Multi-Omics Integration Unravels the Genetic and Hormonal Regulatory Mechanisms Underlying Increased Main Stem Node Number in Soybean

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study addresses an agronomically important trait and very relevant in the plant field, with a coherent biological narrative linking node number and hormone signaling.

 Comments about the figures:

The figures are organized logically, but some images are dense and could use re-formatting.

It was hard to read the measurement tape scales in Fig 1A. It might also be helpful to to label the varieties directly in the image.

I would suggest to color asterisks in red due to accessibility concerns.

Fig 2B: Labeled text very small

Fig 3A: It’s unclear what the axis label “significant” means on its own.

What do the numbers next to the bar in Fig 3B refer to?

Fig 6: Overlapping red boxes make it hard to read the text. Maybe a legend or a figure reference could help.

Fig 7: Calibri font seems different than other figures with Arial font.

Methods

The workflow is mostly clear, but some parts can use more details for reproducibility, including growth criteria, and how overlap/candidate thresholds were chosen.

I assumed data processing was performed using R with packages like fastp, but R was not explicitly stated. Please mention if another terminal was used and the version.

Overall, the experimental concept and analyses are solid, but the weakness is presentation quality. The manuscript contains language problems and several places where the figures and text appear inconsistently formatted. I would also want more restraint in the discussion for a gene that has not yet been functionally validated in this study, and also no limitations of the study were mentioned.

Comments on the Quality of English Language

English language:

Some basic English proofreading needed, such as in areas of:

• L128: 29.5, Gb (no comma)
• All subtitles: Need spacing instead of period (depending on journal guidelines) 2.5.Phyto
• L410: Need spacing after numbered list 1) Euclidean Distance

Author Response

Review #1:

This study addresses an agronomically important trait and very relevant in the plant field, with a coherent biological narrative linking node number and hormone signaling.

Reply: Thank you very much for this positive and encouraging assessment. We have carefully addressed all of your specific suggestions in the point‑by‑point responses below.

 

Comments about the figures:

1. The figures are organized logically, but some images are dense and could use re-formatting.

Reply: Thank you. We have revised the figures accordingly as below.

 

2. It was hard to read the measurement tape scales in Fig 1A. It might also be helpful to to label the varieties directly in the image.

Reply: Thank you for your comment. We agree that direct labeling would be helpful, but adding artificial markers to finalized photographs may compromise data authenticity. The scales are visible upon magnification, and the figure primarily serves to show similar plant heights between varieties—consistent with Fig 2C statistics. The variety correspondence is explained in the legend.

 

3. I would suggest to color asterisks in red due to accessibility concerns.

Reply: Thanks. Asterisks are now red and enlarged.

Revision: See the main Figures 1.

 

4. Fig 2B: Labeled text very small

Reply: Thank you. Font size has been increased.

Revision: See the main Figures 2B.

 

5. Fig 3A: It’s unclear what the axis label “significant” means on its own.

Reply: Thank you for pointing this out. The axis label has been changed to "Gene number".

Revision: See the main Figures 3A.

 

6. What do the numbers next to the bar in Fig 3B refer to?

Reply: Thanks. The numbers next to each bar in Fig 3B indicate the gene number. This has now been clearly labeled in the revised figure.

Revision: See the main Figures 3B.

 

7. Fig 6: Overlapping red boxes make it hard to read the text. Maybe a legend or a figure reference could help.

Reply: Thank you. We have provided a detailed description of all genes in the figure legend.

