Dietary Fishmeal Replacement by Methanol-Extracted Cottonseed Meal with Amino Acid Supplementation for Juvenile Cobia Rachycentron canadum
, Guangde Wu 1,†, Delbert M. Gatlin III 3, Kunpeng Lan 1, Yun Wang 1, Chuanpeng Zhou 1 and Zhenhua Ma 1,2
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe present manuscript summarizes a promising strategy to replace fishmeal in a carnivorous fish species, which is valuable and worthy to work in aquaculture. The use of herbal-based protein as a fishmeal replacement has been successfully evaluated for a range of fish species and cottonseed meal could represent a practical option for partial replacement of fishmeal, the quantity and price of which are a constraint to the expansion of aquaculture. Back to the manuscript, it is well-designed, well-written, and easy to understand and read. Therefore, I highly recommend to publish the manuscript in JMSE after considering the following comments:
Please use a term control or full fishmeal diet instead of “reference” in the whole manuscript.
L19: with an average initial…
The authors should perform broken line tests for the candidate parameters to find the optimal level of fishmeal replacement by CSM.
L86: Is there any research that has determined the nutritional requirements of cobia and the authors are aware of how much fishmeal is needed in ideal conditions in order to replace it in cobia fry?
L96: Citation please.
It is needed to clarify in the M&M section that how free gossypol was measured.
L346: What is the maximum allowed or tolerable amount of gossypol as an anti-nutritional factor in bony fish diet?
Based on Table 2, why the authors used only two replications? It is commonly to test at least three replications.
L107: What was the internal reference detection columns? Why the authors did not measure tryptophan?
How IPF was measured from the fish?
It is strange for me that the final weight in the full fishmeal diet was higher than other treatments, but the WG was lower than CSM25 and CSM50. Please clarify.
How can IPF ratio help the authors to justify the obtain results? Is there any relationship with other analyzed parameters like growth performance or carcass quality or hepatic enzymes?
L388: Why? What is the physiological pathway(s)? What is the reason(s) for the alternation of hepatic enzymes in response to replaced fishmeal diets?
Author Response
Comments and Suggestions for Authors
The present manuscript summarizes a promising strategy to replace fishmeal in a carnivorous fish species, which is valuable and worthy to work in aquaculture. The use of herbal-based protein as a fishmeal replacement has been successfully evaluated for a range of fish species and cottonseed meal could represent a practical option for partial replacement of fishmeal, the quantity and price of which are a constraint to the expansion of aquaculture. Back to the manuscript, it is well-designed, well-written, and easy to understand and read. Therefore, I highly recommend to publish the manuscript in JMSE after considering the following comments:
Response: We sincerely appreciate the encouragements from Reviewer.
Please use a term control or full fishmeal diet instead of “reference” in the whole manuscript.
Response: As Reviewer suggested, the term “Control” is used.
L19: with an average initial…
Response: “an” has been added before “average” (L19).
The authors should perform broken line tests for the candidate parameters to find the optimal level of fishmeal replacement by CSM.
Response: Following Reviewer’s suggestion, second-order polynomial regression based on weight gain (WG), feed efficiency (FE), and protein efficiency ratio (PER) were performed to estimated the optimal level of fishmeal replacement by CSM (Figure. 1), the result and conclusion have been revised accordingly (L32).
L86: Is there any research that has determined the nutritional requirements of cobia and the authors are aware of how much fishmeal is needed in ideal conditions in order to replace it in cobia fry?
