Dual-Transcriptome Dissection of the Mechanisms Underlying Alfalfa Phenotypic Differences Induced by Two Rhizobial Isolates

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In the manuscript (MS) titled “Dual-transcriptome dissection of the mechanisms underlying alfalfa biomass differences induced by two rhizobial isolates“ Jian Guan and co-authors presented the results of a study on the influence of two strains of Sinorhizobium meliloti G3L13 and WWL2 on the formation of nodules in alfalfa. The authors presented the results of a comparison of gene expression in two strains and showed that strain G3L13 exhibited overall stronger expression of key regulatory and structural genes responsible for nitrogen fixation and associated electron transport/microaerobic respiration modules. For the WWL2 strain, differences in the expression of genes associated with the regulation of nodule formation signals and surface structures, chemotaxis, motility, as well as nutrient transport and assimilation were shown. Overall, the work is interesting and consistent with the scope of the journal Microorganisms.

However, there are a number of criticisms. For example, the introduction should have provided information explaining the reasoning behind the selection of these strains.

Line 25. “genes(ENOD93” add space before the parenthesis

Line 69 and throughout the text. All Latin names of bacteria, including S. meliloti, and genes should be italicized.

Line 91-92. MG575932.1 refers to Sinorhizobium meliloti strain WWL2, while MG575945.1 refers to Sinorhizobium meliloti strain WE2. Why do the authors use the strain code G3L13 instead of WE2 in this manuscript?

Line 174-176. Provide a description of the determination of nitrogenase activity or a reference to the method.

Line 249. For BP, CC, and MF, provide the explanation in the caption.

In Figure 20 (above), where did the reference to variant G9L8 come from? This section, which contains a reference to Figure 20, is talking about bacterial cultures WWL2 and G3L13.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 1

 

Dear Reviewer 1,

 

Thank you very much for your valuable comments and constructive suggestions. We have carefully revised the manuscript accordingly. Below, we respond to each comment point by point. All changes have been highlighted in the revised manuscript for ease of review.

 

 

Comments 1: introduction should have provided information explaining the reasoning behind the selection of these strains

Response 1: We added a brief rationale in the Introduction explaining why these two strains were selected. Under identical experimental conditions, they showed markedly different symbiotic phenotypes, including plant growth and nitrogenase activity, and were used as a high and low efficiency strain pair to analyze coupled host and symbiont transcriptional responses at the mature nodule stage.

 

 

Comments 2: Line 25. “genes(ENOD93” add space before the parenthesis

Response 2: We corrected this spacing issue and checked the manuscript to fix similar cases throughout.

 

 

Comments 3: Line 69 and throughout the text. All Latin names of bacteria, including S. meliloti, and genes should be italicized

Response 3: We formatted all bacterial Latin names (for example, Sinorhizobium meliloti) and gene symbols (for example, nifH and fixK) in italics according to the journal style. We also italicized strain codes to ensure consistency.

 

 

Comments 4: Line 91-92. MG575932.1 refers to Sinorhizobium meliloti strain WWL2, while MG575945.1 refers to Sinorhizobium meliloti strain WE2. Why do the authors use the strain code G3L13 instead of WE2

Response 4: We clarified in the Methods that MG575945.1 corresponds to strain WE2 in the database, and G3L13 was an internal code used for the same isolate. We replaced G3L13 with WE2 throughout the manuscript to avoid confusion and to keep strain naming consistent.

 

 

Comments 5: Line 174-176. Provide a description of the determination of nitrogenase activity or a reference to the method. Line 249. For BP, CC, and MF, provide the explanation in the caption. In Figure 20 (above), where did the reference to variant G9L8 come from? This section, which contains a reference

Response 5: We added a description and an appropriate reference for the acetylene reduction assay used to measure nitrogenase activity in Section 2.

 

 

Comments 6: Line 249. For BP, CC, and MF, provide the explanation in the caption

Response 6: We added the full names of BP, CC, and MF at their first occurrence and clarified them in the relevant figure caption to improve readability.

 

 

Comments 7: In Figure 20 (above), where did the reference to variant G9L8 come from

Response 7: We verified that the G9L8 label was a leftover identifier from an earlier draft and was unrelated to this study. We corrected it to WWL2 in the figure and ensured consistent strain naming throughout the manuscript.

