QTL-Seq Identifies Extra QTLs and Candidate Genes Controlling High Haploid Induction Rate in Maize

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents a high-resolution genetic analysis of the Haploid Induction Rate (HIR) in maize using a QTL-seq (Bulked Segregant Analysis) approach. The study effectively targets the gap in stabilizing HIR within diverse genetic backgrounds, particularly when the major loci qhir1 and qhir8 are already fixed. By integrating high-throughput sequencing with KASP marker validation, the authors identified novel candidate genes, specifically WEB1 and JAR1a, which contribute significantly to the phenotypic variance. The methodology is robust, leveraging an S2 population of 337 lines and achieving high sequencing depth. The identification of a missense mutation in WEB1 associated with blue-light-induced chloroplast movement provides a compelling, though currently theoretical, link to the physiological basis of haploid induction.

Major Revisions

Functional Validation Gap: While the association of WEB1 and JAR1a with HIR is statistically strong ($R^2$ up to 0.72), the study lacks direct functional evidence (e.g., CRISPR/Cas9 knockout or overexpression assays). At a minimum, the authors must strengthen the Discussion by providing a more detailed mechanistic model of how chloroplast relocation or jasmonate signaling (JAR1a) could biologically influence the fertilization failure or chromosome elimination required for haploid induction.

Data Accessibility: There is a project number, but it is inaccessible.

Refining the QTL-Seq Thresholds: The selection criteria for the 147 "high-confidence" SNPs from the initial pool of over 5,000 need clearer statistical justification. Provide the exact ∆(SNP-index) confidence interval used to define these candidate regions.

Minor Revisions

Nomenclature Consistency: Ensure that the "qhir1+/qhir8+" notation is clearly defined in the Materials and Methods. It should explicitly state whether this refers to the presence of the inducer alleles in a homozygous state.

Graphical Quality: The font size and resolution of axis labels in Figures 3 and 4 are insufficient for high-impact publication. These figures should be redrawn to ensure clarity at standard print dimensions.

Amino Acid Notation: When discussing the R to Q substitution, it is standard practice to provide the specific position of the residue within the protein sequence  to allow for easier orthology comparisons.

Author Response

Dear Reviewer 1,

We appreciate your valuable suggestions for further improving the manuscript. Therefore, the manuscript has been revised as your suggestions. Where modifications were made, the highlights are blue. The details of the revision are given below.

 

Point 1: Functional Validation Gap: While the association of WEB1 and JAR1a with HIR is statistically strong ($R^2$ up to 0.72), the study lacks direct functional evidence (e.g., CRISPR/Cas9 knockout or overexpression assays). At a minimum, the authors must strengthen the Discussion by providing a more detailed mechanistic model of how chloroplast relocation or jasmonate signaling (JAR1a) could biologically influence the fertilization failure or chromosome elimination required for haploid induction.

Response 1: We thank the reviewer for pointing out this issue. We have added details on how genes might lead to haploid induction.

            Lines 260-262:

“From this point, we believe that the WEB1 gene may be involved in the avoidance movement of chloroplasts, causing ROS in maize leaves. This leads to damaged pollen grains and haploid seeds”.

 

Lines 276-278:

“From this, we believe that the JAR1a gene may be involved in pollen grain development. This leads to immature pollen grains and haploid seeds”.

 

Point 2: Data Accessibility: There is a project number, but it is inaccessible.

Response 2: We thank the reviewer for pointing out this issue. We will share the public link once the publication is available.  In the meantime, here is the link for data access: https://dataview.ncbi.nlm.nih.gov/object/PRJNA1403211?reviewer=d7c0vpqmdgggeu5ei7dc8nnlv6

 

 

Point 3: Refining the QTL-Seq Thresholds: The selection criteria for the 147 "high-confidence" SNPs from the initial pool of over 5,000 need clearer statistical justification. Provide the exact ∆(SNP-index) confidence interval used to define these candidate regions.

Response 3: We thank the reviewer for pointing out this issue. We added the ∆SNP index to the main text.

 

Lines 176-182:

“We evaluated the R² values for 5,224 SNPs within the four QTLs including qHI2 (∆SNP index =0.39), qHI3 (∆SNP index =0.40), qHI6 (∆SNP index =0.35), and qHI8 (∆SNP index =0.38) and found that 147 SNPs with R² values greater than 0.3 were associated with qHI2, qHI3, and qHI8 and six candidate genes: GRMZM2G140156, GRMZM2G359746, GRMZM2G440943, AC198725.4, GRMZM2G091276, and GRMZM2G134738 (Table 3). Ten of these SNPs caused missense mutations in three genes: GRMZM2G359746, AC198725.4, and GRMZM2G091276.”