Response: There are some research estimated the nutritional requirements, e.g., protein, amino acids , and lipid (Chou et al., 2001; Wang et al., 2005; Zhou et al., 2004, 2006, 2007; Lunger et al., 2006, 2007a, b, c; Fraser and Davies, 2009; Salze et al., 2011, 2012; Ren et al., 2014; Chi et al., 2020). Nevertheless, till now, most of the related research were on juvenile fish, and fish fry (with smaller size) was much less studied. Craig et al. (2006) used cobia fry with an average initial weight of 7.4 g, and fed them with diets containing two levels of crude protein (CP; 40 and 50%) and three levels of lipid (6, 12 and 18%). They found that weight gain of experimental fish was not significantly affected by protein or lipid levels and ranged from 1099% in fish fed the diet containing 40% CP/18% lipid to 1305% in fish fed the diet containing 50% CP/12% lipid. Feed efficiency ratio values, visceral somatic and hepatosomatic indices were significantly affected by protein and/or lipid. Muscle and liver lipid were impacted by dietary lipid. Muscle protein was significantly impacted by dietary protein levels, while liver protein was affected by both main effects. Dietary protein and lipid had no impact on muscle ash. Till now, how much fishmeal is needed in ideal conditions in order to replace it in cobia fry is not clear.
L96: Citation please.
Response: A citation has been added. (L100)
It is needed to clarify in the M&M section that how free gossypol was measured.
Response: The method for measuring gossypol has been supplemented (L128-132).
L346: What is the maximum allowed or tolerable amount of gossypol as an anti-nutritional factor in bony fish diet?
Response: There is a wide variation in gossypol tolerance among different fish species. For example, dietary incorporation of regular CSM in replacement of soybean meal at a level of 100% (647 mg gossypol/kg diet) did not impair survival of common carp Cyprinus carpio (Wang et al., 2014), and replacement of FM with regular CSM up to 100% (9160 mg gossypol/kg diet) did not impair survival of juvenile tilapia Oreochromis sp. (Mbahinzireki et al., 2000). Dietary gossypol concentration up to 900 mg/kg diet produced no reduction in the growth of channel catfish (Dorsa et al., 1982) Alam et al.(2018) found that a gossypol level of 3466 mg/kg diet did not significantly affect the survival of juvenile southern flounder Paralichthys lethostigma. Anderson et al. (2016) was able to replace 100% of FM protein with low-gossypol CSM protein supplemented with lysine, without negatively affecting the growth performance of juvenile black sea bass Centropristis striata.
Even within the same species, various tolerance and toxic levels of free gossypol have been reported. For example, in channel catfish. Dorsa et al. (1982) demonstrated that juvenile channel catfish could tolerate a diet containing free gossypol from gossypolacetic acid or CSM at approximately 900 mg/kg without affecting growth. Growth was reduced in the fish fed a diet containing free gossypol CSM (FGCSM) at 1,137 mg/kg. However, Barros et al. (2000, 2002) reported that growth was reduced, but total cell count, red blood cell count, hematocrit, and hemoglobin remained unchanged in juvenile channel catfish fed a diet containing FGCSM at 671 mg. A diet containing FGCSM 336 mg/kg did not affect growth but increased glycogen concentration in the liver. However, Yildirim et al. (2003) examined the growth response of juvenile channel catfish fed chemically defined diets and found that an level of 300 mg/kg depressed growth.
Hence, the tolerance of gossypol of fish seems to be species-specific and related to the nutritional formulation and nutrients composition of diets.
Alam, M.S.; Watanabe, W.O.; Carroll, P.M.; Gabel, J.E.; Corum, M.A.; Seaton, P.; Wedegaertner, T.C.; Rathore, K.S.; Dowd, M.K. Evaluation of genetically-improved (glandless) and genetically-modified low-gossypol cottonseed meal as alternative protein sources in the diet of juvenile southern flounder Paralichthys lethostigma reared in a recirculating aquaculture system. Aquaculture 2018, 489, 36–45. doi.org/10.1016/j.aquaculture.2018.02.006.
Barros, M. M., C. Lim, and P. H. Klesius. 2002. Effect of soybean meal replacement by cottonseed meal and iron supplementation on growth, immune response, and disease resistance of channel catfish (Ictalurus punctatus) to Edwardsiella ictaluri challenge. Aquaculture 207:263-279.
Barros, M. M., C. Lim, J. J. Evans, and P. H. Klesius. 2000. Effect of iron supplementation to cottonseed meal diets on the growth performance of channel catfish, Ictalurus punctatus. Journal of Applied Aquaculture 10(1):65-86.
