Effect of Cadmium and Nickel Exposure on Early Development in Zebrafish (Danio rerio) Embryos

Round 1

Reviewer 1 Report

Dear authors,

This manuscript has interesting information regarding exposure to metals, especially when analysed together and can have implications on the development of aquatic organisms. However, there are some questions that need to be clarified in the methodology, such as the experimental unit, the rationale to use these concentrations, among others. It would be an improvement to see a discussion/ statement about the practical implication of these results (environment, research, fish maintenance regarding water quality, etc).

 

Abstract

Please add a clear aim, and general implication(s) of the study’s results.

 

Introduction

• Line 55: the word “However” does not fit here.
• Line 79-86: Reporting all these studies, please describe the novelty/ improvement/ complement of your study.

 

Material and Methods

• Add the conductivity and pH where the animals live, develop and when they are exposed to the metals.
• Explain the rationale to use these concentrations. You can add references and describe which are the concentrations of these metals in watercourses, public waters…
• Please clarify the experimental unit. The well is the experimental unit, so how many wells you did per group? You said that each well had 2-3 embryos, thus, for each parameter did you do the average of their values or just use one of the animals in each analysis? With this, please clarify the number of experimental units used to study each parameter.
• 3: the 30 randomly selected embryos were one from each well and so you analysed 30 experimental units for each group?
• Line 140: you stated that you counted the number of movements, but in the results and later you described that you record if a movement occurred or not in 60 seconds (line 216-217). Please clarify this here. And here you should also describe the nature of that movement inside the chorion as you describe later. Why didn’t you analyse the number of movements per animal in 60 seconds? That analysis would give you a more accurate profile of the behaviour.
• Line 143-144: for how long did you record the distance after the touch? Until the animal stopped moving? Here you should already describe that you divided the distance into two categories/ ranges and count the number of animals that fit in each. Why didn’t you use the value of the distance travelled by the animals instead of dividing the distances in two ranges? With your analysis, you may lose information.
• Line 145: Please confirm the anaesthetic overdose used to kill zebrafish larvae, as they are resistant to MS222 at this age and the dose reported is very low. The 48 and 72 hpf larvae were fixed in NBF, that could be a confirmatory method of euthanasia, but not 24hpf larvae. How did you handle those larvae after MS222? Did you confirm the death?
• Please describe the analysis of staining quantification that you performed (2.5).
• Line 180: add “number of” spontaneous movements
• You used ANOVA but did you assess data distribution and the homogeneity of variances between groups?
• Line 184: for movement assessment, did you use 30 embryos for both movements? You mentioned 10 embryos for elicited movement. Please see my comment regarding the experimental unit and how many experimental units you have per group. When you refer the number of animals per each parameter, is the same as the experimental unit (each embryo was from a different well)? Please refer to the sample size calculation you performed.
• For what did you use students’ t-test? I could not identify in the manuscript. And why to use this test for several groups and not use ANOVA as it is already made for comparisons of more than two groups? Maybe for figure 8E, but you described ANOVA in the legend.
• Line 220: you can put the data not shown as supplementary data and put in the manuscript the difference in nickel.
• Line 233-225: “We…distance).”- this is methodology.
• This comment is general to all the manuscript, as you refer this word several times: The description of a p value as a trend is a subjective analysis and should be avoided, using the analysis described there were no differences between groups.
• Line 234-236: “At 48 hpf, exposure to nickel alone at the two highest concentrations decreased total body length (ANOVA, p<0.00001) as compared to exposure to the two lower nickel concentrations.” – please verify as the graph only show that 0.5 mg/L NiCl2 was significantly different than all other NiCl2 concentrations.
• Line 239: “In contrast,” - It is not in contrast to what you refer before, it is also no differences.
• Line 249: Add the information of no differences at 48hpf when using combined exposure regarding eye diameter.
• Line 247-248: “… the combined lower concentrations increased eye area at 72 hpf” - compared with what group? as compared with control it decreases, I think you are comparing with the higher concentrations, please clarify in the text.
• You can joint figure 3 and 4 in one as it is the same parameter.
• Line 257: add the p value of the comparison between the highest concentration of cadmium vs other cadmium concentrations.

 

Discussion

• Line 313: instead of “of toxicants” replace by “these specific toxicants”. The situation you described seems that the chlorpyrifos did nothing to the hatching rate of the ZF but only the nickel, so even if you had an additive effect it would be equal to nickel, as chlorpyrifos was innocuous.
• Line 332: If you analyse how often the embryo moves and not only registered if the embryo moved or not in 60 sec, maybe you could find some statistical alterations. Thus, you should acknowledge this limitation and a ceiling effect on your analysis.
• Line 340: similar to the previous observation, maybe if you use the distance travelled by the animals instead of dividing the distances in two ranges, you could find some alterations, as your analysis may be limited.
• Line 350-351: This result may mean probably a no-delay in muscle fibre development which is in line with fewer problems in hatching, and no alterations in movement. Please discuss that.
• Line 355-361: Do you think that embryos at 48hpf are making a great effort to hatch, but are not able to do it in combined exposure (hatching rate is low), but that effort is leading to an increase in myosin expression? In which way these metals alter the chorion permeability and rigidity?

 

Conclusions

• Line 382-383: “The combination… spontaneous movement.”- We cannot observe that because you did not show all the results of the individual and combined metals exposures regarding elicited or spontaneous movements, please do so.
• Line 387-389: I do not think this last sentence is clear, please read it carefully and re-phrase it.

 

Table 1: It is not described between which groups are the differences marked with a *.

Table 2: Why did you analyse more exposed embryos than controls? Wouldn't this interfere with the statistical analysis of Chi-square?  It would be good to have also values of % in each group, as the number of animals analysed are not the same.

In the figures: Why don’t you use SD as it is more representative of each group variability? The standard error is not the most adequate measure for variability when you want to compare differences between groups (https://academic.oup.com/bja/article/90/4/514/274340).