Revision: Legend of Figure 6: “Figure 6. Volcano plots showing differential expression between LSD914 and HN48 in leaf tissue at the R2 stage (L.R2). (A) Gene‑level volcano plot. Each point represents a gene. The horizontal dashed line indicates adjusted p‑value = 0.01 (Benjamini‑Hochberg); vertical dashed lines indicate log₂ fold change = ±1. Red points: significantly up‑regulated genes; blue points: significantly down‑regulated genes; gray points: not significant. Five DEGs have been presented, including Gmax_RTD_Gm09G019730, Gmax_RTD_Gm03G021380, Gmax_RTD_Gm17G006930, Gmax_RTD_Gm18G003560, and Gmax_RTD_Gm16G008640. (B) Transcript‑level (isoform‑level) volcano plot. Five DTGs have been presented including, Gmax_RTD_Gm14G019190.RTD.1, Gmax_RTD_Gm18G006640.RTD.1, Gmax_RTD_Gm07G012400.RTD.2, Gmax_RTD_Gm17G004100.RTD.1, and Gmax_RTD_Gm12G005580.RTD.1. Their functional annotations are based on GO and KEGG analyses (see Methods).”

 

8. Fig 7: Calibri font seems different than other figures with Arial font.

Reply: Thanks, The font has been changed to Arial.

Revision: See the main Figure 7.

 

Methods

9. The workflow is mostly clear, but some parts can use more details for reproducibility, including growth criteria, and how overlap/candidate thresholds were chosen.

Reply: Thank you for this suggestion. Additional details have been added to the Materials and Methods to improve reproducibility.

Revision: 4.1. Plant materials and phenotyping: “Growth criteria included plant height (cm), main stem node number, branch number, internode length (cm), days to flowering (R1), days to maturity (R8), pod number per plant, and 100‑seed weight (g). All measurements were taken at maturity (R8) over four years (2021–2024). ”

4.2. Construction and sequencing for bulk segregant analysis by sequencing (BSA-seq): “F2 individuals were sampled at the V2 stage (third trifoliate leaf). Young leaves were collected individually, immediately frozen in liquid nitrogen, and stored at −80 °C. After maturity phenotyping, 32 individuals with extreme high‑node and 32 with extreme low‑node phenotypes were selected, and equal amounts of leaf tissue were pooled to form the two bulks for DNA extraction and sequencing”.

 

10. I assumed data processing was performed using R with packages like fastp, but R was not explicitly stated. Please mention if another terminal was used and the version.

Reply: Thank you for your comment. R and relevant package versions are now specified. Other processing used a Linux system. The relevant details have been added to the Methods section.

Revision: 4.9. Expression quantification and differential analysis: “Differential expression was assessed using R package “DESeq2” (version 1.36.0) [52].......Enrichment was tested using the hypergeometric test implemented in the “clusterProfiler” R package (version 4.4.4) [53].”

 

11. Overall, the experimental concept and analyses are solid, but the weakness is presentation quality. The manuscript contains language problems and several places where the figures and text appear inconsistently formatted. I would also want more restraint in the discussion for a gene that has not yet been functionally validated in this study, and also no limitations of the study were mentioned.

Reply: Thank you for your overall positive assessment of our experimental concept and analyses. We fully acknowledge the concerns you raised regarding presentation quality, language, formatting consistency, discussion restraint, and the absence of study limitations. We have carefully revised the presentation and language according to your comments. We agree that claims regarding the candidate gene (e.g., GmGASA32) should be tempered given the lack of functional validation in this study. This limitation has been added to the Discussion.

Revision: Discussion: “It should be emphasized that, despite the consistent differential expression and genomic localization of GmGASA32, we cannot rule out the possibility that it is linked to rather than directly responsible for the causal mutation. GmGASA32 is presented as a high‑priority candidate that requires functional validation (e.g., gene editing or complementation) to establish causality. Thus, the inferred role of GmGASA32 in regulating node number remains correlative and requires direct experimental confirmation.”

 

Comments on the Quality of English Language

English language:

Some basic English proofreading needed, such as in areas of:

12. L128: 29.5, Gb (no comma)

Reply: Thanks. This point has been revised.

Revision: 2.2. BSA-seq data analysis and variant detection: “A total of approximately 29.5 Gb of clean data were generated from the four samples (Table S1).”

 

13. All subtitles: Need spacing instead of period (depending on journal guidelines) 2.5.Phyto

Reply: Thank you. The correction has been made throughout the manuscript.

Revision: See the main text for details.

