PREreview del Single-nucleus transcriptome analyses uncover a dynamic transcriptional landscape of soybean roots in response to soybean cyst nematode infection

Summary

This preprint presents a single-nucleus RNA sequencing analysis examining plant-pathogen interactions, specifically focusing on the development of feeding structures (syncytia) in soybean roots during infection by soybean cyst nematode (SCN). The authors analyse gene expression changes across three time points (1, 3, and 7 days post-infection) to understand the molecular mechanisms underlying syncytium formation. The study provides useful insights into transcriptional changes during infection of resistant and susceptible soybean cultivars, but it could be further strengthened with additional experimental validation, methodological clarification, and improved presentation.

Major Remarks

Experimental validation: Several conclusions drawn from this analysis would be further enhanced by experimental validation. The manuscript mentions spatial transcriptomics (lines 118, 327, 332, and 334) but no experimental details are provided. For example, Glyma.18G022400 is identified as enriched in epidermal cells (Figure 3b), and it would add significant strength if this finding were supported by experimental evidence.

Similarly, the claim about syncytial cell development from procambium cells refers to classical histological and ultrastructural studies (line 272), but these are not shown. We recommend including:

• Cell lineage tracing experiments to directly demonstrate the procambium-to-syncytium differentiation pathway.

• Histological validation of cell type assignments would provide strong support for key conclusions.

Methodological details and consistency: The integration of different cell clustering methods shows discrepancies in cell type assignments (Figure 2). Readers would benefit from a clear explanation of how these methods were reconciled and which was ultimately used for downstream analysis and Figure 2e. Additionally, adding clarity on the following points would improve transparency:

• Normalisation procedures for gene expression analysis.

• Statistical methods for differential expression analysis.

• References supporting the selection of 1, 3, and 7 dpi as sampling timepoints (lines 104–106).

• Additional background on resistant and susceptible cultivars, particularly the timing of necrosis in resistant cultivars after syncytium formation. Does this timing coincide with sample collection timepoints? The only background states necrosis occurs "shortly after" syncytium formation (line 63).

Clarity and presentation: Some figures are challenging to interpret without extensive cross-referencing to the main text. Enhancing figure legends with essential details (axes, abbreviations, experimental conditions) would make them more self-contained. For example, Figure 4 is particularly difficult to interpret without additional context, and naming conventions (gene vs. locus identifiers) should be standardized for consistency (e.g. Figure 3).

Minor Remarks

Latin names for soybean cyst nematode Heterodera glycines are inconsistent between lines 48 and 83.

Supplementary Figure 8 is referenced but not included. Similarly, supplementary data for Figure 5e is mentioned in the legend ("experiments performed 3 times with similar results") should be provided.

Figure-Specific Comments

Figure 1

• In the figure legend, it would be helpful to describe how acid fuchsin stains SCN for Figure 1a as this is not mentioned in the methods.

Figure 2

• Including the unassigned cell cluster UMAP alongside assigned cell cluster UMAP would make comparisons easier.

• Separate colour legends for Figure 2b-d versus 2e, as "unknown" assignments only apply to specific panels.

• Expand the figure legend to specify how Figure 2e was generated: How were different clustering methods integrated, or which method was used for cluster assignment? Address how discrepancies across clustering methods to assign cell types were handled.

Figure 3

• Standardising the dot plot scales, specifically dot sizes, across panels to avoid misinterpretation.

• Clarify terms "expression proportion" and "average gene level".

• Explain normalisation approach for gene expression.

• Addition of statistical analysis to compare expression pattern differences between cultivars and timepoints.

• Use consistent gene naming conventions throughout.

Figure 4

• Provide additional context for the upset plot (Figure 4a), as its interpretation is not obvious.

• Correct figure legends: Figure 4b should read "...between PI 88788 and Williams 82..." and Figure 4c should read "...between Forrest and Williams 82..."

• Gene ontology analysis is mentioned in lines 216-217 but not shown.

• Boxplots may better show data distribution than histograms for Figures 4d-g.

• Figures 4d-g are not mentioned in the text, and their relation to the rest of the figure is unclear.

Figure 5

• Clarify statistical methodology in the Figure 5d legend. The authors mention a two-way ANOVA when multiple group comparisons may be more appropriate.

• Provide supplementary data for Figure 5e as mentioned in the legend ("experiments performed 3 times with similar results").

• Include WGCNA analysis in the figure as mentioned.

• Clarify how depth perception is achieved for Figure 5f.

Figure 6

• Replace "re-differentiation" with more accurate terminology (e.g., "diverted differentiation pathway").

• Supporting the procambium-to-syncytium development claim with experimental data would be very valuable.

• Add UMAP axis labels to Figure 6b.

Figure 7

• Address the discrepancy between pseudotime inference and infection timepoints, since they do not align confidently.

Recommendations

1. Broaden the conceptual framework: Consider framing the work more broadly around how pathogens alter plant development, including discussions of gall and nodule formation in other plant-microbe interactions for comparison.

2. Expand background information: Providing additional background on resistant and susceptible cultivars and citing references for the rationale behind timepoints used for sample collection would aid readers new to the topic. A diagram of syncytium feeding structure in soybean roots in the introduction or Figure 1 would also be a valuable addition.

3. Add validation experiments: Complementing bioinformatic findings with experimental validation (e.g. histology, lineage tracing and spatial transcriptomics) would strengthen the study considerably.

4. Improve figures: Ensuring figures can be interpreted independently through comprehensive legends and standardised formatting would improve accessibility.

5. Clarify methods: Include more detail for clustering integration, normalisation, and statistical analysis.

6. Supplementary data: Include all referenced supplementary figures.

Conclusion

This research provides a valuable transcriptomic dataset of soybean response to SCN, particularly regarding how pathogens reprogram host development. Through this analysis, the authors identify novel target genes that show potential for developing SCN-resistant soybean cultivars, highlighting the advantages and importance of single-cell transcriptomics in plant-pathogen research. Addressing points related to validation, methodological clarity, and presentation would enhance the impact and ensure the findings are accessible and convincing to a broad readership.

Competing Interests

The authors declare no competing interests.

Competing interests

The authors declare that they have no competing interests.