In the figures’ legend: please put the range of the number of animals used in each group. Was each value of the figure/ statistic an average of 3-4embryos (as 3-4 is the number of embryos you put in one well for exposure, as the well is the experimental unit)? When you refer 3 replicate experiments are you saying you had 3 exposure wells for each group or how many? Please clarify this in the methodology in line with my previous comments regarding the number of experimental units.

Figure 3: Line 646: this p value was already stated before and see my previous comment about trends.

Author Response

Response to Reviewer #1

AbstractPlease add a clear aim, and general implication(s) of the study’s results.

Response:

Information concerning the aim of the study has been added to the beginning of the abstract. The general implications have been clarified at the end of the abstract

 

Introduction

Line 55: the word “However” does not fit here. 

Response:The sentence containing “However” has been removed.

Line 79-86: Reporting all these studies, please describe the novelty/ improvement/ complement of your study. 

Response:We have added information to more clearly identify the fact that we exposed embryos to combinations of nickel and cadmium, which is the improvement our study design

 

Materials and Methods.

Add the conductivity and pH where the animals live, develop and when they are exposed to the metals.

Response: Conductivity was not measured; thus, this information cannot be added to the manuscript.  The pH of the medium in which the embryos are reared and exposed to metals is 7.20; the composition and pH of the medium is detailed in the methods. 

 Explain the rationale to use these concentrations. You can add references and describe which are the concentrations of these metals in watercourses, public waters…

Response:Text has been added to the methods that describes the environmental contamination levels for cadmium and nickel that have been observed in various water sources around the world.

 

Please clarify the experimental unit. The well is the experimental unit, so how many wells you did per group? You said that each well had 2-3 embryos, thus, for each parameter did you do the average of their values or just use one of the animals in each analysis? With this, please clarify the number of experimental units used to study each parameter.

Response: We have included an explanation at the start of the Statistical Analysis section in the Materials and Methods.  Zebrafish embryos were collected on a weekly basis and assigned to control and experimental metal exposures.  Due to the large number of experimental groups used in this study, not all metals and combinations of metals could be assessed in a single week. Thus, for each concentration of cadmium, nickel, or combination of cadmium and nickel, the exposures were repeated a minimum of three times to ensure that a sufficient number of embryos were tested in this study.  Each week we would run (or test) the complete series of concentrations of cadmium, nickel, or the combinations of nickel and cadmium.  Three to four embryos were pooled in one well of the 24-well plate, but only one embryo from each well was used for a evulation of a given parameter.  Thus, we would have tested 30-80 individual embryos for each expeirmental treatment; and each embryo was taken from a different well.

 

3: The 30 randomly selected embryos were one from each well and so you analyzed 30 experimental units for each group?

Response:Yes.  As explained above, that is correct.  We would analyzed 30 or more individual embryos from different experimental units for each control or experimental group.  

 

Line 140: you stated that you counted the number of movements, but in the results and later you described that you record if a movement occurred or not in 60 seconds (line 216-217). Please clarify this here. And here you should also describe the nature of that movement inside the chorion as you describe later. Why didn’t you analyze the number of movements per animal in 60 seconds? That analysis would give you a more accurate profile of the behaviour.

Response: In developing ZFE, spontaneous movements of tail musculature are initiated at approximately 17 hpf, peak at 19 hpf, and continue until hatching [37].  Spontaneous tail movements or “twitches” are attributed to the establishment of motor neural circuits between the spinal cord and developing skeletal muscles [37].  We assessed spontaneous movement of ZFE at 24 hpf in controls and all experimental groups (Table 1).  Data were evaluated by Chi-square analysis.  Because the embryos did not move or moved so infrequently in the 60-second period (i.e., less than 5 times), movements were categorized as no movement observed in 60 seconds or movement observed in 60 seconds.  We explained this in the Results when reported the results. 

 

Line 143-144: for how long did you record the distance after the touch? Until the animal stopped moving? Here you should already describe that you divided the distance into two categories/ ranges and count the number of animals that fit in each. Why didn’t you use the value of the distance travelled by the animals instead of dividing the distances in two ranges? With your analysis, you may lose information.

Response: The elicited movement was assessed by applying a single slight touch of the embryo’s tail to initiate swimming.  The distance swam in response to that one single stimulus was measured, and that measurement was made just once.

 

Line 145: Please confirm the anaesthetic overdose used to kill zebrafish larvae, as they are resistant to MS222 at this age and the dose reported is very low. The 48 and 72 hpf larvae were fixed in NBF, that could be a confirmatory method of euthanasia, but not 24hpf larvae. How did you handle those larvae after MS222? Did you confirm the death?  Response:

Response:  As stated in the methods, all embryos were exposed to MS222, even the 24-hpf embryos.  The concentration of MS222 used was as reported in the cited publish reference.  The embryos were all examined as they were transferred to NBF.  No movement was observed.

 

Please describe the analysis of staining quantification that you performed (2.5).

Response:  We have included this statement, “Fixed embryos were immunostained for myosin heavy chain protein; staining intensity in experimental embryos was expressed as a percent of staining in control embryos.” In the Methods.  We have also included a succinct statement in the Results and Figure Legend 7 to remind that reader that staining intensity in experimental embryos was expressed as a percent of staining in controls.

 

Line 180: add “number of” spontaneous movements

Response:  This phrase was inserted.

 

You used ANOVA but did you assess data distribution and the homogeneity of variances between groups?

Response:  A separate test for homogeneity was not carried out, because no outliers were observed in the data and the variances that were generated in the ANOVAs were not very different. 

 

Line 184: for movement assessment, did you use 30 embryos for both movements? You mentioned 10 embryos for elicited movement. Please see my comment regarding the experimental unit and how many experimental units you have per group. When you refer the number of animals per each parameter, is the same as the experimental unit (each embryo was from a different well)? Please refer to the sample size calculation you performed.

Response:  As described above, 3-4 embryos pooled in a single well and reared/treated in control medium (no metals) or experimental medium with metal(s).  Spontaneous movement was evaluated in approximately 100 individual embryos, each one randomly selected from different wells.  This has been clarified in the methods.  Elicited movement was evaluated in 40-80 individual embryos, each one randomly selected from different wells. This has also been clarified in the methods.  