Dorsa, W. J., H. R. Robinette, E. H. Robinson, and W. E. Poe. 1982. Effects of dietary cottonseed meal and gossypol on growth of young channel catfish. Transactions of the American Fisheries Society 111:651–655.
Dorsa, W.J.; Robinette, H.R.; Robinson, E.H.; Poe, W.E. Effects of dietary cottonseed meal and gossypol on growth of young channel catfish. Trans. Am. Fish. Soc. 1982, 111, 651–655.
Mbahinzireki, G.B.; Dabrowski, K.; Lee, K.J.; El-Saidy, D.; Wisner, E.R. Growth, feed utilization and body composition of tilapia (Oreochromis sp.) fed with cottonseed meal-based diets in a recirculating system. Aquac. Nutr. 2000, 7, 189–200. doi.org/10.1046/j.1365-2095.2001.00172.x
Wang, X.F.; Li, X.Q.; Leng, X.J.; Shan, L.L.; Zhao, J.X.; Wang, Y.T. Effects of dietary cottonseed meal level on the growth, hematological indices, liver and gonad histology of juvenile common carp (Cyprinus carpio). Aquaculture 2014, 428–429, 79–87. doi.org/10.1016/j.aquaculture.2014.02.040
Yildirim, M., C. Lim, P. J. Wan, and P. H. Klesius. 2003. Growth performance and immune response of channel catfish (Ictalurus punctatus) fed diets containing graded levels of gossypol-acetic acid. Aquaculture 219: 751-768.
Based on Table 2, why the authors used only two replications? It is commonly to test at least three replications.
Response: We agree with Reviewer’s opinion that it is commonly to test at least three replications. We used only two replications because the amino acids measurement is quite expensive, also the values of the two measurements were similar, so two replications night be acceptable, so it was only tested twice.
L107: What was the internal reference detection columns? Why the authors did not measure tryptophan?
Response: The internal standards are norvaline and norleucine. The tryptophan was not measured because all of the other amino acids can be measured by acidic hydrolysis, while the determination of tryptophan requires an additional alkaline hydrolysis of the sample, so measure the content of tryptophan will double the cost. Also, according to previous study, the tryptophan content in cottonseed meal is close to that of fishmeal, based on these considerations the present study didn’t test it.
How IPF was measured from the fish?
Response: The IPF was measured by separate the fat in the abdominal cavity, weighed, and the ratio to body weight is calculated. As mentioned below, since the Lipid content (in Table 4) can indicate the lipid deposition properties in fish fed either Control or CSM diets, the IPF parameter has been deleted.
It is strange for me that the final weight in the full fishmeal diet was higher than other treatments, but the WG was lower than CSM25 and CSM50. Please clarify.
Response: As shown in Table 2, the average final weight of fish fed the Control diet (the full fishmeal diet) was 120.1±5.7 g, it was lower than that of fish fed CSM25 diet (166.7±5.7 g), and was higher than that of fish fed CSM50 diet (115.4±16.4 g), correspondingly, its WG was lower and higher than that of CSM25 and CSM50, respectively.
How can IPF ratio help the authors to justify the obtain results? Is there any relationship with other analyzed parameters like growth performance or carcass quality or hepatic enzymes?
Response: As Reviewer asked, we checked the manuscript, and found that the IPF does not help too much to justify the obtained results including carcass quality or hepatic enzymes. Fed fish diet with a higher level of lipid may led to a higher IPF, in this case, the ratio of muscle to body weigh may lower. But, it does not mean that a higher IPF would impact the quality of muscle, for example the level of muscle ash. We measured IPF because we want to know whether feeding CSM diet would lead to more fat being deposited in the fish. Since the Lipid content can shown the lipid deposition properties (Table 4), the IPF was deleted in the revision.
After L388: Why? What is the physiological pathway(s)? What is the reason(s) for the alternation of hepatic enzymes in response to replaced fishmeal diets?
Response: Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are two of the liver enzymes. High levels of ALT or AST may be a sign of liver is suffering toxins or undergoing an abnormal protein metabolism. In the present study, serum ALT and AST activities of fish fed diets with 50% or greater CSM were lower than those of fish fed the Control and CSM25 diets, indicated that the gossypol in these diets did not induce a toxic effect on the liver or cause protein metabolism disorders. Nevertheless, the effects of gossypol on hepatic enzymes, e.g., the physiological pathway(s) need further study.