 

14. L410: Need spacing after numbered list 1) Euclidean Distance

Reply: Thanks. We have revised this point.

Revision: See the main text for details.

Reviewer 2 Report

Comments and Suggestions for Authors

This study used an integrated multi-omics strategy to investigate why the soybean mutant LSD914 produces significantly more main stem nodes than its parental line HN48. The authors identified major candidate genomic regions on chromosome 18 and proposed GmGASA32 as a central regulatory gene because it was consistently downregulated across tissues and developmental stages. They further suggest that altered hormonal balance, especially reduced jasmonate, salicylate, and auxin metabolites together with increased cis-zeatin, may shift growth from internode elongation toward node initiation, thereby increasing node number without substantially increasing plant height. These findings provide useful molecular targets for improving soybean architecture and yield. Some suggestions for authors:

1. Is GmGASA32 the causal gene, or only linked to the causal mutation?
2. How does GmGASA32 mechanistically regulate node initiation? The downstream signaling network connecting GmGASA32 to meristem activity or node formation remains unclear.
3. Please consider it as a perspective, perform CRISPR/Cas9 knockout and overexpression of GmGASA32. Because this would directly test causality and clarify whether reduced expression increases node number. Applying cytokinin, GA, or auxin modulators to mutant and wild-type plants could verify hormone interactions.
4. Please deepen the mechanistic discussion of hormone crosstalk. The manuscript mentions GA, auxin, JA, SA, and cytokinin changes, but a clearer integrated signaling model would improve impact.
5. The manuscript presents a strong multi-omics framework integrating transcriptomics and hormone profiling; however, the discussion of tissue-specific transcriptional regulation underlying increased node number would benefit from the inclusion of more recent high-resolution literature. The recent study https://doi.org/10.1016/j.tplants.2026.03.003 provides important context on cell-type-specific transcriptomic responses and spatial regulatory heterogeneity, which is relevant for interpreting developmental node initiation processes in soybean. Incorporating this study will enhance the mechanistic perspective and highlight the novelty of applying bulk multi-omics to a complex architectural trait.
6. What is the role of cytokinin cis-zeatin accumulation in the mutant? Whether elevated cis-zeatin is a cause or consequence of increased node number needs testing.
7. Are the identified chromosome 18 QTLs stable across environments and backgrounds?
   The mutant was tested mainly in one genetic background; broader validation is needed.
8. Please clarify the distinction between the node number and internode length regulation.
   This is central to the phenotype and should be more explicitly explained in Results and Discussion.
9. The phytohormone analysis is valuable; however, the interpretation of reduced auxin-related metabolites and hormone crosstalk would benefit from citing recent advances in auxin signaling biology. The recent review 10.1111/ppl.70165 provides comprehensive mechanistic insight into auxin biodynamics and its interactions with other phytohormones, directly relevant to the altered auxin–cytokinin balance proposed in this study. Incorporating this study will strengthen the hormonal framework and better emphasize the significance of your findings on soybean node architecture.
10. Please add more details in the figure legends.

Author Response

Review #2:

This study used an integrated multi-omics strategy to investigate why the soybean mutant LSD914 produces significantly more main stem nodes than its parental line HN48. The authors identified major candidate genomic regions on chromosome 18 and proposed GmGASA32 as a central regulatory gene because it was consistently downregulated across tissues and developmental stages. They further suggest that altered hormonal balance, especially reduced jasmonate, salicylate, and auxin metabolites together with increased cis-zeatin, may shift growth from internode elongation toward node initiation, thereby increasing node number without substantially increasing plant height. These findings provide useful molecular targets for improving soybean architecture and yield. Some suggestions for authors:

Reply: We sincerely thank the reviewer for the thoughtful and positive assessment of our work. We are glad that the reviewer recognized the value of our integrated multi-omics strategy and the identification of GmGASA32 as a central candidate, as well as the proposed hormonal shifts that may underlie the increased node number phenotype in the LSD914 mutant. The reviewer then provided a series of constructive suggestions. We have carefully addressed each of these suggestions in our point-by-point responses below.

