing antibiotic use is implied, a clearer definition would strengthen the paper's premise. For instance, does it solely mean antibiotic-free, or does it also encompass welfare, environmental impact, and feed sustainability?
L 44-47: The introduction states that Bacillus subtilis (BS) can "germinate rapidly" in the intestine. While true, the extent and location of germination are critical for efficacy and can be highly variable. It would benefit from mentioning the complexities affecting spore germination, such as gut transit time and bile salt concentrations, which can differ between avian species.
L 55-60: The introduction mentions CAL can "reduce the feed conversion ratio" and has shown effects in swine and poultry. It would be more impactful to cite specific quantitative improvements (e.g., a specific percentage reduction in FCR in broilers) from the referenced studies [6, 7] to set a stronger precedent for the current experiment.
L 69-71: You correctly identify a research gap regarding CAL's effect on breeder geese. However, the introduction could be strengthened by briefly hypothesizing why geese might respond differently than chickens or swine, perhaps due to their reproductive patterns or unique digestive physiology such as longer gut, different fiber digestion capacity or etc.
L 87-93: The experimental design section states that 2 separate sheds were assigned to each treatment as "independent production groups". This raises a critical statistical issue. The true experimental unit for statistical analysis should be the shed (n=2 per treatment), not the individual goose. Treating individual geese as replicates when they share a common environment (shed) leads to pseudoreplication, which can artificially lower the p-value and lead to incorrect conclusions of significance. The statistical analysis should account for this. In additon, there are notable differences in the number of geese per treatment group (e.g., Con group has ~1733 females, 60 ppm group has ~1464 females). The authors should explain the reason for this unequal distribution and confirm if stocking density, which can be a confounding variable affecting stress and performance, was kept consistent across all sheds.
Table 1: The provided diet composition table lacks information on the basal diet's composition before the CAL premix was added. It lists "Vitamin-mineral Premix" at 5%, which is an unusually high inclusion rate for a standard premix. Typically, premixes are included at 0.25-1%. This may be a typographical error or represent a custom formulation that requires further explanation.
L 111: Serum was collected from only two female geese per shed for hormone analysis. This sample size (n=4 per treatment) is extremely small and likely underpowered to detect statistically significant differences, especially given the high variability in reproductive hormone levels. This limitation should be explicitly stated in the discussion when concluding there was "no significant difference".
L 124-127: Follicles are classified as "hierarchical follicles (>2 cm)" and "large yellow follicles (10 mm < size ≤ 2 cm)" While functional, this classification could be more detailed. Goose ovaries often have a more complex hierarchy. The paper would be more precise if it followed a standard avian follicle classification system (e.g., F1, F2, F3 for hierarchical follicles) to allow for better comparison with other literature.
L 146-149: The paper states that one-way ANOVA was used for comparisons. As noted in above comment #5, if the shed is the experimental unit, a nested ANOVA or a mixed-effects model with 'shed' as a random effect would be more appropriate to avoid pseudoreplication.
L 165-168: The 60 ppm dose significantly decreased egg weight and egg shape index (Table 3). This is a very unusual and important finding. The discussion offers no hypothesis for this negative effect. Could an intermediate dose disrupt the gut microbiome in a non-beneficial way that a higher, more competitive dose does not? This warrants significant exploration.
L 186-190: The number of hierarchical follicles was significantly increased in the 100 ppm group, but the total number of large follicles (hierarchical + large yellow) was not significantly different. This suggests that CAL does not increase the total pool of developing follicles but rather accelerates their transition from the "large yellow" stage into the final "hierarchical" stage, which is a crucial insight that could be emphasized more.
L 210-212: The results state that CAL supplementation "reduced the diversity of fecal microorganisms" (Figure 2-A), which is often interpreted as a negative outcome. However, in the context of probiotic action, a reduction in diversity can be positive if it results from the suppression of pathogenic or non-beneficial taxa and the dominance of beneficial ones. This nuance is critical and should be explained in the discussion.
L 220-224: The LEfSe analysis (Figure 3-C) identifies Bacillaceae-Bacillus and Lactococcus as significantly more abundant in the 100 ppm group.This is strong evidence. However, the discussion should include other less common microbial groups to give a fuller view of the microbial changes.
L 308-311: The discussion posits that BS may act indirectly on the HPG axis because no significant changes in hormones were detected. This is a reasonable hypothesis, but given the small sample size for hormone analysis (n=4), the lack of significance could be a Type II error (a false negative). The authors should acknowledge this limitation when making this conclusion.
L 332-335: The discussion suggests that Short-Chain Fatty Acids (SCFAs) produced by Firmicutes could supply energy to ovarian tissues. This is an interesting idea linking gut health to reproduction. To make it stronger, the authors could measure SCFA levels in the cecum or blood to directly support this.
L 341-344: You highlight the enrichment of Lactococcus for its bacteriocin production, which is valuable. To improve the discussion, please consider adding that Lactococcus also produces lactic acid, which lowers the gut pH. This acidic environment further inhibits pathogens and supports the growth of beneficial bacteria, thus playing a key role in the observed shifts in the gut microbial ecosystem.
L 347-349: The conclusion recommends a dose of 100 ppm CAL. This is well-supported by the egg production and follicle data. However, the paper should also explicitly advise against the 60 ppm dose, given its significant negative impact on egg quality, making the practical advice more complete and cautious.