 

 

Conclusion

Thank you for your valuable comments and constructive suggestions. Your suggestions make the topic of the article more prominent and the results more rigorous. We hope our responses and revisions address all concerns raised by the reviewers. Please feel free to contact us if there are additional questions or suggestions. Thank you for your time and consideration.

Reviewer 2 Report

Comments and Suggestions for Authors

attached 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

attached 

Author Response

Dear Reviewer,

Thank you for your valuable comments and constructive suggestions, which have significantly improved our manuscript. Below, we provide a detailed response to each comment. All revisions have been highlighted in the manuscript for clarity.

 

Comments 1: There are some grammatical errors in the whole article.. check and correct, please.

Response 1: Thank you for noting the grammatical issues. We have carefully checked and corrected grammatical errors throughout the manuscript.

 

 

Comments 2: The abstract is overly dense and highly technical, which may reduce accessibility.....Simplify the abstract by reducing jargon and shortening long sentences.

Response 2: Thank you for pointing out issues with the abstract. We have rewritten the abstract following your suggestions to improve clarity and reduce jargon.

 

 

Comments 3: The introduction occasionally repeats similar concepts (e.g., strain differences and dual regulation) without adding new insight. and Research objectives are described narratively but not explicitly stated as clear aims or hypotheses...Reduce redundancy in the background section to improve focus and flow..End the introduction with explicit, numbered research objectives or hypotheses.

Response 3: Thank you for this suggestion. We have rewritten the Introduction by integrating your comments with those of another reviewer to reduce redundancy and to state clear objectives and hypotheses.

 

 

Comments 4: Materials and Methods......The rationale for sample size (n = 3 biological replicates) is not statistically justified. Include a brief justification for biological replicate number (or acknowledge limitations)

Response 4: Thank you for this comment. We have added a brief statement in the Discussion acknowledging the limitation associated with using three biological replicates and the resulting constraints on statistical power.

 

 

Comments 5: Module score calculation, while innovative, lacks sufficient validation.Clarify whether module scores were validated against independent metrics.

Response 5: Thank you for this important suggestion. We agree that the module scoring approach is intended as a descriptive summary of coordinated expression changes and therefore should be accompanied by evidence of biological relevance. In the revised manuscript, we clarified that we performed an internal consistency check (sanity check) using independent readouts obtained from the same biological replicates. Specifically, the direction of module score shifts agreed with independent phenotypic indicators measured at 21 dpi, including nitrogenase activity assessed by the acetylene reduction assay and shoot fresh weight (Figure 1). In addition, RT qPCR measurements of representative marker genes from the corresponding modules supported the same direction as the RNA seq trends (Figure 19). We also explicitly stated the limitation that this check provides internal support but does not constitute external validation on an independent cohort; further validation across additional samples, time points, or independent datasets will be required to assess generalizability. These statements were added to the Discussion section on study limitations.

 

 

Comments 6: Some methodological choices (e.g., reference genome selection for rhizobial mapping) could introduce bias but are not discussed.

Response 6: Thank you for this important comment. We agree that the choice of reference genome for rhizobial read mapping can introduce bias when strains differ in accessory genes or sequence divergence. In this study, we mapped symbiont side reads to the Sinorhizobium meliloti Sm1021 reference genome because it provides a complete and well curated annotation, enables consistent gene identifiers and functional assignment, and facilitates cross sample comparisons under a unified annotation framework; moreover, Sm1021 is widely used as a reference genome for read mapping and comparative analyses in S. meliloti transcriptomic studies. Accordingly, we added a brief statement in the study limitations section acknowledging that mapping WE2 and WWL2 reads to Sm1021 may under represent strain unique genes and may reduce sensitivity for highly diverged loci. We therefore clarify that our rhizobial side interpretations mainly focus on conserved orthologous genes and functional modules with reproducible directional differences across biological replicates, supported by independent phenotypic and RT qPCR evidence. We further note that future analyses using complete strain specific genome assemblies and annotations for WE2 and WWL2, together with strain specific or pan genome based mapping strategies, will be required to evaluate strain specific transcriptional programs and to further assess generalizability.

 

 

Comments 7: Explicitly discuss potential biases introduced by mapping rhizobial reads to Sm1021 rather than strain-specific genomes.