 

For what did you use students’ t-test? I could not identify in the manuscript. And why to use this test for several groups and not use ANOVA as it is already made for comparisons of more than two groups? Maybe for figure 8E, but you described ANOVA in the legend.

Response:  Comparisons between means for each treatment were made using the multiple comparisons Duncan’s difference procedure or the Student’s t-test and the Bonferroni correction was applied as needed. 

 

Line 220: you can put the data not shown as supplementary data and put in the manuscript the difference in nickel. 

Response:  We appreciate the suggestion to place the data that were not shown into supplementary figures.  However, because the data that were not shown were all not significant, it does not seem useful to show the readers the details of the non-significant data.  Previous journals and editors have suggested that, in general, such data are really not really useful and unnecessarily add to the length of manuscripts.  We would venture to suggest that readers would not examine supplementary data, if such data were not significant.

 

Line 233-225: “We…distance).”- this is methodology.

Response:  We moved the statement “We used Chi-square analysis to assess elicited movement; data were categorized into two groups (0 to 25 mm swim distance and 26 to 50 mm distance)”;to the Methods section.  

 

This comment is general to all the manuscript, as you refer this word several times: The description of a p value as a trend is a subjective analysis and should be avoided, using the analysis described there were no differences between groups.

Response:  The point raised by the reviewer is valid and has been taken into consideration by the authors. The phrase, “although there was a trend for decreased swim distances for each single metal exposure,” has been removed from the text.

 

Line 234-236: “At 48 hpf, exposure to nickel alone at the two highest concentrations decreased total body length (ANOVA, p<0.00001) as compared to exposure to the two lower nickel concentrations.” – please verify as the graph only show that 0.5 mg/L NiCl2 was significantly different than all other NiCl2 concentrations. CHECK THIS ONE

Response:  This original text did not accurately describe the effects of nickel exposure on total body length at 48 hpf.  That portion of text has been edited to accurately reflect the data and presentation of data in Figure 2A.  “At 48 hpf exposure to nickel did not alter body length, whereas at 72 hpf ZFE exposed to the two highest nickel concentrations were significantly shorter than controls and shorter than ZFE exposed to the two lower nickel concentrations.”

 

Line 239: “In contrast,” - It is not in contrast to what you refer before, it is also no differences.

Response:  The phrase, “In contrast,” has been removed. 

 

Line 249: Add the information of no differences at 48h pf when using combined exposure regarding eye diameter.

Response:  A statement now is included to describe that combined exposure to nickel and cadmium did not affect eye diameter in 48-hpf embryos. 

 

Line 247-248: “… the combined lower concentrations increased eye area at 72 hpf” - compared with what group? as compared with control it decreases, I think you are comparing with the higher concentrations, please clarify in the text. 

Response:  The observed effects of combined exposure to nickel and cadmium on eye area at 72 hpf has been clarified in the following text: “ Combined exposure to nickel and cadmium did not alter eye area at 48 hpf.  However, when embryos were exposed to the combination of nickel and cadmium (ANOVA, p=0.000005), the combined higher concentrations decreased eye area (p=0.006) as compared to age-matched controls (Fig 3D). ”     

 

You can join figure 3 and 4 in one as it is the same parameter.

Response:  Figures 3 and 4 have been combined; the respective data for eye diameter and eye are now are presented in Figure 3.  As such, the subsequent four figures have been renumbered.  

 

Line 257: add the p value of the comparison between the highest concentration of cadmium vs other cadmium concentrations.  

Response:  We ask for the Reviewer’s patience.  Line 257 in the original manuscript is in the portion of text that reported data for movement.  However, the exposure to the highest concentration of cadmium did significantly increase yolk sac area in 72-hpf embryos as compare to controls and as compared to all other cadmium concentrations with a p value of 0.004 (vs. other cadmium concentrations).  ‘p = 0.004’ has been inserted into the text.     

 

Discussion

Line 313: instead of “of toxicants” replace by “these specific toxicants”. The situation you described seems that the chlorpyrifos did nothing to the hatching rate of the ZF but only the nickel, so even if you had an additive effect it would be equal to nickel, as chlorpyrifos was innocuous.

Response:  The phrase, “of toxicants” has been replaced by “these specific toxicants”, as suggested.

 

Line 332: If you analyze how often the embryo moves and not only registered if the embryo moved or not in 60 sec, maybe you could find some statistical alterations. Thus, you should acknowledge this limitation and a ceiling effect on your analysis.

Response:  The reviewer has raised a valid point and additional discussion of this point has been added to the text. 

 

Line 340: similar to the previous observation, maybe if you use the distance travelled by the animals instead of dividing the distances in two ranges, you could find some alterations, as your analysis may be limited.

Response:  When movements are elicited in 72-hpf ZFEs, the movements are a reflex escape behavior.  When touched, the ZFE exhibit a short burst of swimming behavior and then stop.  When the data were analyzed using multiple ranges, no additional alterations were observed.  Thus, the analysis that was presented in this study was determined to be the most straightforward analysis to use

 

Line 350-351: This result may mean probably a no-delay in muscle fibre development which is in line with fewer problems in hatching, and no alterations in movement. Please discuss that.

Response:  We have added discussion concerning hatching and muscle development.  Please note that there were subtle but significant differences in spontaneous and elicited movements at some metal exposures.

 

Line 355-351: Do you think that embryos at 48hpf are making a great effort to hatch, but are not able to do it in combined exposure (hatching rate is low), but that effort is leading to an increase in myosin expression?  In which way these metals alter chorion permeability and rigidity?

Response:  The reviewer has proposed an interesting explanation for the increase in myosin expression; that is a possible increase protein synthesis in response to greater physical activity.  Perhaps in a future study actual protein synthesis could be measured.  To the second point, we had explained that nickel might decrease chorionase expression or its activity.

