Population Structure and Genetic Diversity of the Spotted Sleeper Odontobutis interrupta (Odontobutidae), a Fish Endemic to Korea

Round 1

Reviewer 1 Report

The authors analyzed the population structure and genetic diversity of Odontobutis interrupta by means of microsatellites combined with mitochondrial DNA, which provided a valuable data basis for genetic diversity assessment and population management of this species. However, there are still several issues should be further addressed.

1. The abstract is too broad, and some important data and technical methods should be described in detail.

2. Geographical differences are discussed in the discussion, and it is suggested to introduce the geographical features and differences of the study sites in the introduction.

3. In the sample collection section, the setting of the sampling sites should be properly described, or the sampling map (Figure 1) should be added the legend to describe the river to which each sampling sites belongs.

4. Phylogenetic analysis should be added to determine the relationship between different geographic populations, and comprehensive analysis combined with the median-joining network will be more convincing.

5. The chart information is incomplete, please add. Such as, figure 1 lacks a scale and some symbols are not marked; the meaning of microsatellite searching numbers in Table S2 is not clearly explained.

6. There are five river systems shown in the supplementary material (Table S1), while there are only four blocks in Figure 1. Please clarify.

7. Are the contents of Figure 2 and Table 2 in the manuscript repeated? Is it more intuitive to reflect the number of samples that share haplotypes in Figure 2?

Author Response

Response author

Q1.

1. The abstract is too broad, and some important data and technical methods should be described in detail.

A1. Thanks for the review. Edited based on reviewer comments.

Q2.

2. Geographical differences are discussed in the discussion, and it is suggested to introduce the geographical features and differences of the study sites in the introduction.

A2. Thanks for the review. Edited based on reviewer comments.

[The Korean Peninsula is genetically different because biogeographically, the uplift of mountain ranges hinders genetic flow [12]. Identifying these genetic differences is an important factor in conservation planning [4].]

[The results of this study differed from previously reported results, indicating clear genetic differences between watersheds owing to geographic differences [59]. This might suggest the effects of human-mediated artificial translocations among genetically divergent populations from different river drainages in South Korea [60,61].

Freshwater fishes typically exhibit genetic differences depending on the geographical water system and little genetic differentiation within the same water system. Interestingly, SCR exhibited significant genetic differences with SOD and SGD. The SOD, SCR, and SGD populations in the Seomjingang River water system were expected to exhibit low genetic differences owing to gene exchange before the dam construction. However, a high level of genetic differences was observed. These genetically distinct differences are corroborated by the DAPC and STRUCTURE results. Solid genetic differentiation can be explained by non-natural processes, such as anthropogenic activities or fragmentation, despite being in the same water system. However, since the latter is a short duration (50 years) for substan-tial genetic differentiation to occur because of fragmentation, these results are likely at-tributed to the translocation owing to anthropogenic activities.

Conversely, the populations of the Geumgang and Hangang Rivers were geograph-ically separated and were expected to exhibit distinct genetic differences. Nonetheless, the populations of the Geumgang and Hangang Rivers, excluding GBG, showed little genetic difference. It is not known which populations of the Hangang and Geumgang Rivers first mixed, but considering the significant genetic differences between GBG vs. GGP and GNS in the same water system of the Geumgang River, an artificial genetic admixture is ex-pected to have occurred. This was corroborated by the DAPC and STRUCTURE results. However, GBG also exhibited a mixed genotype with the Hangang River populations; therefore, we presume that genetic mixing of the Hangang River water system population occurred.]

Q3.

3. In the sample collection section, the setting of the sampling sites should be properly described, or the sampling map (Figure 1) should be added the legend to describe the river to which each sampling sites belongs.

A3. Thanks for the review. Edited based on reviewer comments.

[Figure 1]

Q4.

4. Phylogenetic analysis should be added to determine the relationship between different geographic populations, and comprehensive analysis combined with the median-joining network will be more convincing.

A4. Thanks for the review. Edited based on reviewer comments.

[Figure 3]

Q5.

5. The chart information is incomplete, please add. Such as, figure 1 lacks a scale and some symbols are not marked; the meaning of microsatellite searching numbers in Table S2 is not clearly explained.