Response 7: Thank you for this important comment. We agree that the choice of reference genome for rhizobial read mapping can introduce bias when strains differ in accessory genes or sequence divergence. In this study, we mapped symbiont side reads to the Sinorhizobium meliloti Sm1021 reference genome because it provides a complete and well curated annotation, enables consistent gene identifiers and functional assignment, and facilitates cross sample comparisons under a unified annotation framework; moreover, Sm1021 is widely used as a reference genome for read mapping and comparative analyses in S. meliloti transcriptomic studies. Accordingly, we added a brief statement in the study limitations section acknowledging that mapping WE2 and WWL2 reads to Sm1021 may under represent strain unique genes and may reduce sensitivity for highly diverged loci. We therefore clarify that our rhizobial side interpretations mainly focus on conserved orthologous genes and functional modules with reproducible directional differences across biological replicates, supported by independent phenotypic and RT qPCR evidence. We further note that future analyses using complete strain specific genome assemblies and annotations for WE2 and WWL2, together with strain specific or pan genome based mapping strategies, will be required to evaluate strain specific transcriptional programs and to further assess generalizability.

 

 

Comments 8: Add a short subsection on statistical assumptions and limitations.

Response 8: We have added Section 4.6, Statistical assumptions and limitations, to address statistical assumptions and study limitations.

 

 

Comments 9: Results......Some sections are excessively descriptive, with limited synthesis.----Certain claims (e.g., “more complete establishment of the nitrogen-fixation program”) verge on interpretation rather than direct evidence.----Figure density is high, which may overwhelm readers. So please Reduce repetition between text, tables, and figure legends.----Clearly distinguish observed results from interpretive statements.---- Consider moving some detailed DEG listings to supplementary materials.---- Improve figure captions to be more self-explanatory.

Response 9: We revised subjective statements that implied a more complete establishment of the nitrogen fixation program. Throughout the Results and Discussion, we replaced interpretive wording, such as strategy, investment, commitment, free living like state, and more complete establishment or efficient stage transition, with descriptions strictly tied to direct observations, including measured nitrogenase activity assessed by the acetylene reduction assay and differential expression patterns of nif, fix, cbb3 and other functional modules. We avoided inferring bacteroid differentiation status or stage transition without direct experimental evidence. In addition, we streamlined and rewrote the text, reduced repetition between the main text, tables and figure legends, moved detailed DEG listings and several figures to the Supplementary Materials, and revised the figure captions to be clearer and more self explanatory.

 

 

Comments 10: Discussion and Conclusion-----Some interpretations imply causality without direct experimental proof. and also The distinction between correlation and mechanism is sometimes blurred.  and Future directions could be more concrete..... my ecommendations in this pont are Tone down causal language and clearly label interpretations as hypotheses.---Explicitly state which conclusions are supported by data and which are inferred.---Propose specific follow-up experiments (e.g., mutant validation, time-course analysis).

Response 10: We revised interpretive statements throughout the manuscript, clarified the distinction between correlation and mechanism, and expanded the future directions with specific follow up experiments, including mutant validation and time course designs.

 

 

Conclusion

Thank you for your valuable comments and constructive suggestions. Your suggestions make the topic of the article more prominent and the results more rigorous. We hope our responses and revisions address all concerns raised by the reviewers. Please feel free to contact us if there are additional questions or suggestions. Thank you for your time and consideration.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear All,

This is a well-designed and technically solid dual RNA-seq study addressing strain-dependent differences in Medicago sativa–Sinorhizobium meliloti symbiosis at the mature-nodule stage (21 dpi). The dataset is extensive, internally consistent, and the integration of host and symbiont transcriptional responses is a clear strength. The manuscript is generally well written and logically organized.

However, the study currently overstates mechanistic causality based on correlative transcriptomic patterns at a single time point. Several interpretations require tempering, and some methodological and conceptual clarifications are necessary before the work can be considered fully robust.

Major comments

1. Over-interpretation of “mechanistic framework”

Pages 19–23; Figures 18–20

The proposed framework coupling host substrate supply/microaerobic homeostasis with rhizobial investment in nitrogen fixation is plausible but remains inferential. All conclusions are derived from transcript abundance at a single developmental stage.

• The manuscript should explicitly frame this as a working or hypothetical model, not a demonstrated mechanism.

• Claims of “investment strategies” and “resource allocation” should be softened unless supported by metabolic fluxes, oxygen profiles, or carbon partitioning data.

Recommendation:
Revise language throughout Results (Sections 3.5) and Discussion (Section 4.5) to consistently emphasize association and coupling, not causation.

2. Single time-point limitation is under-acknowledged

Pages 2–3; Methods and Discussion

Sampling exclusively at 21 dpi is justified, but the manuscript implicitly assumes comparable developmental synchrony between strains.