 

Conclusions

Line 382-383: “The combination… spontaneous movement.”- We cannot observe that because you did not show all the results of the individual and combined metals exposures regarding elicited or spontaneous movements, please do so.

Response:  We have presented data for those effects that were statistically significant.

 

Line 387-389: I do not think this last sentence is clear, please read it carefully and re-phrase it.

Response:  We have rephrased the text in the conclusion section so that is more clear.

 

Figures and Tables

Table 1: It is not described between which groups are the differences marked with a *. 

Response:  This information has been added to Table 1.

Table 2: Why did you analyze more exposed embryos than controls? Wouldn't this interfere with the statistical analysis of Chi-square? It would be good to have also values of % in each group, as the number of animals analyzed are not the same.

Response:  We had more exposed embryos than controls.  This does not interfere with the analysis.  The Chi-Square test is quite robust.  It does not require that the sample sizes of the study groups be equal.  The Chi-Square test is valid for groups that are of unequal sample size.  It also is true that the groups may be of equal size as well.  Either situation is valid.  The Chi-square test is described as a distribution-free test.  The Chi-Square test is used to assess 1- or 2-dimensional arrays of data.  Please note that more than two groups can be assessed for expected results and observed results.  Most calculators for Chi-Square will assess up to ten groups.  It is recommended that one does not include more than 20 groups.  Two groups is the minimum.  There are quite a few references that discuss the Chi-Square test; here is one: Preacher, K. J. (2001, April). Calculation for the chi-square test: An interactive calculation tool for chi-square tests of goodness of fit and independence [Computer software]. Available from http://quantpsy.org.

In the figures: Why don’t you use SD as it is more representative of each group variability? The standard error is not the most adequate measure for variability when you want to compare differences between groups (https://academic.oup.com/bja/article/90/4/514/274340).

Response:  It is true that one can report variability using standard deviation (SD) or standard error of the mean (SEM).  SEM is typically used to indicate the accuracy of a sample mean. In other words, the SEM describes how precisely the mean of the sample reflects the actual or “true” mean of the population.  SD is used to indicate the amount of variability in the data set.  Since we were more interested in the means of the populations, we chose to use SEM.  We examined the article that was cited (Misuse of standard error of the mean (sem) when reporting variability of a sample. A critical evaluation of four anaesthesia journals P. Nagele BJA: British Journal of Anaesthesia, Volume 90, Issue 4, April 2003, Pages 514–516). We note that not everyone understands the differences between SD and SEM, as was mentioned in the article.  However, we would point out that there are reasons to use SEM.  We concluded that because we were interested assessing the means of the populations and not just the variability of the groups, we should present SEM.

 

In the figures’ legend: please put the range of the number of animals used in each group. Was each value of the figure/ statistic an average of 3-4embryos (as 3-4 is the number of embryos you put in one well for exposure, as the well is the experimental unit)? When you refer 3 replicate experiments are you saying you had 3 exposure wells for each group or how many? Please clarify this in the methodology in line with my previous comments regarding the number of experimental units.

Response:  At the start of the Statistical Analysis in the Materials and Methods we have explained the expeirmental unit.  Zebrafish embryos were collected on a weekly basis and assigned to control and experimental groups.  Due to the large number of experimental groups used in this study, not all metals and combinations of metals could be assessed in a single week. Thus, for each concentration of cadmium, nickel, or combination of cadmium and nickel, the exposures were repeated a minimum of three times to ensure that a sufficient number of embryos were tested in this study.  Each week we would run (or test) the complete series of concentrations of cadmium, nickel, or the combinations of nickel and cadmium.  Three to four embryos were pooled in one well of the 24-well plate, but only one embryo from each well was used for a evulation of a given parameter.  Thus, we would have tested 30-100 individual embryos for each expeirmental treatment; and each embryo was taken from a different well.

 

Figure 3: Line 646: this p value was already stated before and see my previous comment about trends. 

Response:  This phrase that referred to the trend has been removed. 

Reviewer 2 Report

 

The paper needs improvements.

In introduction: omit the text in lines 56-59 since a human is rather far from zebrafish.

In materials and methods the link in line 116 did not open.  Please, explain carefully how the fish were maintained? You say: “ultrapure water with low concentration of specific ions”. Which ions you had and which concentrations? What was the hardness of this water since increased hardness is known to reduce LD50-values of Ni according to EPA?  Some ions are also known to effect on solubility of Cd²⁺ and Ni²⁺.   

In lines 121-125 you explain Ni and Cd solutions. Are the concentration only for Ni and Cd or are they for these salts meaning NiCl₂  and CdCl₂, which are sold as NiCl₂∙6H₂O or as CdCl₂∙H₂O? How exactly were the solutions done?

Your figures are not good. I would change them to tables where the concentrations would be rows and the 48 hpf and 72 hpf would be the columns. That would allow to compare the change during the time.  For statistical differences I would use letters a, b, c … If both results have a, there is no difference between them and ab would be similar with both a and b. It is also common to show with stars ( ⁕ or ⁕⁕ or ⁕⁕⁕ ) the level of significance.

The other possibility is to join in Fig 1 both A and D as well B and E. Usually a table with numbers is better for a person who has studied science. Your ♯ and especially † are so unusual, that do not use them!  

The journal has the tables and figures  with their captions in text not at the end or not partly in text and partly at the end.

If the major part of caption is the same, you can ask the reader to see the rest  in Table 1 or Figure 2 and it is not necessary to repeat all of it again.  

All the numbers indicated to references should be as [1] and see from instructions for authors (home page of journal) how they should be in reference list.

Author Response

The paper needs improvements.

Response:  The comment is appreciated.  The manuscript has been edited in light of the reviewers’ comments and suggestions, and the manuscript has been greatly improved. 

 

Introduction

In introduction: omit the text in lines 56-59 since a human is rather far from zebrafish.

Response:  The information concerning the similarities in genomes has been removed from the Introduction.