Reviewer 2 Report
Comments and Suggestions for AuthorsPlease see attached file
Comments for author File: 
Comments.pdf
Author Response
Comments on article entitled Dietary fishmeal replacement by methanol-extracted cottonseed meal with amino acid supplementation for juvenile Cobia Rachycentron canadum
Materials and methods
Line 171: Livers, viscera and intestinal fat? Do you mean tissue instead of fat?
Response: Sorry for any confusion. This sentence has been revised into “Livers, viscera, and fat on intestine were collected......” (L202).
Line 172-173: Samples were prepared as described previously [34]? What kinds of samples were prepared? Such as those indicated in the line 171? But this does not coincide with the reference.
Response: We are very sorry for the mistake in endnote cross-referencing. The citation should be [32]. The sampling methods for blood samples, intestinal samples, and liver samples were rewritten (L205-215).
- Wang, J.; Lan, K.P.; Wu, G.D.; Wang, Y.; Zhou, C.P.; Lin, H.Z.; Ma, Z.H. Effect of dietary carbohydrate level on growth, feed utilization, energy retention, body composition, and digestive and metabolic enzymes activities of juvenile cobia, Rachycentron canadum. Aquacul. Rep. 2022, 25, 101211. doi.org/10.1016/j.aqrep.2022.101211
Line 171-190: These paragraphs need clarification. Please clarify: 1) which enzyme in which tissue, 2) how did/did not, you prepare the tissue for the measurement of enzymes activity, 3) it is mentioned that enzymes activities were determined with commercial kits but it is not mentioned if protein levels have been determined in the samples and how. This is very important step for the results obtained for enzymes activity results and interpretations, 4) the references mentioned do not coincide/related with the methods and measures parameters indicated in this study.
Response: Sorry for any confusion, the manuscript has been revised as reviewer suggested: 1)the alanine aminotransferase (ALT) and aspartate aminotransferase (AST) enzymes were from blood samples; the lipase, amylase, and trypsin enzymes were from intestinal samples; the antioxidant enzymes (SOD and GSH-Px) and MDA samples were from liver (L216-234). 2) The methods of sampling for tissues, and preparation o f blood sample were provided in the manuscript (L205-215). 3) The protein contents in the samples were quantified by the Bradford (1976) method using a total protein quantification kit (Jiancheng, Nanjing, China) with bovine serum albumin as the standard (L235-237). 4) again, we are sorry for the mistake in the endnote cross-referencing. We firstly compose the manuscript in Word version, after converting to the format required by the journal, there are some changes in the citation of references. The reference cited has been corrected.
Statistical analysis
Lines 205-213: It is not mentioned which tests have been used for normality (normal
distribution) and homoscedasticity (homogeneity of variance) of the data. Explain/give the reason (s) why you have not be used post hoc tests but orthogonal polynomial contrasts tests.
Response: Thanks for this valuable suggestion. The description of Statistical analysis has been revised. The normality and homoscedasticity were tested using Levene Test in SPSS software (IBM Corp., Armonk, NY, USA). The post hoc tests were also performed using Duncan's multiple range Test if the ANOVA reached a significant level to compare variables between each other, and the superscript letters indicating significance of difference were based on the post hoc tests. The manuscript has been revised accordingly (L269-280).
Results
3.4. Digestive enzymes activities Line 278-283: Results are presented in a peculiar way. They are not understandable by a reader. The expressed Units of Enzymes activities are doubted.
Response: we are sorry for any confusion. As the difference in activities of these enzymes between dietary groups is not significant, to be concise, we did not list these values in a Table, and only described the trend of variation in the text. According to the reviewer's suggestion, these data have shown in Table 6.
The expressed Units of Enzymes activities are doubted.