A5. Thanks for the review. Edited based on reviewer comments.

[Figure 1, Table S2]

Q6.

6. There are five river systems shown in the supplementary material (Table S1), while there are only four blocks in Figure 1. Please clarify.

A6. Thanks for the review. Edited based on reviewer comments.

[Figure 1]

Q7.

7. Are the contents of Figure 2 and Table 2 in the manuscript repeated? Is it more intuitive to reflect the number of samples that share haplotypes in Figure 2?

A7. Thanks for the review. It is repeated, but since it is intuitive that both are present in the text of the manuscript, they have not been moved to the supplementary table.

Author Response File: Author Response.docx

Reviewer 2 Report

It is not clear, in some parts of manuscript four main river systems are distinguished, while in other parts of text, five river systems are indicated.

L18 please provide Hd and π values

L53-55 change to “Another species of the family Odontobutidae, O. potamophila closely related to O. interrupta, is a promising aquaculture species in China that can grow up to 11 cm”

L56 not clear what do you mean by this sentence, please specific, what strategies are needed? Or rephrase sentence.

L58-60 please add reference

L66-68 Is O. interrupta an invasive species? Does this species do the harm to the environment or native species? If no, it is introduced but not invasive species. Change the sentence declaring that introduced species that become invasive can negatively affect native species, ecology niches etc.

L77 please add thus far/until now/ to date

L81 at first mentioning, you should write full name of the gene and abbreviation in the brackets

L77-85 not clear which exact mtDNA loci were examined, comprehensively or just pilot studies, whether or not the sequences of the complete mtDNA genome are known for this species

L261, 425 please use the abbreviation mtDNA

L86-91 please provide in this paragraph which markers were applied for the study

L93, subsection headings must be written using a style where the first letter of word is capitalized

L94 “Iwata and Jeon (1985)” delete authors of species, if no you should provide authors for all species mentioned in the manuscript. Furthermore, authors must be added at first mentioning of the species

L96 some important information will be only in supplementary material, therefore I suggest to shortly describe Table S1. Twelve samples from four river systems (which ones), sample size ranging from 20 to 23. Also modify TableS1 it is not clear which samples represents river systems.

L100 microsatellite loci

L103 please add details about colors of oval clouds, which river systems they represent?

L140 what was the length of PCR extension?

L142 it is critical remark. Why cox1 sequences obtained were not deposited in the GenBank? All sequences obtained must be deposited in GenBank. Sequences are raw data and they are essential for future comparisons and investigations.

L150-154 it is not clearly defined which software was used for the estimating bottleneck by the second method.

L157 revise English of this sentence.

L158-159 it should be moved to introduction

L159 Parameters of haplotype diversity and nucleotide diversity…

L163-174 it is not clear from the description if the same software and analysis were performed for both data groups mtDNA and microsatellite, whether specific analyses were applied to microsatellite and mtDNA data

L186-187 where it is displayed in the TableS3 that there is no evidence of scoring of null alleles?

L199 please change the order, at first mtDNA and then microsatellite, since in the Table at first data on mtDNA is presented

Table 1 it is not clear which ID represents which river systems. For example, you can use shadowing with different colors the ID column to present the data. The same remark is for Table 4

L200-202 π: nucleotide diversity. Change in the table nucleotide diversity to π. In methods, please add how Fu’s values, D: Tajima’s values were estimated, which programs, significance of these values were checked or not?

L188-222 The short summary is needed which samples exhibited low genetic diversity by microsatellite, mtDNA or both markers; the same about high genetic diversity. I do not agree that NJH has low genetic diversity by microsatellites

L217 you should clearly state that COI diversity was low in the species examined.

L237-243 here you present results obtained using three models, while in methods two models were described (L149)

L235 the reference here is necessary, since different authors give different figures

L283-284 more detail description is needed, what was the similarity between two analyses?

Table 5 how it could be microsatellite on shared haplotypes? Haplotypes are of mtDNA. It is not clear how AMOVA groping for microsatellite data was performed.

L314, 437, 458 Full genus names must be written at the beginning of the sentence.