 

In materials and methods the link in line 116 did not open. Please, explain carefully how the fish were maintained? You say: “ultrapure water with low concentration of specific ions”. ADD DETAILS Which ions you had and which concentrations? What was the hardness of this water since increased hardness is known to reduce LD50-values of Ni according to EPA? Some ions are also known to effect on solubility of Cd and Ni.

Response:  The composition of the embryo medium (in mM, 13.7 NaCl, 0.54 KCl, 0.025 Na₂PO₄, 0.44 KH₂PO₄, 1.3 CaCl₂, 1.0 MgSO₄, and 4.3 NaHCO₃ and pH adjusted to 7.20 as per Westerfield [Ref #31]) is now included in the Methods.  The hardness of ultrapure water (Millipore® system) used to make the metal stock solutions and the embryo was not directly measured.  The conductivity of ultrapure water is typically about 0.055 uS/cm at 25C and the resistivity of 18.2 MOhm.  The hardness of the ultrapure water was not determined, but is typically extremely low.  In this case, we do not believe the hardness of the ultrapure water or the total or individual concentration of the ions would have significantly reduced the median lethal dose (LD50) of Ni or Cd. 

 

In lines 121-125 you explain Ni and Cd solutions. Are the concentration only for Ni and Cd  YES or are they for these salts meaning NiCl and CdCl , which are sold as NiCl ·6H O or unless otherwise stated as CdCl ·H O? How exactly were the solutions done?  Response:

The reviewer is correct; the concentration of nickel or cadmium were for the metal itself, not nickel chloride or cadmium chloride.  We have clarified this in the methods in the description of how stock and experimental solutions were prepared and in the text of the legends and labeling of the graphs and tables.  In the methods, we explain that separate stock solutions of cadmium and nickel were prepared from dichlorocadmium hydrate (CdCl₂^.H₂O) and nickel chloride hexahydrate (NiCl₂^.6H₂O).  Stock solutions were prepared at 1.0 g of cadmium or nickel per liter (g/L) using sterile water, and then diluted with fresh embryo medium to prepare treatment solutions.  Effects of nickel alone were examined using embryo medium with 0.1, 0.5, 2.5, and 5 mg/L nickel, and the effects of cadmium alone were using medium with 0.015, 0.15, 0.705 and 1.5 mg/L cadmium.  For combined exposure to nickel and cadmium, ZFE were exposed simultaneously to 0.5 mg/L nickel and 0.15 mg/L cadmium or 2.5 mg/L nickel and 0.7 mg/L cadmium.  In the figures, legends and tables, we no longer refer to nickel chloride or cadmiium chloride, but refer to simply nickel or cadmium.

 

Your figures are not good. I would change them to tables where the concentrations would be rows and the 48 hpf and 72 hpf would be the columns. That would allow to compare the change during the time. For statistical differences I would use letters a, b, c … If both results have a, there is no difference between them and ab would be similar with both a and b. It is also common to show with stars ( ⁕ or ⁕⁕ or ⁕⁕⁕) the level of significance.

Response:  We appreciate the suggestion to present the data in tabular form, but elect to still present the data in a graphs.  We have followed the suggestion of this reviewer and Reviewer #3 and now use more conventional markers of statistical similarities and differences.  In the revised figures when values for experimental treatments and values for the respective age-matched control are not statistically different, the respective values for experimental and control treatment are marked with the letter ‘a’.  Values for experimental treatments marked with a b are significantly different than values for the age-matched control (p<0.05).  In other cases, the letters, c, d, e, and f are used to indicate statistically significant differences between different concentrations of nickel or cadmium, or the different combinations of nickel and cadmium.

 

The other possibility is to join in Fig 1 both A and D as well B and E.  Usually a table with numbers is better for a person who has studied science.  Your ♯ and especially † are so unusual, that do not use them!

Response:  We understand that Reviewer #2 is not in favor of presenting the data as graphs and has recommended that tables be used instead.  It is the case, however, that most readers prefer to see data presented as graphs rather than tables.  The visual representations of numbers are much easier to comprehend than numerical data in tables.  Since the other two reviewers did not offer the same recommendation, we decided to not convert the graphs into tables.  We have followed the recommendation to combine figures 3 and 4 as suggested by reviewer #1.  We feel that it would be too confusing for the readers to join A and D and B and E in Figure 1.  It is quite easy to look across the figure to make the necessary comparisons. 

We agree with the suggestion to use letters (a, b, c…) to indicate similarities and differences in the figures.  We have changed the symbols to letters for the figures as suggested.

 

The journal has the tables and figures with their captions in text not at the end or not partly in text and partly at the end.

Response:  In the revised manuscript, each table and each figure along with the respective legend are imbedded within the text of the manuscript. 

 

If the major part of caption is the same, you can ask the reader to see the rest in Table 1 or Figure 2 and it is not necessary to repeat all of it again.

Response:  The data and information regarding Table 1 and Table 2 are fundamental different and thus, have not been markedly edited.  However, the suggestion to not repeat information that is common to all figures, e.g., use of the letter ‘a’ to indicate that control data and identified data for experimental groups are statistically similar, is well taken.  In the revised manuscript such information is stated in first and once in the legend for Figure 1.

 

All the numbers indicated to references should be as [1] and see from instructions for authors (home page of journal) how they should be in reference list.

Response:  All numbers for references are cited within brackets, e.g., [3,4,5], as per the Journal’s instructions for authors. 

Reviewer 3 Report

The authors have studied the effects of low concentrations of cadmium (Cd), nickel (Ni) and their binary mixtures on various endpoints in embryo-larval zebrafish. The manuscript is written in a clear and concise way. Most of the methods are described well, the results have been mostly presented in a clear way, and the results have been discussed well. The introduction, however, fails to present the rationale of the study, and it could be significantly improved. The authors do not report having measured the metal concentrations in the exposure media, and the immunohistochemistry image analysis has not been described. The issues with the introduction make me rate the manuscript as "major revision" (as I think it will not be minor work to do those changes). 

Here are some more detailed comments that I hope will help in improving the manuscript.

