PREreview of Alternative splicing of a potato disease resistance gene maintains homeostasis between development and immunity

Alternative splicing of a potato disease resistance gene maintains homeostasis between development and immunity, and functions as a novel process for pathogen surveillance

By Anam Siddiqui^1*, Mauricio P. Contreras^1*, Shona Strachan^1* and Yufei Li^*1

1: The Sainsbury Laboratory, University of East Anglia, Norwich, UK.

*All authors contributed equally to the review, authors are ordered in alphabetical order according to their first names.

This article delves into the regulatory mechanisms that govern the activity of NLR immune receptors during pathogen infection. In this study, the authors study the potato NLR gene RB (aka Rpi-blb1) and its cognate effector, IPI-O1 from Phytophthora infestans. They report that RB exhibits alternative splicing (AS) to produce two transcriptional isoforms that modulate the balance between plant immunity and growth. Intriguingly, in the absence of pathogen infection, the RB gene primarily exists as the intron-retained isoform, RB_IR, encoding a truncated non-functional protein, ensuring unhindered plant growth. Upon late blight infection triggered by Phytophthora infestans, a transition occurs whereby the intron-spliced isoform RB_CDS becomes predominant, encoding a functional R protein that mediates disease resistance. Using a luciferase reporter system, they pinpoint the role of the pathogen effector IPI-O1, which directly interacts with the potato splicing factor StCWC15, in facilitating this alternative splicing event. This provides a novel mechanism for effector recognition in plants. This remarkable finding underscores the intricate mechanisms plants employ to modulate their immune responses, aligning with a balance between growth and immunity. Overall, the manuscript is well-written, with a strong foundation in previous research, and presents a compelling narrative. The conclusions are well supported by the data presented and the findings represent a conceptual advance for the field.

After reading this preprint, we have a few comments and suggestions that we hope may be of use to the authors. We believe some of our suggestions could make the manuscript more compelling and accessible to the reader. Nonetheless, the manuscript is already of very high quality.

General comments

• Some of the figures are quite information dense, with several subpanels, and contain a lot of information. Would it be possible to split up some of the figures or to move some of the information to supplementary materials?

• In general, while the evidence for alternative splicing and the experiments showing accumulation of transcripts are super convincing and backed up with a lot of functional assays (HR, infection with P. infestans, etc.), if feasible to check, it would be nice if protein levels could be checked, not just cDNA. Does transcript abundance result in increased translation/protein accumulation levels?

Figures

• Figure 1

  • Would it be possible to make the asterisks red (or any other more visible colour) in Figure 1 panel D? The black asterisks are a bit hard to see.

  • Would it be possible to add a ladder to the RT-PCR gel?

• Figure 2

  • In panel A, while the internal controls look convincing, the levels of RNA extracted for the P. infestans treatment look lower than P. capsici and the mock control - is there a reason for this, is there another representative gel from a different rep which could be used? Is this a reproducible and consistent phenomenon across reps?

• Figure 3

  • In panel B, it would be helpful to add the ID labels (RB, RB_CDS, etc.) to the bottom of the boxplot to allow interpretation of the results.

  • It would be nice to have data on the expression levels as a supp figure. Ideally, information on protein levels would be amazing.

•  Figure 4

  • In panel A, it would have been ideal to test silencing of these spliceosome core components with another R-gene unrelated to RB to determine if silencing spliceosome components such as U1-70K or SF3B1 specifically affects RB or whether it generally has an impact on other R proteins or on immune signaling in general.

• Figure 5

  • In panel A, the schematic could be made easier to understand, maybe by removing the Exon 2 which isn’t present in the top construct (makes it easier to understand that it isn’t expressed and so no luciferase activity)

  • Narratively from the beginning of the paper it is eluded to that the authors know what is causing the alternative splicing, but it isn’t until fig 5 that the experimental proof is given, IPI-O1 is mentioned in fig 4 but isn’t explained until fig 5

  • Are the SRE screened present in the JH19 isolate or were they chosen because the were previously published to be SREs in P. infestans?

  • IPI-O1 is referred to as SRE8 in panel B and IPI-O1 in panel C. It would perhaps be better to keep the naming consistent and stick to IPI-O1?

• Figure 6

  • Maybe the RT-qPCR results could be added as a supplementary figure, to support the splicing ratios in panel F? 

  • The figure legend doesn’t reflect what is in the figure panel (text for some of the panels is missing and there is text that refers to the wrong figure). Figure legend needs to be revised.

• Figure 7

  • Although multiple orthogonal techniques have been used to confirm this interaction, it would’ve been nice to have another control which wasn’t EV.

  • Related to the previous comment, it would be nice to have a treatment in the Y2H or the CoIP experiments which can confirm that IPI-O1 is not a sticky protein. For example, IPI-O1 co-expressed with another unrelated protein. Having said that, we appreciate the inclusion of another effector, Pi04314, as a stringent control.

  • The figure legend does not reflect what is in the figure panel (text for some of the panels is missing and there is text that refers to the wrong figure). Figure legend needs to be revised.

• Figure 8

  • Would’ve been good to add a wildtype infection assay plant to panel A/B (as was done in Figure 9A). This is not crucial though.

  • Showing the complementation with the synthetic/codon-scrambled variantwas superb! This sort of approach would have been a good addition to the experiment in Figure 4a.

  • The figure legend does not reflect what is in the figure panel (text for some of the panels is missing and there is text that refers to the wrong figure). Figure legend needs to be revised.

Concluding remarks:

All in all, we thoroughly enjoyed reading this paper. The narrative was clear and easy to follow. We particularly enjoyed the inclusion of a working model at the end! Some additional thoughts that arose during reading:

• It would be nice to expand further on this new proposed effector recognition mechanism and how it differs from other previously described strategies (direct, guardee, decoy, etc.). This is a new concept and should be highlighted more!

• In the discussion, the authors mention that “it remains to be tested whether the truncated CC domain competes with the CC domain from full-length RB to form an inactive oligomerized resistosome during normal growth in order to avoid autoactivation due to RB self-oligomerization.” Have the authors considered co-expressing RB-IR simultaneously with RB-CDS along with IPI-O1 in N. benthamiana to test if the truncated form has an effect on RB-CDS signaling? This may be slightly off-scope but would be very straightforward to test and may add some clarity regarding any potential negative regulatory roles of RB-IR on RB-CDS.

• You may want to make it clearer that the binding of CWC15 with IPI-O1 is required/essential for the pathway to be initiated. The authors mention the re-localisation to the nucleolus (and the loss of relocalization in the mutant) but could go a step further and discuss how they speculate this could be necessary for the interaction/recognition.

• Zhao et al., (2021, Plant Commun.) previously reported that the P. infestans effector IPI-O4 can block IPI-O1 recognition by RB. Part of this suppression mechanism involves IPI-O4/IPI-O1 association, which is hypothesized to block IPI-O1 avirulence/RB recognition. While this is completely outside the scope of this paper, we were wondering if perhaps IPI-O4 association with IPI-O1 also interferes with the latter effector’s interaction with splicing machinery components that are required for activation of RB splicing and subsequent recognition.

Competing interests

The author declares that they have no competing interests.