Response: There are several ways to calculate the enzyme activity, for example, the activity of
glutathione peroxidase (GSH-Px) is defined as “at a certain temperature (e.g., 37 °C), 1 mg of protein (enzyme) that can catalyzing 1 nmol of NADPH per minute is defined as 1 unit of enzyme activity”, it can be calculated based on protein concentration (the unit is nmol/min/mg prot), the mass of samples (nmol/min/g), the cell number (nmol/min/104 cell), and the liquid volume (nmol/min/mL), and the units varied accordingly. In the present study, the activities were measured using a commercial kit, the value is calculated based on the liquid volume, the unit is nmol/mL/min, that is μmol/L, which is used in the present study.
3.5. Antioxidant enzymes activities and serum parameters Line 285-291 and Table 6: Results in the text do not coincide with the results indicated in the table. There is a missing data in the table for some enzymes. The problem with the expression of Units in enzymes activity remains. There is a peculiar presentation of statistical analysis of the data in the table.
Response: We are sorry for our careless on missing data. The values of activities of enzymes measured were listed in the table (Table 6). For the statistical analysis, we firstly tested the normality and homoscedasticity of all values with Levene Test, and then use ANOVA to determine if the inclusion levels of CSM significantly (P < 0.05) affected the observed responses. If the difference is significant, a Duncan's multiple range test was performed to compare the differences between each dietary group. Also, the orthogonal polynomial contrasts was performed to determine if the effect was linear and/or quadratic. The superscript letters on the value in each line indicate whether the difference reached a significant level. In addition, following another Reviewer’s suggestion, we conducted a second-order polynomial regression based on WG, FE, and PER to estimated the optimal level of fishmeal replacement by CSM.
Discussion
Line 309-391: Effort to explain results obtained in this study and connect them with the results with literature have been made only for responses and indicators related to fish growth performance and feeds and feeding indices. Discussion related to enzymes activity and related responses are totally missing.
Response: Following Reviewer’s suggestion, we supplemented discussion related to enzymes activity and related responses (L453-465).
Reviewer 3 Report
Comments and Suggestions for AuthorsAuthors are advised to include an additional paragraph discussing "methanol-extracted cottonseed" in greater detail, providing specific information to justify its selection in the study.
Furthermore, the addition of a table outlining the nutrient composition of "methanol-extracted cottonseed" is recommended.
It is suggested that authors incorporate the methodology for amino acid analysis of various diets into the manuscript.
Regarding Table 3, the term "reference" should be replaced with "initial parameters recorded for the fish."
Author Response
Comments on article entitled Dietary fishmeal replacement by methanol-extracted cottonseed meal with amino acid supplementation for juvenile Cobia Rachycentron canadum
Materials and methods
Line 171: Livers, viscera and intestinal fat? Do you mean tissue instead of fat?
Response: Sorry for any confusion. This sentence has been revised into “Livers, viscera, and fat on intestine were collected......” (L202).
Line 172-173: Samples were prepared as described previously [34]? What kinds of samples were prepared? Such as those indicated in the line 171? But this does not coincide with the reference.
Response: We are very sorry for the mistake in endnote cross-referencing. The citation should be [32]. The sampling methods for blood samples, intestinal samples, and liver samples were rewritten (L205-215).
- Wang, J.; Lan, K.P.; Wu, G.D.; Wang, Y.; Zhou, C.P.; Lin, H.Z.; Ma, Z.H. Effect of dietary carbohydrate level on growth, feed utilization, energy retention, body composition, and digestive and metabolic enzymes activities of juvenile cobia, Rachycentron canadum. Aquacul. Rep. 2022, 25, 101211. doi.org/10.1016/j.aqrep.2022.101211
Line 171-190: These paragraphs need clarification. Please clarify: 1) which enzyme in which tissue, 2) how did/did not, you prepare the tissue for the measurement of enzymes activity, 3) it is mentioned that enzymes activities were determined with commercial kits but it is not mentioned if protein levels have been determined in the samples and how. This is very important step for the results obtained for enzymes activity results and interpretations, 4) the references mentioned do not coincide/related with the methods and measures parameters indicated in this study.