Introduction:

The introduction is rather scattered, telling about effects of Cd and Ni in terrestrial systems, mammals, etc., but not giving important background information; e.g. concentrations of Cd and Ni in aqueous systems that would be important for assessing how environmentally relevant the used concentrations were (and essential if you want to make conclusions about the environmental risk of these compounds).  The introduction should concentrate on what is known about the effects of Cd and Ni in fish. The rationale of the study is rather vague in the present form of the manuscript, and the authors should make it clearer what the knowledge gap is that will be filled with this study. There is plenty of literature on the effects of Cd and Ni in zebrafish - what is novel in this manuscript?

Material and Methods:

Most of the methods have been described adequately. The authors should tell how the exposure concentrations were chosen. The image analysis for immunohistochemistry has not been described at all in M&M. It says in the figure caption of Fig. 8 that “staining intensity in experimental embryos was expressed as a percent of staining in control embryos”, but the method should be described in M&M.

The authors do not report having measured concentrations of the metals in the exposure solutions. It would be good to have measured values, especially when you are using exposure vessels made of plastic, as some of the metals will stick to the walls of the containers.

Results:

The results section is clear, and the figures are also rather clear and informative. It would be still clearer for the reader if the statistical significance over the bars was marked so that the same letter/symbol above the bar would denote statistical similarity. So all the treatments that are not statistically different from each other would get the same letter/symbol (“belong to the same group”) and the ones that are statistically different would have different symbols from each other.

Discussion:

The discussion repeats the results somewhat. I think this is a borderline issue - some repetition is inevitable, but I think it could be improved a little. The conclusions are supported by the results, and the flow of ideas is good.

Author Response

Introduction

The introduction is rather scattered, telling about effects of Cd and Ni in terrestrial systems, mammals, etc., but not giving important background information; e.g. concentrations of Cd and Ni in aqueous systems that would be important for assessing how environmentally relevant the used concentrations were (and essential if you want to make conclusions about the environmental risk of these compounds).  The introduction should concentrate on what is known about the effects of Cd and Ni in fish.  The rationale of the study is rather vague in the present form of the manuscript, and the authors should make it clearer what the knowledge gap is that will be filled with this study.  There is plenty of literature on the effects of Cd and Ni in zebrafish what is novel in this manuscript?

Response:  We have edited the introduction to focus more on aquatic contamination by heavy metals and the known effects on embryonic development and reproduction. We have emphasized that this study examined the combined effects of Ni and Cd in our model system.

Materials and Methods

Most of the methods have been described adequately. The authors should tell how the exposure concentrations were chosen.  The image analysis for immunohistochemistry has not been described at all in M&M. It says in the figure caption of Fig. 8 that “staining intensity in experimental embryos was expressed as a percent of staining in control embryos”, but the method should be described in M&M.

Response:  Text has been added to the methods that describes the environmental contamination levels for cadmium and nickel that have been observed in various water sources around the world.   We have included this statement, “Fixed embryos were immunostained for myosin heavy chain protein; staining intensity in experimental embryos was expressed as a percent of staining in control embryos.” in the Methods.  We have also included a succinct statement in the Results and Figure Legend 7 to remind that reader that staining intensity in experimental embryos was expressed as a percent of staining in controls.

Comment: The authors do not report having measured concentrations of the metals in the exposure solutions. It would be good to have measured values, especially when you are using exposure vessels made of plastic, as some of the metals will stick to the walls of the containers.

Response:  The concentrations of cadmium and nickel in exposure solutions were not measured.  Certain plastics can bind metals. 

Results:

Comment: The results section is clear, and the figures are also rather clear and informative. It would be still clearer for the reader if the statistical significance over the bars was marked so that the same letter/symbol above the bar would denote statistical similarity. So all the treatments that are not statistically different from each other would get the same letter/symbol (“belong to the same group”) and the ones that are statistically different would have different symbols from each other.

Response:  A similar suggestion was made by Reviewer #2.  In the revised figures when values for experimental treatments and values for the respective age-matched control are not statistically different, the respective values for experimental and control treatment are marked with the letter ‘a’.  Values for experimental treatments marked with a bare significantly different than values for the age-matched control (p<0.05).  In other figures, the letters, c, d, e, and f are used to indicate statistically significant differences between different concentrations of nickel or cadmium, or the different combinations of nickel and cadmium.

Discussion:

The discussion repeats the results somewhat. I think this is a borderline issue - some repetition is inevitable, but I think it could be improved a little. The conclusions are supported by the results, and the flow of ideas is good.

Response:  We have attempted to keep repetition in the discussion to a minimum.

Round 2

Reviewer 1 Report

Dear authors,

 

Your clarifications improved the manuscript and I have only minor considerations:

• Alter the first two sentences in the abstract:

“Exposure to even low concentrations of heavy metals can be toxic to aquatic organisms, especially during embryonic development. Thus, this study aims to study the toxicity of nickel and cadmium…”

 

• Just to clarify, you did not measure the pH of the metal solutions, correct? A pH alteration can also have some implications in larvae behaviour.

 

• You stated that “the variances that were generated in the ANOVAs were not very different”, but you need to perform a test to confirm that (Levene’s test for example). Moreover, to use parametric tests like ANOVA, the data must have a normal distribution and you did not state if you verify that.

 

• Still, you report using ANOVA (“One-way analysis of variance (ANOVA) was used to determine whether differences existed among the remaining responses to the different concentrations of heavy metals.”), but then you refer to use Student’s t-test to compare means for each treatment, and this is also what one-way ANOVA does but for more than two groups that are the cases of this analysis. Thus, I still do not understand what parameter did you use Student’s t-test. In the manuscript, you refer to this test when you analyse the myosin expression, but you also mention the ANOVA for the same type of comparisons (which is the correct to use). Please choose to only use one-way ANOVA with posthoc tests for these types of comparisons and correct the analysis according to this.

 

• My request regarding the rest of the data (data not shown) in supplementary figures is in the name of publication transparency and because the raw data is often useful when you are planning a similar experiment. But this will be an editorial decision.