Comments on the Quality of English Language

The English quality of the research paper is generally good and conveys the scientific findings clearly. However, there are some areas where the language could be refined for better flow, precision, and adherence to academic writing conventions.

Author Response

We sincerely appreciate the reviewer’s suggestions. In response, we have revised the manuscript accordingly, as detailed below, if the modification does not meet the requirements, please point out again:

The abstract and introduction use the term "green breeding" and "green and healthy development" without a precise operational definition within the context of the study. While the link to reducing antibiotic use is implied, a clearer definition would strengthen the paper's premise. For instance, does it solely mean antibiotic-free, or does it also encompass welfare, environmental impact, and feed sustainability?

A: We have made the required revisions. For details, please refer to lines 34-37 and 77-81 of the manuscript.

L 44-47: The introduction states that Bacillus subtilis (BS) can "germinate rapidly" in the intestine. While true, the extent and location of germination are critical for efficacy and can be highly variable. It would benefit from mentioning the complexities affecting spore germination, such as gut transit time and bile salt concentrations, which can differ between avian species.

A：We have revised and replaced the references as required. For details, please refer to the manuscript. Lines 46-53.

L 55-60: The introduction mentions CAL can "reduce the feed conversion ratio" and has shown effects in swine and poultry. It would be more impactful to cite specific quantitative improvements (e.g., a specific percentage reduction in FCR in broilers) from the referenced studies [6, 7] to set a stronger precedent for the current experiment.

A：We have revised：In breeding experiments on swine and poultry, CAL has improved growth performance (e.g., reducing broiler feed conversion ratio by 1.46%-6.92% [6] and 5.52% overall [7]) and enhanced immunity. Lines 64-67

L 69-71: You correctly identify a research gap regarding CAL's effect on breeder geese. However, the introduction could be strengthened by briefly hypothesizing why geese might respond differently than chickens or swine, perhaps due to their reproductive patterns or unique digestive physiology such as longer gut, different fiber digestion capacity or etc.

A:We have revised：This phenomenon may stem from geese's distinctive physiological adaptations, particularly their elongated intestinal tract that enhances nutrient retention time, specialized cecal fermentation systems capable of processing high-fiber diets, and seasonally-regulated reproductive cycles that drive unique metabolic requirements during egg production phases. Lines 94-99

5：L 87-93: The experimental design section states that 2 separate sheds were assigned to each treatment as "independent production groups". This raises a critical statistical issue. The true experimental unit for statistical analysis should be the shed (n=2 per treatment), not the individual goose. Treating individual geese as replicates when they share a common environment (shed) leads to pseudoreplication, which can artificially lower the p-value and lead to incorrect conclusions of significance. The statistical analysis should account for this. In additon, there are notable differences in the number of geese per treatment group (e.g., Con group has ~1733 females, 60 ppm group has ~1464 females). The authors should explain the reason for this unequal distribution and confirm if stocking density, which can be a confounding variable affecting stress and performance, was kept consistent across all sheds.

A: Thank you for your valuable advice! We are sorry for missing important grouping information in this section. It is necessary to clarify the design of the experimental units in this study: Each treatment group is set up with 2 greenhouses, and each greenhouse contains 4 independent small greenhouses. In fact, the small greenhouses are used as the real experimental units, so each treatment group has 8 replicates (n=8)：Three dietary treatments (0, 60, and 100 ppm of CAL) were established. Each treatment was assigned 2 separate sheds, with 4 independent sub-sheds per shed. The sub-sheds served as the true experimental units, resulting in 8 replicates per treatment (n=8). Stocking density was strictly maintained at 1.5 m²/geese across all sub-sheds to control confounding effects Lines 114-118

Table 1: The provided diet composition table lacks information on the basal diet's composition before the CAL premix was added. It lists "Vitamin-mineral Premix" at 5%, which is an unusually high inclusion rate for a standard premix. Typically, premixes are included at 0.25-1%. This may be a typographical error or represent a custom formulation that requires further explanation.