L316 spp. not italic

L416-421 the AMOVA results should be discussed with a caution, since different samples were grouped using mtDNA and microsatellite data, which may affect the results.

minor editing is needed. 

Author Response

Q1.

It is not clear, in some parts of manuscript four main river systems are distinguished, while in other parts of text, five river systems are indicated.

A1.

Thanks for the review. Edited based on reviewer comments.

Q2.

L18 please provide Hd and π values

A2.

Thanks for the review. Edited based on reviewer comments.

[However, mitochondrial DNA exhibited low genetic diversity (Hd: 0.000-0.674, π: 0.00000-0.00159).]

Q3.

L53-55 change to “Another species of the family Odontobutidae, O. potamophila closely related to O. interrupta, is a promising aquaculture species in China that can grow up to 11 cm”

A3.

Thanks for the review. Edited based on reviewer comments.

Q4.

L56 not clear what do you mean by this sentence, please specific, what strategies are needed? Or rephrase sentence.

A4.

Thanks for the review. Edited based on reviewer comments.

[Breeding strategies are needed to conserve endemic species that show potential for aq-uaculture [14,15].]

Q5.

L58-60 please add reference

A5.

Thanks for the review. Edited based on reviewer comments.

[Damage to these habitats reduces population size, and genetic admixture can result in unexpected negative consequences [4]. Hence, it is imperative to conduct a genetic as-sessment to establish effective conservation strategies and genetic management units [4].]

Q6.

L66-68 Is O. interrupta an invasive species? Does this species do the harm to the environment or native species? If no, it is introduced but not invasive species. Change the sentence declaring that introduced species that become invasive can negatively affect native species, ecology niches etc.

A6.

Thanks for the review. This species is presumed to engage in breeding competition with species with similar ecological niches. Therefore, we have changed the sentence from invasive speices -> introduced, so please understand.

Q7.

L77 please add thus far/until now/ to date

L81 at first mentioning, you should write full name of the gene and abbreviation in the brackets

A7.

Thanks for the review. Edited based on reviewer comments.

[Thus far genetic research on O. interrupta has only centered around the characteristics of the mitochondrial genome]

[Mitochondrial Cytochrome c oxidase I (COI) regions are typically used as barcoding regions and are useful for tracking translocation because they facilitate discrimination be-tween species and populations within species [24–26].]

Q8.

L77-85 not clear which exact mtDNA loci were examined, comprehensively or just pilot studies, whether or not the sequences of the complete mtDNA genome are known for this species

A8.

Thanks for the review.

Thus far, only the characteristics of the mtDNA genome were studied, and no other COI genes were studied. This is the first study in this study. I hope you understand.

Q9.

L261, 425 please use the abbreviation mtDNA

A9.

Thanks for the review. Edited based on reviewer comments.

Q10.

L86-91 please provide in this paragraph which markers were applied for the study

A10.

Thanks for the review. Edited based on reviewer comments.

[Additionally, this study is the first to investigate the genetic diversity and structure of O. interrupta for microsatellite and mtDNA marker, and thus provide valuable genetic insights for population management in the development of conservation strategies.]

Q11.

L93, subsection headings must be written using a style where the first letter of word is capitalized

A11.

Thanks for the review. Edited based on reviewer comments.

[Materials and methods]

Q12.

L94 “Iwata and Jeon (1985)” delete authors of species, if no you should provide authors for all species mentioned in the manuscript. Furthermore, authors must be added at first mentioning of the species

A12.

Thanks for the review. Edited based on reviewer comments.

[Odontobutis interrupta (1985) are endemic to Korea; therefore, animal ethics approval was waived.]

Q13.

L96 some important information will be only in supplementary material, therefore I suggest to shortly describe Table S1. Twelve samples from four river systems (which ones), sample size ranging from 20 to 23. Also modify TableS1 it is not clear which samples represents river systems.

A13.

Thanks for the review. Edited based on reviewer comments.

[Twelve populations of O. interrupta were sampled location and latitude longitude (more details are provided in Figure 1 and Table S1).]

Q14.

L100 microsatellite loci

A14.

Thanks for the review. Edited based on reviewer comments

Q15.