 

• In the last sentence of the manuscript add “that”: “Thus, these results indicate that…”

 

• Just for you to know, the concentration of euthanasia is too low for embryos, but the legislation does not require to complete euthanasia in such a young stage of zebrafish (i.e. not protected by EU directive, only with 6dpf). Nevertheless, science is always evolving, and we do not actually know when suffering starts. Thus, I recommend to always complete euthanasia. In the case of the embryos that are going to be fixed, this procedure complete euthanasia, but not in the 24hpf embryos, and they can stop moving without being dead, you had to confirm that by lack of blood circulation.

Author Response

Reviewer 1.

 

Alter the first two sentences in the abstract: “Exposure to even low concentrations of heavy metals can be toxic to aquatic organisms, especially during embryonic development. Thus, this study aims to study the toxicity of nickel and cadmium…”

 

**We have changed the first two sentences in the abstract to reflect the suggested changes indicated by reviewer 1.  The suggestions have improved the text.  Thank you.

 

 

Just to clarify, you did not measure the pH of the metal solutions, correct? A pH alteration can also have some implications in larvae behaviour.

 

**That is correct; we did not measure the pH of the embryo medium after the metals were added.  This is a condition of the experiments.

 

 

You stated that “the variances that were generated in the ANOVAs were not very different”, but you need to perform a test to confirm that (Levene’s test for example). Moreover, to use parametric tests like ANOVA, the data must have a normal distribution and you did not state if you verify that.

 

**The homogeneity of variances of data has been analyzed by the D’Agostino-Pearson test.  Non-normally distributed data were analyzed using Kruskal-Wallis test (ANOVA).  This information has been added to the manuscript. 

 

 

Still, you report using ANOVA (“One-way analysis of variance (ANOVA) was used to determine whether differences existed among the remaining responses to the different concentrations of heavy metals.”), but then you refer to use Student’s t-test to compare means for each treatment, and this is also what one-way ANOVA does but for more than two groups that are the cases of this analysis. Thus, I still do not understand what parameter did you use Student’s t-test. In the manuscript, you refer to this test when you analyze the myosin expression, but you also mention the ANOVA for the same type of comparisons (which is the correct to use). Please choose to only use one-way ANOVA with posthoc tests for these types of comparisons and correct the analysis according to this.

 

**We have complied with this request.  All the post-hoc tests were carried out using Student’s t-test WITH the Bonferroni correction.  The Bonferroni correction also called the Bonferroni type adjustment, is a valid method to use to avoid Type 1 errors in statistical analysis.  In fact, it is a quite conservative test.

 

 

My request regarding the rest of the data (data not shown) in supplementary figures is in the name of publication transparency and because the raw data is often useful when you are planning a similar experiment. But this will be an editorial decision.

 

**Thank you. We agree that this should be an editorial decision.

 

 

In the last sentence of the manuscript add “that”: “Thus, these results indicate that…”

 

**The suggested text (the word “that”) has been added to the last sentence of the manuscript.

 

Just for you to know, the concentration of euthanasia is too low for embryos, but the legislation does not require to complete euthanasia in such a young stage of zebrafish (i.e. not protected by EU directive, only with 6dpf). Nevertheless, science is always evolving, and we do not actually know when suffering starts. Thus, I recommend to always complete euthanasia. In the case of the embryos that are going to be fixed, this procedure complete euthanasia, but not in the 24hpf embryos, and they can stop moving without being dead, you had to confirm that by lack of blood circulation.

 

**We thank this reviewer for their comments.  While we cannot definitely state that we checked every single one of the several thousand embryos that were used in this study, we were quite mindful of the need for adequate euthanasia.  When the embryos were placed in Petri dishes with the MS222 (hatched or not), they were closely examined for the cessation of movement.  The embryos were chilled on ice when they were in the MS222 and when they were in the formalin fixative.  The 24-hpf embryos were placed in formalin, even though they were not used in the immunohistochemistry studies included in this particular study.  Therefore, as you stated, the use of fixation completed euthanasia for all the embryos in this study.  We have changed the text to indicate that all the embryos were placed in fixative after treatment with MS222.  Here is the adjusted text in the Methods section at the end of section 2.3:   “All embryos were then immersed in 10% neutral buffered formalin (NBF) for 24 hours at 4°C.  The embryos were transferred to phosphate-buffered-saline (PBS; pH 7.2) and stored at 4°C.  The 48- and 72-hpf embryos were subsequently processed for morphometric and immunohistochemical analyses.” 

Reviewer 2 Report

Still

1, In line 188 omit the superscript. 

 

2. In Fig 2 A

Statistically: all treatments marked with a means, that there is no statistical difference between these treatments. All the other letters (b, c, d etc) mean that there is statistically significant difference. Or? 

What means ac; is the treatment Ni 0.5 mg/L  statistically significantly different or similar?      

3. Correct still all references as given in instructions. 

Author Response

Reviewer 2

1. In line 188 omit the superscript.

**This superscript has been removed.  It was ‘36’, which was the original number assigned to Westerfield, 2000 and did not delete in the first revision.  [In the original manuscript all citations were mistakenly superscripted.] 

 

2. In Fig 2 A

Statistically: all treatments marked with a means, that there is no statistical difference between these treatments. All the other letters (b, c, d etc) mean that there is statistically significant difference. Or?

What means ac; is the treatment Ni 0.5 mg/L statistically significantly different or similar?

**We understand that Reviewer 2 had some difficulty with our labels on the graphs.  We have revised the letters used to indicate significant differences and have simplified the labeling in the graphs to comply with the suggestion of Reviewer 2 and Reviewer 3.

 

3. Correct still all references as given in instructions.

**We have reformatted all references in compliance with the style of papers published in ‘Water’ to the best of our ability.  

Reviewer 3 Report

I am satisfied with most changes the authors have made, and I think the manuscript may be published after minor revision. Here is a short list of issues that need to be improved or corrected.