A:Thank you for your feedback. The composition information shown in Table 1 is the composition information of the base diet itself before adding CAL premix - since a trace amount of CAL does not change the core component structure of the base diet, it can be directly used to represent the composition of the base diet before addition. The addition amount of premix is a customized design based on the unique egg-laying requirements of the target goose, rather than a universal commercial formula.

L 111: Serum was collected from only two female geese per shed for hormone analysis. This sample size (n=4 per treatment) is extremely small and likely underpowered to detect statistically significant differences, especially given the high variability in reproductive hormone levels. This limitation should be explicitly stated in the discussion when concluding there was "no significant difference".

A：We apologize for missing important grouping information in this section. It is necessary to clarify the design of the experimental units in this study: Each treatment group is set up with 2 greenhouses, and each greenhouse contains 4 independent small greenhouses. In fact, the small greenhouses are used as the real experimental units, so each treatment group has 8 replicates (n=8).

L 124-127: Follicles are classified as "hierarchical follicles (>2 cm)" and "large yellow follicles (10 mm < size ≤ 2 cm)" While functional, this classification could be more detailed. Goose ovaries often have a more complex hierarchy. The paper would be more precise if it followed a standard avian follicle classification system (e.g., F1, F2, F3 for hierarchical follicles) to allow for better comparison with other literature.

A:Regarding follicular classification, you proposed in question 11 to explore the influence of CAL on the transition of follicles from the "large yellow stage" to the "hierarchical stage". We also believe that this result is more meaningful. Therefore, we plan to use our previous classification standards. If you think it is necessary to distinguish according to F1, F2, and F3, we fully respect your opinion and continue to make modifications

8.L 146-149: The paper states that one-way ANOVA was used for comparisons. As noted in above comment #5, if the shed is the experimental unit, a nested ANOVA or a mixed-effects model with 'shed' as a random effect would be more appropriate to avoid pseudoreplication.

9.L 165-168: The 60 ppm dose significantly decreased egg weight and egg shape index (Table 3). This is a very unusual and important finding. The discussion offers no hypothesis for this negative effect. Could an intermediate dose disrupt the gut microbiome in a non-beneficial way that a higher, more competitive dose does not? This warrants significant exploration.

A:Thank you for your suggestion. We have added specific details in the discussion section: lines 292-300

11.L 186-190: The number of hierarchical follicles was significantly increased in the 100 ppm group, but the total number of large follicles (hierarchical + large yellow) was not significantly different. This suggests that CAL does not increase the total pool of developing follicles but rather accelerates their transition from the "large yellow" stage into the final "hierarchical" stage, which is a crucial insight that could be emphasized more.

A: Thank you for your suggestion. We have added specific details in the discussion section: lines338-344

12.L 210-212: The results state that CAL supplementation "reduced the diversity of fecal microorganisms" (Figure 2-A), which is often interpreted as a negative outcome. However, in the context of probiotic action, a reduction in diversity can be positive if it results from the suppression of pathogenic or non-beneficial taxa and the dominance of beneficial ones. This nuance is critical and should be explained in the discussion.

A: We have added this part of discussion as required: lines 363-371

13.L 220-224: The LEfSe analysis (Figure 3-C) identifies Bacillaceae-Bacillus and Lactococcus as significantly more abundant in the 100 ppm group.This is strong evidence. However, the discussion should include other less common microbial groups to give a fuller view of the microbial changes.

A: We have added this part of the discussion as required: lines 382-391, and supplemented the references: 31,32.

14.L 308-311: The discussion posits that BS may act indirectly on the HPG axis because no significant changes in hormones were detected. This is a reasonable hypothesis, but given the small sample size for hormone analysis (n=4), the lack of significance could be a Type II error (a false negative). The authors should acknowledge this limitation when making this conclusion.

A: The point you raised is highly constructive and precisely points out the key limitations of the research! We fully agree. We are sorry for missing important grouping information in this section. It is necessary to clarify the design of the experimental unit in this study: Each treatment group was set up with 2 greenhouses, each of which contained 4 independent small greenhouses. In fact, the small greenhouses were used as the real experimental units, so each treatment group had 8 replicates (n=8). In the subsequent research, we will prioritize the sample size of hormone detection and further rigorously verify the indirect regulation hypothesis of BS on the HPG axis to enhance the reliability of the conclusion.

15.L 332-335: The discussion suggests that Short-Chain Fatty Acids (SCFAs) produced by Firmicutes could supply energy to ovarian tissues. This is an interesting idea linking gut health to reproduction. To make it stronger, the authors could measure SCFA levels in the cecum or blood to directly support this.