Response: Sorry for any confusion, the manuscript has been revised as reviewer suggested: 1)the alanine aminotransferase (ALT) and aspartate aminotransferase (AST) enzymes were from blood samples; the lipase, amylase, and trypsin enzymes were from intestinal samples; the antioxidant enzymes (SOD and GSH-Px) and MDA samples were from liver (L216-234). 2) The methods of sampling for tissues, and preparation o f blood sample were provided in the manuscript (L205-215). 3) The protein contents in the samples were quantified by the Bradford (1976) method using a total protein quantification kit (Jiancheng, Nanjing, China) with bovine serum albumin as the standard (L235-237). 4) again, we are sorry for the mistake in the endnote cross-referencing. We firstly compose the manuscript in Word version, after converting to the format required by the journal, there are some changes in the citation of references. The reference cited has been corrected.
Statistical analysis
Lines 205-213: It is not mentioned which tests have been used for normality (normal
distribution) and homoscedasticity (homogeneity of variance) of the data. Explain/give the reason (s) why you have not be used post hoc tests but orthogonal polynomial contrasts tests.
Response: Thanks for this valuable suggestion. The description of Statistical analysis has been revised. The normality and homoscedasticity were tested using Levene Test in SPSS software (IBM Corp., Armonk, NY, USA). The post hoc tests were also performed using Duncan's multiple range Test if the ANOVA reached a significant level to compare variables between each other, and the superscript letters indicating significance of difference were based on the post hoc tests. The manuscript has been revised accordingly (L269-280).
Results
3.4. Digestive enzymes activities Line 278-283: Results are presented in a peculiar way. They are not understandable by a reader. The expressed Units of Enzymes activities are doubted.
Response: we are sorry for any confusion. As the difference in activities of these enzymes between dietary groups is not significant, to be concise, we did not list these values in a Table, and only described the trend of variation in the text. According to the reviewer's suggestion, these data have shown in Table 6.
The expressed Units of Enzymes activities are doubted.
Response: There are several ways to calculate the enzyme activity, for example, the activity of
glutathione peroxidase (GSH-Px) is defined as “at a certain temperature (e.g., 37 °C), 1 mg of protein (enzyme) that can catalyzing 1 nmol of NADPH per minute is defined as 1 unit of enzyme activity”, it can be calculated based on protein concentration (the unit is nmol/min/mg prot), the mass of samples (nmol/min/g), the cell number (nmol/min/104 cell), and the liquid volume (nmol/min/mL), and the units varied accordingly. In the present study, the activities were measured using a commercial kit, the value is calculated based on the liquid volume, the unit is nmol/mL/min, that is μmol/L, which is used in the present study.
3.5. Antioxidant enzymes activities and serum parameters Line 285-291 and Table 6: Results in the text do not coincide with the results indicated in the table. There is a missing data in the table for some enzymes. The problem with the expression of Units in enzymes activity remains. There is a peculiar presentation of statistical analysis of the data in the table.
Response: We are sorry for our careless on missing data. The values of activities of enzymes measured were listed in the table (Table 6). For the statistical analysis, we firstly tested the normality and homoscedasticity of all values with Levene Test, and then use ANOVA to determine if the inclusion levels of CSM significantly (P < 0.05) affected the observed responses. If the difference is significant, a Duncan's multiple range test was performed to compare the differences between each dietary group. Also, the orthogonal polynomial contrasts was performed to determine if the effect was linear and/or quadratic. The superscript letters on the value in each line indicate whether the difference reached a significant level. In addition, following another Reviewer’s suggestion, we conducted a second-order polynomial regression based on WG, FE, and PER to estimated the optimal level of fishmeal replacement by CSM.
Discussion
Line 309-391: Effort to explain results obtained in this study and connect them with the results with literature have been made only for responses and indicators related to fish growth performance and feeds and feeding indices. Discussion related to enzymes activity and related responses are totally missing.
Response: Following Reviewer’s suggestion, we supplemented discussion related to enzymes activity and related responses (L453-465).
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors carefully addresses the comments and accepted.
Author Response
Response: We sincerely appreciate the encouragements from Reviewer, and thanks again for all the valuable comments and suggestions.
Author Response File: 
Author Response.docx