 

1. Is GmGASA32 the causal gene, or only linked to the causal mutation?

Reply: Thank you for this important question. At this stage, GmGASA32 is a strongly prioritized candidate but causality remains unproven. We have clarified this in the Results and Discussion.

Revision: 2.8. Integrated analysis of BSA-seq candidates and differential expressions: “Fourteen potential QTLs were localized to the GmGASA32-containing region, suggesting that this region may influence expression patterns and contribute to the MSN difference in LSD914 (Table S16). However, whether GmGASA32 itself is the causal gene or is merely linked to the causative mutation requires functional validation.”

Discussion: “It should be emphasized that, despite the consistent differential expression and genomic localization of GmGASA32, we cannot rule out the possibility that it is linked to rather than directly responsible for the causal mutation. GmGASA32 is presented as a high‑priority candidate that requires functional validation (e.g., gene editing or complementation) to establish causality. Thus, the inferred role of GmGASA32 in regulating node number remains correlative and requires direct experimental confirmation.”

 

2. How does GmGASA32 mechanistically regulate node initiation? The downstream signaling network connecting GmGASA32 to meristem activity or node formation remains unclear.Please consider it as a perspective, perform CRISPR/Cas9 knockout and overexpression of GmGASA32. Because this would directly test causality and clarify whether reduced expression increases node number. Applying cytokinin, GA, or auxin modulators to mutant and wild-type plants could verify hormone interactions.

Reply: Thank you for this insightful question. We agree that mechanistic dissection is important but beyond the current study’s scope. We have acknowledged this limitation and framed our model as speculative.

Revision: See the Revision of Question #1.

 

3. Please deepen the mechanistic discussion of hormone crosstalk. The manuscript mentions GA, auxin, JA, SA, and cytokinin changes, but a clearer integrated signaling model would improve impact.

Reply: Thank you for this suggestion. In the revised Discussion, we have added an integrative model linking auxin, JA, SA, and cZ changes to a shift from "elongation-dominant" to "node-initiation-permissive" states.

accumulation of cis-zeatin as a positive cell-division signal.

Revision: Discussion: “Collectively, the reduction of auxins, JA, and SA may lower the threshold for node initiation by attenuating signals that typically promote elongation and stress‑related growth suppression. The concurrent accumulation of cZ, a cytokinin known to stimulate cell division in shoot meristems, likely provides a localized positive cue for new node formation. Thus, the hormone landscape in LSD914 appears to shift from a ‘elongation‑dominant’ state (higher auxins, JA, SA) toward a ‘node‑initiation‑permissive’ state (elevated cZ, reduced counteracting hormones).”

 

4. The manuscript presents a strong multi-omics framework integrating transcriptomics and hormone profiling; however, the discussion of tissue-specific transcriptional regulation underlying increased node number would benefit from the inclusion of more recent high-resolution literature. The recent study https://doi.org/10.1016/j.tplants.2026.03.003 provides important context on cell-type-specific transcriptomic responses and spatial regulatory heterogeneity, which is relevant for interpreting developmental node initiation processes in soybean. Incorporating this study will enhance the mechanistic perspective and highlight the novelty of applying bulk multi-omics to a complex architectural trait.

Reply: Thank you for the suggestion. We have cited the recommended review and added a sentence on single-cell/spatial transcriptomics as a future direction.

Revision: Discussion: “Finally, while our bulk transcriptome analysis captured average expression changes across whole tissues, emerging technologies such as single‑cell and spatial transcriptomics could in the future resolve cell‑type‑specific regulatory programs within the shoot apical meristem, potentially uncovering the precise spatial domains where GmGASA32 and hormone signals operate [42].”

 

5. What is the role of cytokinin cis-zeatin accumulation in the mutant? Whether elevated cis-zeatin is a cause or consequence of increased node number needs testing.