 

Point 4: Nomenclature Consistency: Ensure that the "qhir1+/qhir8+" notation is clearly defined in the Materials and Methods. It should explicitly state whether this refers to the presence of the inducer alleles in a homozygous state.

Response 4: We thank the reviewer for pointing out this issue. We added the homozygous state to the Materials and Methods section.

Lines 316-318:

“A population of 337 S2 haploid inducers was developed from crossing K8 (qhir1-/qhir8-_ homozygous dominant, low HIR, tropical) x BHI306 (qhir1+/qhir8+_ homozygous recessive, HIR of 10-15%, temperate) (Figure 4).”

 

Point 5: Graphical Quality: The font size and resolution of axis labels in Figures 3 and 4 are insufficient for high-impact publication. These figures should be redrawn to ensure clarity at standard print dimensions.

Response 5: We thank the reviewer for pointing out this issue. We have provided a clear version of the pictures in separate files.

 

Point 6: Amino Acid Notation: When discussing the R to Q substitution, it is standard practice to provide the specific position of the residue within the protein sequence to allow for easier orthology comparisons.

Response 6: We thank the reviewer for pointing out this issue. We removed that part of the protein substitutions from the discussion section.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

To address the instability of population induction efficiency (HIR) in earlier studies, constructing mapping populations and preliminarily screening related QTL loci，which might play a significant role in elucidating the molecular mechanisms underlying the induction function of induction lines. However, the study failed to further elucidate how the two candidate QTL loci enhance induction efficiency or whether they could improve the stability of induction efficiency in induction lines.

Thus,  the discussion parts should be strengthened to help readers to understand:  

how the two identified candidate genes might influence （improve） haploid induction and if they can and how to maintain the stability of their effects in the induced populations are needed further to explore and prospect in the discussion section.  

Comments on the Quality of English Language

The expression of the article needs to be further standardized and improved.

1）The letter labels B-E in F.3 are inadequately explained in the main text and should be supplemented with appropriate explanations to clarify the relationship between gene structure and SNP loci.

2）Standardize figure numbering in the main text. Use consistent numbering from F.1 to F.6. It is recommended to place FS1 directly in the main text and eliminate its footnote-style. Clearly explain how FS1 aids in differentiation within the main text. Avoid duplicate numbering, such as two F.4 figures.

3） Several grammatical issues include inconsistencies, repetitions, and errors. For example, line 71 and 17 show inconsistencies between DH word expression; multiple instances of abbreviations with full forms are repeated; KASP is first mentioned without its full form; Table 3 and Table 4 contain discrepancies between Chr. and Chromosome; line 261 incorrectly uses 'pollen' as an uncountable noun, and the predicate is misused.

Author Response

Dear Reviewer 2,

We appreciate your valuable suggestions for further improving the manuscript. Therefore, the manuscript has been revised as your suggestions. Where modifications were made, the highlights are blue. The details of the revision are given below.

 

Point 1: Line To address the instability of population induction efficiency (HIR) in earlier studies, constructing mapping populations and preliminarily screening related QTL loci, which might play a significant role in elucidating the molecular mechanisms underlying the induction function of induction lines. However, the study failed to further elucidate how the two candidate QTL loci enhance induction efficiency or whether they could improve the stability of induction efficiency in induction lines.

Thus, the discussion parts should be strengthened to help readers to understand: 

how the two identified candidate genes might influence （improve） haploid induction and if they can and how to maintain the stability of their effects in the induced populations are needed further to explore and prospect in the discussion section.

Response 1: We thank the reviewer for pointing out this issue. We added the discussion on improving the HIR and stability in lines 272-277.

            Lines 308-312:

“Nevertheless, we found that these two genes could improve HIR by an average of 12.78% (qhir1+/qhir8+/web1+/jar1a+), up from 6.66% (qhir1+/qhir8+). However, stability remains low due to the wide range of HIR in the population with qhir1+/qhir8+/web1+/jar1a+ (4.6%-23.01%). This suggests that other genes or QTLs may play a role in this stability.”

 

Point 2: The letter labels B-E in F.3 are inadequately explained in the main text and should be supplemented with appropriate explanations to clarify the relationship between gene structure and SNP loci.

Response 2: We thank the reviewer for pointing out this issue. We added the explanation for Figure 3B–E in lines 181-184.