• The observed transcriptional differences may partially reflect asynchronous nodule maturation rather than divergent strategies per se.

• This issue is mentioned briefly (Discussion 4.4) but should be more explicitly integrated into interpretation, especially for motility/chemotaxis signals in WWL2.

Recommendation:
Add a concise paragraph clarifying how strain-specific developmental pacing could contribute to the observed profiles and how future time-course data would resolve this.

3. Rhizobial read depth and sensitivity

Page 3; Table S4

Rhizobial reads account for only ~1.6–5.0% of total reads.

• While this is typical for nodule dual RNA-seq, the manuscript occasionally draws fine-scale functional distinctions (e.g., between transport systems) that may be sensitive to depth.

• It would be useful to clarify minimum read counts per DEG used for rhizobial analyses.

Recommendation:
Explicitly state any read-count thresholds or robustness checks applied to rhizobial DEGs to reduce false positives.

4. Module score construction needs clearer justification

Page 19; Section 3.5.1; Table 3

The module score approach is interesting but methodologically non-standard.

• The rationale for z-score averaging across predefined gene sets should be more clearly justified.

• It is unclear whether results are robust to alternative normalization strategies or gene-set definitions.

Recommendation:
Briefly justify this approach relative to common alternatives (e.g., GSVA-like logic), and clarify that module scores are descriptive summaries, not inferential statistics.

Minor comments

1. Abstract (lines 24–35):
   Replace “mechanistic framework” with “integrative working model” to avoid overstatement.

2. Methods (Page 4, Section 2.7):
   Clarify why FPKM (host) and RPKM (rhizobium) were retained instead of TPM for downstream visualization.

3. Results (Page 6, lines 229–231):
   The statement that G3L13 hosts are “more transcriptionally active” is vague. Specify which functional domains drive this conclusion.

4. Figures 7–10:
   Consider adding sample size (n) directly in figure legends for clarity.

5. Discussion (Section 4.3):
   The phrase “free-living-like physiological state” should be qualified as relative, given that bacteria are still nodule-resident.

6. Terminology consistency:
   Use either “microaerobic respiration” or “low-oxygen respiration” consistently throughout.

Author Response

Response to Reviewer

 

Dear Reviewer,

 

Thank you very much for your valuable comments and constructive suggestions. We have carefully revised the manuscript accordingly. Below, we respond to each comment point by point. All changes have been highlighted in the revised manuscript for ease of review.

 

 

Comments 1: Over-interpretation of “mechanistic framework”

Response 1: We agree that the current dataset mainly supports associations rather than demonstrated causal mechanisms. We revised the language throughout the Results and Discussion by replacing strong causal or strategic wording, such as mechanistic framework, resource allocation, and investment, with more cautious terms, including integrative working model, working hypothesis, and transcriptional association.

 

 

Comments 2: Single time-point limitation is under-acknowledged

Response 2: We expanded the Discussion to more explicitly acknowledge the limitation of sampling only one time point at 21 dpi. We clarified that strain specific developmental pacing may contribute to the observed profiles and that asynchronous nodule maturation or bacteroid differentiation could partially explain some transcriptional differences. We also stated that future time course sampling combined with nodule staging will be needed to distinguish developmental asynchrony from functional regulation.

 

 

Comments 3: Rhizobial read depth and sensitivity

Response 3: To reduce false positives caused by low read counts in the rhizobial libraries, we filtered out lowly expressed rhizobial genes using edgeR filterByExpr, which applies a CPM threshold adjusted for library size and group size. We then performed differential expression analysis on raw counts using TMM normalization and controlled the false discovery rate.

 

 

Comments 4: Module score construction needs clearer justification

Response 4: We clarified the rationale and interpretation boundary of the module score approach in both Methods and Results. We stated that the scoring scheme is intended as a descriptive gene set summary and is conceptually related to single sample gene set scoring approaches such as ssGSEA and GSVA, with per gene z scoring used to mitigate scale differences among genes. We also clarified that module scores should be interpreted as relative expression tendencies rather than inferential statistics or direct measures of causal investment.

 

 

Comments 5: Abstract (lines 24–35):Replace “mechanistic framework” with “integrative working model” to avoid overstatement.

Response 5: We replaced mechanistic framework with integrative working model in the Abstract and adjusted related wording to avoid overstatement.