Reply: Thank you for this important question. Our data clearly show that cis-zeatin (cZ) accumulates specifically in the LSD914 mutant and is undetectable in the wild-type HN48. However, based on the current correlative data, we cannot determine whether elevated cZ is a cause or a consequence of increased node number. This distinction would require functional experiments such as exogenous cZ application to wild‑type plants or genetic manipulation of cZ biosynthesis genes. We have now explicitly acknowledged this limitation in the Discussion, stating that the observed cZ accumulation is a correlative signature and that causal testing is beyond the scope of this study.

Revision: Discussion: “However, the specific accumulation of cis‑zeatin (cZ) in LSD914 is a striking observation, but our data do not distinguish whether elevated cZ drives node initiation or is a secondary consequence of altered shoot architecture. Determining causality would require exogenous cZ application, tissue‑specific cZ measurement during early node development, or genetic manipulation of cZ biosynthetic genes.”

 

6. Are the identified chromosome 18 QTLs stable across environments and backgrounds?The mutant was tested mainly in one genetic background; broader validation is needed.

Reply: Thank you for this constructive comment. We agree that broader validation across additional genetic backgrounds and environments is necessary. In the revised manuscript, we have modified the corresponding sentence in the Discussion to reframe this limitation as a future direction. Specifically, the broader detection of GmGASA32-associated QTLs across diverse genetic backgrounds and environments is a logical next step for soybean breeding.

Revision: Discussion: “Additionally, our analyses were performed using two soybean varieties (HN48 and LSD914) grown under field conditions in a single geographic location. Broader validation of the GmGASA32-associated QTLs across diverse genetic backgrounds and environments is a logical next step for soybean breeding.”

 

7. Please clarify the distinction between the node number and internode length regulation.This is central to the phenotype and should be more explicitly explained in Results and Discussion.

Reply: Thank you. In the revised manuscript, we have now explicitly clarified this distinction in both the Results and Discussion sections. Specifically, we emphasize that LSD914 and HN48 have comparable plant heights (Figure 1C), but LSD914 has ~34% more main stem nodes (Figure 1B).

Revision: 2.1. Phenotypic evaluations of parental and segregating population: “Specifically, HN48 averaged 17.98 nodes, while LSD914 reached 24.08. LSD914 displayed ~34% more nodes without height increase. Because total plant height remained similar, the increased node number in LSD914 is necessarily accompanied by a proportional reduction in average internode length. This indicates that the LSD914 mutation promotes node initiation while suppressing internode elongation, resulting in a compact, high-node architecture.”

Discussion: “A key feature of the LSD914 phenotype is the uncoupling of node number from plant height: more nodes are produced without an increase in total stem length, implying shorter internodes. Thus, the underlying genetic and hormonal changes likely affect both node initiation (positively) and internode elongation (negatively), or they shift the balance between these two processes. ”

 

8. The phytohormone analysis is valuable; however, the interpretation of reduced auxin-related metabolites and hormone crosstalk would benefit from citing recent advances in auxin signaling biology. The recent review 10.1111/ppl.70165 provides comprehensive mechanistic insight into auxin biodynamics and its interactions with other phytohormones, directly relevant to the altered auxin–cytokinin balance proposed in this study. Incorporating this study will strengthen the hormonal framework and better emphasize the significance of your findings on soybean node architecture.

Reply: Thank you for the suggestion. We have now cited the recommended review (Ali et al., 2025, Physiologia Plantarum, doi:10.1111/ppl.70165) in the revised Discussion, specifically to contextualize the reduced auxin-related metabolites observed in LSD914 and to support our discussion of auxin-cytokinin crosstalk in node initiation. We have also added a brief statement connecting our findings to the broader framework of auxin biodynamics and hormone interaction networks.

Revision: Discussion: “Recent advances in auxin signaling biology have highlighted the dynamic nature of auxin biodynamics and its integral role in coordinating plant growth and development through complex crosstalk with other phytohormones [41].”

 

9. Please add more details in the figure legends.

Reply: Thank you for this suggestion. All figure legends have been revised with additional details.  

Revision: See main text for the details about all the revised legends.