The introduction is better now than in the previous version, but the background of what is known about toxicity of Cd and Ni (alone) in zebrafish embryos is still somewhat vaguely presented, and I think it can still be improved. I would recommend putting still more emphasis on telling about developmental effects of metals, especially Cd and Ni, in zebrafish, and showing that though there is literature on individual metals and some binary and tertiary mixtures, and complex (environmental or "environmental-mimicking") mixtures, information on mixtures of Ni and Cd are lacking or scarce. Also, you can tell if others have used higher concentrations and yours are lower.

The references are not always adequate for the sentences, e.g. lines 163-165 “fish embryos are being used… to investigate the effects of heavy metals on early embryonic development”, reference 22 tells about toxicity of diclofenac, and 24 about toxicity of a nano-composite. Please replace these with those about studies on metals (especially Cd and Ni), and check for other sentences as well that the references refer to what you say in the sentence.

I think the methods for immunohistochemistry still need clarification. How did you get the numerical values for the staining intensity in the embryos (for calculating percent of staining)? Which image analysis program did you use?

I think you could shortly mention in the text that the metal concentrations are nominal concentrations, and that due to absorption to plastic the "real" concentrations may actually have been lower.

The marking of statistical difference in figures is incorrect. If you mark the similarity with the same letter in a correct way, you do not need to explain in the figure caption anything else but state that “statistical similarity is denoted with the same letter”. Thus, for example Fig. 1B you would mark control and Cd 0.015 with “a”, Cd 0.15 and Cd 0.705 with “ab” and Cd 1.5 with “b”, to mean that there is statistical difference between control and Cd 1.5, and between Cd 0.015 and Cd 1.5. In the same way, Fig. 1D mark control with “a”; Ni 0.1 with “ab” (statistically similar with control, Ni 0.5, and Ni 2.5); Ni 0.5 with “abc” (statistically similar to all); Ni 2.5 with “bc” (statistically similar to all other but control) and Ni 5 with “c” (statistically similar to Ni 0.5 and Ni 2.5). Correct all other figures in the same way.

Also, there are some typos and other mistakes that are either new or that I did not spot in the first version. So I recommend a careful proof-reading. For example, the first sentence of the abstract does not make sense, and the reference list contains many typos.

Author Response

Reviewer 3

The introduction is better now than in the previous version, but the background of what is known about toxicity of Cd and Ni (alone) in zebrafish embryos is still somewhat vaguely presented, and I think it can still be improved. I would recommend putting still more emphasis on telling about. developmental effects of metals, especially Cd and Ni, in zebrafish, and showing that though there is literature on individual metals and some binary and tertiary mixtures, and complex (environmental or “environmental-mimicking") mixtures, information on mixtures of Ni and Cd are lacking or scarce. Also, you can tell if others have used higher concentrations and yours are lower.

**We have revised the next to last two paragraphs in the introduction to emphasize the effects of metals on fish development.  The revised text is indicated in green and purple-blue text.

**We also added the following information to the methods in section 2.2.

“These concentrations were chosen based on reports of heavy metal contamination in different water sources around the globe. For example, cadmium ground water contamination in Pakistan was as high as 0.18 mg/L and nickel was as high as 1.0 mg/L [32].  Cadmium contamination in the Caspian Sea, Iran, has been measured at 120 µg/L [33].  The highest levels of nickel contamination in the ocean near the Philippines was 0.6 mg/L [34].  Thus, the ranges of concentrations used in this investigation were environmentally relevant for highly contaminated waters water sources around the world.”

 

The references are not always adequate for the sentences, e.g. lines 163-165 “fish embryos are being used… to investigate the effects of heavy metals on early embryonic development”, reference 22 tells about toxicity of diclofenac, and 24 about toxicity of a nano-composite. Please replace these with those about studies on metals (especially Cd and Ni), and check for other sentences as well that the references refer to what you say in the sentence.

**The reference by Hallare AV, Köhler HR, Triebskorn R., has been removed.  However, the reference by Weil M, Meißner T, Busch W, Springer A, Kühnel D, Schulz R, et al.  was not removed, as it discusses the effects of a form of iron.

 

I think the methods for immunohistochemistry still need clarification. How did you get the numerical values for the staining intensity in the embryos (for calculating percent of staining)? Which image analysis program did you use?

**Additional information has been added to the methods section (Section 2.5) concerning the measurement of the immunohistochemical staining for myosin. We used Image J for analysis.

 

I think you could shortly mention in the text that the metal concentrations are nominal concentrations, and that due to absorption to plastic the "real" concentrations may actually have been lower.

**We have added the following text to section 2.2.  “We note that these metal concentrations were nominal concentrations, due to the possibility that some absorption to the plastic may have occurred.  Therefore, the actual concentrations to which the ZFEs were exposed may actually have been somewhat lower.”

 

The marking of statistical difference in figures is incorrect. If you mark the similarity with the same letter in a correct way, you do not need to explain in the figure caption anything else but state that “statistical similarity is denoted with the same letter”. Thus, for example Fig. 1B you would mark control and Cd 0.015 with “a”, Cd 0.15 and Cd 0.705 with “ab” and Cd 1.5 with “b”, to mean that there is statistical difference between control and Cd 1.5, and between Cd 0.015 and Cd 1.5. In the same way, Fig. 1D mark control with “a”; Ni 0.1 with “ab” (statistically similar with control, Ni 0.5, and Ni 2.5); Ni 0.5 with “abc” (statistically similar to all); Ni 2.5 with “bc” (statistically similar to all other but control) and Ni 5 with “c” (statistically similar to Ni 0.5 and Ni 2.5). Correct all other figures in the same way.

**As suggested, we have revised the letters used to indicate the significant differences and simplified the labeling in the graphs to comply with the reviewer's suggestion.  In the legend for Figure 1, we have noted “Note: In all figures, within a given panel statistical similarity is denoted with the same letter (p<0.05).”

 

Also, there are some typos and other mistakes that are either new or that I did not spot in the first version. So I recommend a careful proof-reading. For example, the first sentence of the abstract does not make sense, and the reference list contains many typos.

**We have clarified the first sentence of the abstract.  We have checked the manuscript for additional typographical errors, and we have revised the references.