 

 

Comments 6: Methods (Page 4, Section 2.7):Clarify why FPKM (host) and RPKM (rhizobium) were retained instead of TPM for downstream visualization.

Response 6: We clarified that differential expression analysis was conducted on raw counts and therefore does not depend on FPKM or TPM. FPKM for the host and RPKM for the rhizobium were retained only for visualization and descriptive summaries, including module score calculation, to remain consistent with the existing workflow. Because FPKM, RPKM, and TPM are monotonically related within the same dataset, the choice does not change relative trend interpretation.

 

 

Comments 7: Results (Page 6, lines 229–231):The statement that G3L13 hosts are “more transcriptionally active” is vague. Specify which functional domains drive this conclusion.

Response 7: We revised the vague statement more transcriptionally active by specifying the functional domains driving the observed differences, such as transmembrane transport, redox and heme related functions, and pathways related to nodule development and microaerobic homeostasis, and we supported these statements with the corresponding enrichment and DEG category results.

 

 

Comments 8: Figures 7–10:Consider adding sample size (n) directly in figure legends for clarity.

Response 8: We added the sample size per group and the presentation format, including mean and SEM, directly in the relevant figure legends.

 

 

Comments 9: Discussion (Section 4.3):The phrase “free-living-like physiological state” should be qualified as relative, given that bacteria are still nodule-resident.

Response 9: We revised this phrase to a more qualified expression to avoid implying that nodule resident bacteria are equivalent to free living cells.

 

 

Comments 10: Terminology consistency:Use either “microaerobic respiration” or “low-oxygen respiration” consistently throughout.

Response 10: We standardized the terminology across the manuscript and consistently used microaerobic respiration.

 

 

 

Conclusion

Thank you for your valuable comments and constructive suggestions. Your suggestions make the topic of the article more prominent and the results more rigorous. We hope our responses and revisions address all concerns raised by the reviewers. Please feel free to contact us if there are additional questions or suggestions. Thank you for your time and consideration.

Reviewer 4 Report

Comments and Suggestions for Authors

Dual-transcriptome dissection of the mechanisms underlying alfalfa biomass differences induced by two rhizobial isolates.

Consider to change the word “biomass” by “phenotypic differences” in title.

Abstract

Nothing about biomass differences is said.

Keywords are important describing words for the study. However, it is suggested to used different words from that in title.

Introduction

Consider that introduction must provided all the relevant knowledge for readers to understand, to see the extend of what is presented in the study. Introduction is too short.

Materials and methods

As general rule, this section must present all the detailed information for the study to be replicated by a competent fellow. Consider this for each section of materials and methods.

Regarding section. 2.1 Plant material and rhizobial inoculation

What type of sand? Was it sterilized? The nitrogen-free nutrient solution, was self-made, or was a commercial product?

Why not a control group was stablished?  A culture without inoculation.

See comments in MS.

Results

See comments in MS

Discussion

OK

Conclusion

Must be rewritten: conclude clearly on each hypothesis presented at the end of introduction.

General comment.

Consider that the title as it is do not reflect the extend of the study. It is suggested to change the word “biomass” by “phenotypic differences”

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

Thank you for your valuable comments and constructive suggestions, which have significantly improved our manuscript. Below, we provide a detailed response to each comment. All revisions have been highlighted in the manuscript for clarity.

 

Comments 1: Consider to change the word “biomass” by “phenotypic differences” in title. Consider that the title as it is do not reflect the extend of the study. It is suggested to change the word “biomass” by “phenotypic differences”

Response 1: Thank you for your suggestion. We have revised the title accordingly.

 

 

Comments 2: Nothing about biomass differences is said.

Response 2: We replaced the term biomass with phenotypic differences in the title and throughout the manuscript; therefore, the manuscript no longer uses biomass terminology.

 

 

Comments 3: Keywords are important describing words for the study. However, it is suggested to used different words from that in title.

Response 3: We have revised the keywords and replaced terms that duplicate the title.

 

 

Comments 4: Consider that introduction must provided all the relevant knowledge for readers to understand, to see the extend of what is presented in the study. Introduction is too short.

Response 4: We have rewritten the Introduction to provide sufficient background and to better present the scope of the study, integrating your suggestions with those of another reviewer.

 

 

Comments 5: As general rule, this section must present all the detailed information for the study to be replicated by a competent fellow. Consider this for each section of materials and methods. Regarding section. 2.1 Plant material and rhizobial inoculation What type of sand? Was it sterilized? The nitrogen-free nutrient solution, was self-made, or was a commercial product? Why not a control group was stablished?  A culture without inoculation.