L103 please add details about colors of oval clouds, which river systems they represent?

A15.

Thanks for the review. Edited based on reviewer comments

Q16.

L140 what was the length of PCR extension?

A16.

Thanks for the review. Edited based on reviewer comments

[extension at 72 °C for 30 s]

Q17.

L142 it is critical remark. Why cox1 sequences obtained were not deposited in the GenBank? All sequences obtained must be deposited in GenBank. Sequences are raw data and they are essential for future comparisons and investigations.

A17.

Thanks for the review. Edited based on reviewer comments

The mitochondria DNA have been deposited at GenBank (OR352370-OR352377)

Q18.

L150-154 it is not clearly defined which software was used for the estimating bottleneck by the second method.

A18.

Thanks for the review. All bottlenecks were estimated using BOTTLENECK software.

[The method involved the BOTTLENECK software (Ver. 1.2.02) [33], a program for estimating bottlenecks through heterozygous excess testing, and the infinite allele model (IAM) [34]. A two-phase model (TPM) and stepwise mutation model (SMM) [35] were used to estimate, and TPM was performed with 10% variance and 90% SMM. In addition, each model had 10,000 iterations, and significance was verified using the Wilcoxon signed-rank test [36].]

Q19.

L157 revise English of this sentence. Edited based on reviewer comments

A19.

Thanks for the review.

[The effective population size (Ne) was determined using the linkage disequilibrium estimation of LDNe software [37].]

Q20.

L158-159 it should be moved to introduction

A20.

Thanks for the review. Edited based on reviewer comments

[Mitochondrial Cytochrome c oxidase I (COI) regions are typically used as barcoding regions and are useful for tracking translocation because they facilitate discrimination between species and populations within species [24–26]. In this study, COI was selected because it facilitates species identification of introduced populations and similar species. The integrated analysis of microsatellites and mitochondrial DNA (mtDNA) markers can significantly improve the investigation of genetic diversity and structure [27].]

Q21.

L159 Parameters of haplotype diversity and nucleotide diversity…

A21.

Thanks for the review. Edited based on reviewer comments

Q22.

L163-174 it is not clear from the description if the same software and analysis were performed for both data groups mtDNA and microsatellite, whether specific analyses were applied to microsatellite and mtDNA data

A22.

Thanks for the review. Edited based on reviewer comments.

[A median-joining network analysis was performed using Network (Ver. 10.2.0.0) [40] to create a haplotype network for determining the phylogenetic affinity between mtDNA se-quences.

2.5. Population genetic structure analysis

The ARLEQUIN software (Ver. 3.05) [32] of microsatellite and mtDNA was used to analyze the differences in genetics between groups as well as analyze molecular variance (AMOVA). The STRUCTURE software (Ver. 2.3) of microsatellite data set [41] was used to perform genetic structure clustering analysis based on the Bayesian method model. To es-timate the most suitable population, we set the population constant (K) to 1–10, and a suitable admixture model was applied to the mixture of water systems. The burn-in peri-od was repeated 10 times with 10,000 iterations, and Markov chain Monte Carlo with 100,000 iterations. To estimate a population-appropriate constant (K), we analyzed a study by Evanno et al. and the cluster results corresponding to K using STRUCTURE HARVESTER [42,43]. A discriminant analysis of principal components (DAPC) of mi-crosatellite data set was performed on the population using the R package ADEGENET (Ver. 2.1.3) [44], a nonmodel-based genetic clustering method.]

Q23.

L186-187 where it is displayed in the TableS3 that there is no evidence of scoring of null alleles?

A23.

Thanks for the review. Edited based on reviewer comments.

Q24.

L199 please change the order, at first mtDNA and then microsatellite, since in the Table at first data on mtDNA is presented

A24.

Thanks for the review. Edited based on reviewer comments.

Q25.

Table 1 it is not clear which ID represents which river systems. For example, you can use shadowing with different colors the ID column to present the data. The same remark is for Table 4

A25.

Thanks for the review. Edited based on reviewer comments.

Q26.