A：We greatly appreciate your valuable suggestion of measuring SCFA levels in the cecum or blood to directly support the proposed gut-reproduction link, which provides a critical direction for deepening this research; unfortunately, due to several unavoidable objective constraints, we were unable to conduct this detection in the current study—specifically, the initial experimental design focused on the association between gut microbiota structure and reproductive phenotypes, with sample collection and preservation protocols not optimized for SCFA analysis (as SCFAs are highly degradable and require specific cryogenic anaerobic preservation conditions), making existing samples unsuitable for such detection, coupled with limitations in dedicated detection equipment and the experimental funding budget that prevented the inclusion of SCFA testing in this batch of research, and all biological samples have been fully processed and preserved during the collection phase, making it difficult to supplement corresponding samples for retrospective testing in the short term. We have clearly acknowledged this limitation in the discussion section and attach great importance to this scientific question, and we plan to specifically design an SCFA detection module in subsequent studies, simultaneously collect cecal contents and blood samples, and combine targeted metabolomics technology to verify this hypothesis, thereby further improving the evidence chain for the gut-reproductive axis regulatory mechanism.

16.L 341-344: You highlight the enrichment of Lactococcus for its bacteriocin production, which is valuable. To improve the discussion, please consider adding that Lactococcus also produces lactic acid, which lowers the gut pH. This acidic environment further inhibits pathogens and supports the growth of beneficial bacteria, thus playing a key role in the observed shifts in the gut microbial ecosystem.

A: We have made the necessary modifications as required. Lines 379-384

17.L 347-349: The conclusion recommends a dose of 100 ppm CAL. This is well-supported by the egg production and follicle data. However, the paper should also explicitly advise against the 60 ppm dose, given its significant negative impact on egg quality, making the practical advice more complete and cautious.

A: We have revised the conclusion as required. Lines 397-402

Best Wishes,

Huiying Wang

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

General Comments
Dear authors, the article addresses a highly relevant topic, especially involving animal species for which few studies are available. However, several corrections and clarifications are essential to improve the manuscript. Below are specific observations:

Introduction
The introduction is well developed, presenting a comprehensive overview of the concepts related to the topic. However, from line 66 onward, the authors begin to discuss geese, but the explanation of how probiotics may benefit the reproductive aspects of these birds could be expanded. The current description is too general, merely stating that “The balanced state of the microecology is closely related to nutrient absorption, immune regulation, and reproductive function of breeder geese.” This point would benefit from a more detailed justification.

Materials and Methods
This section requires adjustments and clarifications:
Details about the concentration of the commercial product are missing (e.g., CFU/g?).
How was the probiotic incorporated into the feed to maintain 100% of the diet composition? Was an inert ingredient used? How did the authors create increasing inclusion levels without altering the proportions of other ingredients? Were the diets isoproteic and isoenergetic?
Was the diet formulated for a specific production phase?
Although the experimental design is somewhat implied, the text should explicitly state the type of design used.
What was the experimental unit? The absence of a clearly stated design raises doubts about the number and nature of replicates. Please specify the design, the number of replicates, and how each replicate was composed.
Were there records of temperature, ventilation, photoperiod, or relative humidity? These parameters are important since they directly influence the studied variables.
For serum collection, specify whether blood samples were collected in dry tubes, EDTA, or heparin tubes. In addition, mention the analytical method used for the kits (e.g., ELISA, RIA) and the assay sensitivity.
For fecal samples and 16S rRNA sequencing, indicate the number of analyzed samples and whether they were composite samples (by house) or individual ones. Clearly defining the experimental unit helps to eliminate such uncertainties.

Results
The authors describe an increase in egg production in Table 2; however, no statistical evidence is provided to support this statement. It is important to note that differences among treatments must always be interpreted within the framework of statistical analysis.
At line 172, the authors report a markedly different number of incubatable eggs for the 60 ppm group. This discrepancy requires justification, as such variation in data volume may affect statistical reliability.

Discussion and conclusion without considerations.

Author Response

Q1:The introduction is well developed, presenting a comprehensive overview of the concepts related to the topic. However, from line 66 onward, the authors begin to discuss geese, but the explanation of how probiotics may benefit the reproductive aspects of these birds could be expanded. The current description is too general, merely stating that “The balanced state of the microecology is closely related to nutrient absorption, immune regulation, and reproductive function of breeder geese.” This point would benefit from a more detailed justification.

A: Thank you very much for your valuable suggestions. We have completed the modification as per your request: Lines 72-87

Q2: Details about the concentration of the commercial product are missing (e.g., CFU/g?).