Response 5: We revised Section 2.1 and the Materials and Methods to provide the detailed information needed for replication, including substrate type and sterilization, preparation of the nitrogen free nutrient solution, and the rationale for not including an uninoculated control.

 

 

Comments 6: See comments in MS.

Response 6: We added strain name labels to the alfalfa photographs as requested.

 

 

Comments 7: Must be rewritten: conclude clearly on each hypothesis presented at the end of introduction.

Response 7: We rewrote the Conclusions to clearly address each hypothesis presented at the end of the revised Introduction.

 

 

Comments 8: How old were the alfalfa cultures? How many days after inoculation?

Response 8: We clarified the seedling growth stage at inoculation and specified the sampling time after inoculation in Section 2.1 to ensure the information is sufficiently detailed.

 

 

Conclusion

Thank you for your valuable comments and constructive suggestions. Your suggestions make the topic of the article more prominent and the results more rigorous. We hope our responses and revisions address all concerns raised by the reviewers. Please feel free to contact us if there are additional questions or suggestions. Thank you for your time and consideration.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

accept

Comments on the Quality of English Language

attached 

Reviewer 3 Report

Comments and Suggestions for Authors

Dear All,

The revised manuscript has been substantially improved and now appropriately reflects the strength and limits of the dataset. The authors have responded carefully and thoroughly to all previous comments. In particular, they have:

• Replaced overstated mechanistic language with appropriately cautious terminology (e.g., “working model,” “transcriptional association”).

• Explicitly acknowledged the single time-point limitation (21 dpi).

• Clarified rhizobial read filtering strategy and edgeR normalization.

• Clearly defined module score construction and its descriptive nature.

• Standardized terminology (e.g., “microaerobic respiration”).

• Added biological replication information directly in figure legends.

The manuscript now presents a coherent, data-supported interpretation without overstating causality.

 

Major Issues Previously Raised – Now Resolved

Overinterpretation of “mechanistic framework”

The authors replaced causal wording with appropriately cautious expressions such as:

• “working hypothesis”

• “transcriptional association”

• “coordinated functional emphasis”

The Discussion (Section 4.5) now explicitly uses a tiered framework (data facts → functional inference → phenotype correspondence), which is scientifically appropriate.

* Concern resolved.

Single time-point limitation

The limitation is now clearly acknowledged in:

• Section 4.4 (Lines c.a. 750 - 759)

• Section 4.6 (Statistical assumptions and limitations)

The authors explicitly state that interpretations are limited to transcriptional associations at the sampled stage and that asynchronous nodule development cannot be excluded.

* Concern resolved.

Rhizobial read depth and sensitivity

Low-count genes are now filtered using , and normalization (TMM) is clearly described (Section 2.4). The manuscript explicitly notes that bacterial mapping rates (1.64–5.04%) reflect sample composition rather than sequencing quality.

* Concern resolved.

Module score justification

Section 2.7 now clearly states:

• Module scores are descriptive summaries.

• They are conceptually related to ssGSEA/GSVA.

• They should not be interpreted as inferential statistics or causal “investment”.

* Concern resolved.

Terminology and clarity improvements

• “Free-living-like physiological state” has been qualified.

• “Investment” was replaced with “expression emphasis.”

• “Mechanistic framework” replaced by “working framework.”

* Concern resolved.

Scientific Merit

This study provides:

• A robust dual RNA-seq comparison of two rhizobial isolates.

• Clear phenotypic divergence (shoot biomass + ARA).

• Coherent cross-partner transcriptional alignment.

• A testable working model linking host substrate supply and rhizobial nitrogen fixation programs.

The integration of host–symbiont transcriptomes is methodologically sound and biologically meaningful.

The manuscript now avoids overreach and remains within the bounds of transcriptomic inference.

Figures and Presentation

• Figures 1–8 are clear and statistically annotated.

• PCA (Figure 2) appropriately shows replicate clustering.

• Volcano plots and heatmaps are properly interpreted.

• Module scores (Table 3) are clearly described as descriptive metrics.

Supplementary material is extensive and well organized.

Minor Editorial Note

Language is now clear and readable. Only minor stylistic polishing could be applied at proof stage (e.g., reducing sentence length in some Discussion paragraphs), but no further revision is required at peer-review level.