L200-202 π: nucleotide diversity. Change in the table nucleotide diversity to π. In methods, please add how Fu’s values, D: Tajima’s values were estimated, which programs, significance of these values were checked or not?

A26.

Thanks for the review. Edited based on reviewer comments.

Q27.

L188-222 The short summary is needed which samples exhibited low genetic diversity by microsatellite, mtDNA or both markers; the same about high genetic diversity. I do not agree that NJH has low genetic diversity by microsatellites

A27.

Thanks for the review. Edited based on reviewer comments.

[The four loci were analyzed for genetic diversity indices in the 12 populations (Table 1). The number of alleles ranged from 3.75–8.25, the H_O ranged from 0.440–0.756, and the expected heterozygosity ranged from 0.517–0.780. We found that the SCR, HDC, and NJH populations deviated from the HWE, whereas others followed the HWE. Inbreeding was observed solely in the SCR population, with significant values of F_IS (P < 0.001). The GGP populations from other regions exhibited high genetic diversity, with NA equal to 8.25 (H_O = 0.750). The NMG populations presumed to have been translocated from other regions exhibited low genetic diversity, with N_A equal to 3.75 (H_O = 0.529), respectively. Despite the presumption that the HSNC population was translocated, it exhibited high genetic diversity (H_O = 0.732). The SCR and HDC populations exhibited low genetic diversity (H_O = 0.440, 0.467), albeit natural (i.e., non-translocated populations).]

Q28.

L217 you should clearly state that COI diversity was low in the species examined.

A28.

Thanks for the review. Edited based on reviewer comments.

[In this study, the haplotype diversity of COI was found to be low.]

Q29.

L237-243 here you present results obtained using three models, while in methods two models were described (L149)

A29.

Thanks for the review. Edited based on reviewer comments.

[For SMM, we did not mention it because there were no significant values lower than 0.05.]

Q30.

L235 the reference here is necessary, since different authors give different figures

A30.

Thanks for the review. Edited based on reviewer comments.

[The effective population size of the NJH and NMG populations, presumed to be translocation populations, was substantially low, i.e., 9 and 10, respectively, although the 95% CI was 3–39 and 2-infinite, respectively (Table 3).]

Q31.

L283-284 more detail description is needed, what was the similarity between two analyses?

A31.

Thanks for the review. Edited based on reviewer comments.

[Bayesian clustering analysis maximized the ΔK value for population structure at K = 3 (Figure 3). At K = 3, the first group included SOD, SGD, and NMG; the second included SCR, HDC, and GBG. The third group included HWJ, HDM, NJH, GGP, GNS, and HSNC, but exhibited genetic admixture among various groups. In the case of NJH, NMG, and HSNC, genes from the three groups (Hangang River, Geumgang River, Somjingang River) were mixed. NJH was assigned to the Hangang River, NMG to the Seomjingang River, and HSNC were all mixed and assigned, suggesting that these presumed translocated populations exhibited distinct genetic origins. Results of non-model-based DAPC analysis showed similar trends to those of STRUCTURE (Figure 4).

]

Q32.

Table 5 how it could be microsatellite on shared haplotypes? Haplotypes are of mtDNA. It is not clear how AMOVA groping for microsatellite data was performed.

A32.

Thanks for the review. Edited based on reviewer comments.

[Four groups based on the distribution of water system]

Q33.

L314, 437, 458 Full genus names must be written at the beginning of the sentence.

A33.

Thanks for the review. Edited based on reviewer comments.

Q34.

L316 spp. not italic

A34.

Thanks for the review. Edited based on reviewer comments.

Q35.

L416-421 the AMOVA results should be discussed with a caution, since different samples were grouped using mtDNA and microsatellite data, which may affect the results. A1.

A35.

Thanks for the review. It is a result of careful preparation according to water lineage and haplotype. Thank you for your careful review.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Accept after minor revision (corrections to minor methodological errors and text editing).

Author Response

Q1. Accept after minor revision (corrections to minor methodological errors and text editing).

A1. Thank you for your review. 

First, I edited the error between the text and picture in Figure1.
Second, the GenBank deposit of the COI genes was added to the manuscript.

[The COI genes were deposited at Genbank (OR352370-OR352377).]