A: The content of BS C-3102 in the product is ≥1.0×10¹⁰ CFU/g. Lines 108-109

Q3：How was the probiotic incorporated into the feed to maintain 100% of the diet composition? Was an inert ingredient used? How did the authors create increasing inclusion levels without altering the proportions of other ingredients? Were the diets isoproteic and isoenergetic?Was the diet formulated for a specific production phase?

A3:Thank you for your suggestion：CAL was added in powder form with calcium carbonate as an inert carrier. To maintain the overall diet composition, we used a proportional substitution strategy: the carrier calcium carbonate was adjusted according to CAL inclusion levels at 0, 60, and 100 ppm, while the proportions of all other dietary ingredients remained unchanged. This approach ensured that only the CAL dosage varied, without altering the relative proportions of other nutrients. The experimental diets were formulated to be isoenergetic and isoproteic and specifically designed for the egg-laying phase of breeder geese, adhering to their nutritional requirements. The diet was specifically formulated for the egg-laying phase of breeder geese, adhering to their nutritional requirements for this production stage.

Q4:What was the experimental unit? The absence of a clearly stated design raises doubts about the number and nature of replicates. Please specify the design, the number of replicates, and how each replicate was composed.

A4: Thank you for your valuable advice! We apologize for missing important grouping information in this section. It is necessary to clarify the design of the experimental units in this study: Each treatment group is set up with 2 greenhouses, and each greenhouse contains 4 independent small greenhouses. In fact, the small greenhouses are used as the real experimental units, so each treatment group has 8 replicates (n=8)：Three dietary treatments (0, 60, and 100 ppm of CAL) were established. Each treatment was assigned 2 separate sheds, with 4 independent sub-sheds per shed. The sub-sheds served as the true experimental units, resulting in 8 replicates per treatment (n=8). Stocking density was strictly maintained at 1.5 m²/geese across all sub-sheds to control confounding effects Lines 114-118.

Q5: Were there records of temperature, ventilation, photoperiod, or relative humidity? These parameters are important since they directly influence the studied variables.

A5:We have supplemented this part as required：All experimental sheds were equipped with automatic environmental monitoring devices, with temperature (18–25°C) and relative humidity (55–65%) monitored at 30-minute intervals daily, ventilation controlled at 5–8 m³/(h·bird) via frequency-conversion fans (hourly recorded), and photoperiod standardized to 16L:8D (300–500 lx) with programmable LED lights (daily recorded for consistency); all environmental parameters were continuously archived throughout the experiment. Lines 122-128

Q6：For serum collection, specify whether blood samples were collected in dry tubes, EDTA, or heparin tubes. In addition, mention the analytical method used for the kits (e.g., ELISA, RIA) and the assay sensitivity.

A6: We have made the necessary modifications as required： 2.4. Serum Collection

Q7：For fecal samples and 16S rRNA sequencing, indicate the number of analyzed samples and whether they were composite samples (by house) or individual ones. Clearly defining the experimental unit helps to eliminate such uncertainties.

A7：We have made the required modifications in lines 166-169：Composite fecal samples were pooled from 5–8 random spots within each sub-shed, with 1 sample per sub-shed, 8 samples per treatment and 24 samples in total collected during sampling.

Q8：The authors describe an increase in egg production in Table 2; however, no statistical evidence is provided to support this statement. It is important to note that differences among treatments must always be interpreted within the framework of statistical analysis.

A8: We have modified Table 2, lines 201-202

Q9：At line 172, the authors report a markedly different number of incubatable eggs for the 60 ppm group. This discrepancy requires justification, as such variation in data volume may affect statistical reliability.

A9: Thank you very much for pointing out the issue of the difference in the number of incubable eggs in this study. We are deeply aware that the details of egg collection were not clearly stated in advance in the experimental design, which brought uncertainty to the interpretation of the data. We sincerely apologize for this. 60 PPM can hatch CAL group number (14287) and control group (7283), 100 PPM CAL group significant differences (7311), the phenomenon is the result of objective breeding conditions: This group of geese happened to be in the peak period of natural egg production during the experiment, and the egg collection period for this group was relatively longer. However, the goose flocks in the control group and the 100 ppm CAL group were not in the peak egg-laying stage. Meanwhile, the egg collection period was relatively short, which eventually led to deviations in the number of hatchable eggs in each group.

Best Wishes,

Huiying Wang

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

This revised version can be accepted for publication.

Reviewer 2 Report

Comments and Suggestions for Authors

After reading the version submitted by the authors, it was observed that the suggested revisions have been properly incorporated.