Lines 184-187:

“We evaluated the R² values for 5,224 SNPs within the four QTLs including the first SNP located on chromosome 2 in the WEB1 gene (GRMZM2G359746) at exon 2 and the second SNP located on chromosome 8 in the JAR1a gene (GRMZM2G091276) at exon 5, were selected for Kompetitive Allele-Specific PCR (KASP) marker development (Figure 2B–E)”.

 

Point 3: Standardize figure numbering in the main text. Use consistent numbering from F.1 to F.6. It is recommended to place FS1 directly in the main text and eliminate its footnote-style. Clearly explain how FS1 aids in differentiation within the main text. Avoid duplicate numbering, such as two F.4 figures.

Response 3: We thank the reviewer for pointing out this issue. We revised the range of the figures in the manuscript and changed Figure S1 to Figure 5, providing an explanation within the main text.  

 

Point 4: Several grammatical issues include inconsistencies, repetitions, and errors. For example, lines 71 and 17 show inconsistencies between DH word expression; multiple instances of abbreviations with full forms are repeated; KASP is first mentioned without its full form; Table 3 and Table 4 contain discrepancies between Chr. and Chromosome; line 261 incorrectly uses 'pollen' as an uncountable noun, and the predicate is misused.

Response 4: We thank the reviewer for pointing out this issue. We removed the doubled haploid from line 71 and the quantitative trait loci from line 77. As mentioned, we added the full form of KASP. We replaced Chr. with Table 4. Additionally, we replaced "unviable pollen at floral stage 13" with "unviable pollen grains at floral stage 13."

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript “QTL-Seq identifies extra QTLs and Candidate Genes controlling high haploid induction rate in Maize” by Khammona et al studied 30 extreme high-HIR lines and 30 extreme low-HIR lines, to identify two potential genes (WEB1 and JAR1a) associated with high HIR via QTL-seq. 

In this study, the authors provided the details about methods and materials, and investigated the population of 337 S2 to analyze the data via QTL pipeline, from constructing extreme phenotype pools, whole-genome sequencing, to the calculation of SNP-index and Δ(SNP-index), and the development of molecular markers and validation. Current data may provide the insights of two potential genes (WEB1 and JAR1a) associated with high HIR. 

But there are some minor concerns that the authors should address below:

1. Line 133 In table 1, the column name should be  “Alignment (%)”. 
2. Line 280 It should be “Figure 1” instead of “Figure 4”. It is highly recommended that the authors briefly explain figure 1 in the results section (i.e. adding a new section 2.#), even though the details are presented in the methods section.
3. Delete Line 338 “Patents” section, as there is no patent related to the study.

Author Response

Dear Reviewer 3,

We appreciate your valuable suggestions for further improving the manuscript. Therefore, the manuscript has been revised as your suggestions. Where modifications were made, the highlights are blue. The details of the revision are given below.

 

Point 1: Line 133 In table 1, the column name should be “Alignment (%)”.

Response 1: We thank the reviewer for pointing out this issue. We revised that part to be “Alignment (%)” in table 1.

 

Point 2: Line 280 It should be “Figure 1” instead of “Figure 4”. It is highly recommended that the authors briefly explain figure 1 in the results section (i.e. adding a new section 2.#), even though the details are presented in the methods section.

Response 2: We thank the reviewer for pointing out this issue. We revised and gave details to “Figure 1” (lines 280).

 

Lines 110-122:

“We developed a population of 337 S2 haploid inducers by crossing K8 (qhir1-/qhir8-, low HIR, tropical) with BHI306 (qhir1+/qhir8+, 10-15% HIR, temperate). 19 superior S1 lines (qhir1+/qhir8+) were selected and assigned to two groups (Group A: 10 families; Group B: 9 families) based on genetic diversity to maximize recombination, yielding 337 S2 lines. We employed this intercrossing strategy to maintain genetic variation at other loci while fixing the major HIR QTLs. This facilitates the identification of additional minor-effect loci (Figure 4). This revealed that the HIR frequency distribution histogram is skewed toward low HIR values in populations (Figure 1). To identify the genomic region associated with HIR, we performed QTL-seq analysis on haploid inducers derived from a cross between BHI306 (the male parent) and a tropical inducer (K8). A total of 337 S2 haploid inducer lines were used in the study (Table S1). To ensure the quality and robustness of phenotyping, we exploited the differences between haploid (n) and diploid (2n) individuals at the seedling stage to minimize errors (Figure 5).”

 

 

Point 3: Delete Line 338 “Patents” section, as there is no patent related to the study.

Response 3: We thank the reviewer for pointing out this issue. We removed the “Patents section”. 

Author Response File: Author Response.pdf