PREreview of ARalstonia solanacearumtype III effector alters the actin and microtubule cytoskeleton to promote bacterial virulence in plants

PreReview of “A Ralstonia solanacearum type III effector alters the actin and microtubule cytoskeleton to promote bacterial virulence in plants”.

By Amelia H. Lovelace and Mauricio P. Contreras

This article by Hiles & Rogers et al. examines the mechanisms by which type III effector proteins (T3Es) employed by microbial pathogens, specifically Ralstonia solanacearum (Rs), interact with and manipulate the cytoskeleton in plant cells to promote disease. Rs is a soil borne bacterial pathogen and the causal agent of bacterial wilt on several economically important crops. Previous studies have determined that Rs T3Es localize to diverse organelles in Solanum hosts; however, many of these T3Es have not been characterized. This study focuses on the conserved Rs effector RipU^K60 and its impact on the plant host actin and microtubule cytoskeleton, both of which are crucial for cellular homeostasis and immune signaling. The researchers demonstrate that RipU^K60 associates with both actin and microtubules resulting in disruption of their organization, and suppresses key immune responses such as reactive oxygen species production and callose deposition. This cytoskeletal manipulation by RipU^K60 facilitates Rs colonization and virulence, as shown by decreased pathogenicity in Rs mutants lacking RipU^K60. This study advances our understanding of the cytoskeleton's role in immune response and pathogen recognition, and underscores the sophisticated strategies pathogens like Rs utilize to undermine plant defenses, emphasizing the dynamic interplay between host cytoskeleton and pathogen effectors. The manuscript is well-constructed and grounded in prior research, presenting a clear and persuasive narrative. We have a few suggestions that we think could elevate the manuscript even further, although it is already of high quality.

Figure 1: 

-Could the authors clarify what cut-off was used for the Pearson’s correlation coefficient calculation to determine co-localization? Could the p-values for each Pearson’s coefficient calculation be reported?

-It would be very useful if a positive control for association with microtubules or actin be included in the analysis and compared with the effectors via Pearsons’s correlation coefficient.

Figure S1:

-It would be great if a loading control of some sort could be included to show that the fractionation did indeed work?

Figure 2:

-It would be really nice if RipBD could be included as an additional control, given that in Figure 1 it was shown that RipBD does not co-localize with actin and microtubules. 

-Could an asterisk be used to indicate the size of the mature effector? There are multiple bands and it is not immediately clear to the reader which one corresponds to full-length RipU.

Figure 3:

-Could a control with GFP overexpression be used rather than EV in ROS assays? Additionally, although not as crucial, could a control with just buffer be included to determine the effect of agroinfiltration triggered PTI on overall ROS accumulation?

-In panel B, would it be possible for the authors to quantify the callose deposition from the images taken using imaging analysis? 

-Anti-HA western blots would be helpful to confirm that the protein is indeed accumulating in the DEX inducible lines upon DEX treatment.

Figure 4:

-Line 167 “Intriguingly, RipUK60-GFP was unable to suppress BAX-mediated cell death, suggesting that this effector is not likely to function as a general suppressor of cell death (Supplemental Figure 2).”

We were curious about what the rationale is for using the mammalian protein BAX? 

Would it be possible to include cell death triggered by a plant derived protein such as a CC or TIR-NLR mediated cell death?

An effector from a plant pathogen is more likely to have evolved to suppress cell death induced by a plant protein.

-Panel A, the gray shading on the points is very faded and difficult to see. A box plot with data points would be easier to interpret.

-Is it correct to use two Wilcox tests comparing only to the mutant to show statistical differences between strains? Maybe a 1-way ANOVA to compare across all strains within each timepoint and within each genotype would be more appropriate, rather than just comparing WT and the complemented strain to the mutant.

-How were the error bars calculated in panel 4B considering the disease index is a categorical variable? Also, the lines vs. Dashes used to distinguish between WT, RipU mutant and the complemented strain are not easy to see, maybe colors could be used instead?

Figure S3:

-Visualization of the log2fold change for the genes in question would be useful so that the readers can assess how up or downregulated they are in susceptible vs. resistant tomato lines?

Figure 5:

-Could the authors change the plots according to the recommendations made for Figure 4. In general it would be nice to use colors to distinguish Rs K60 vs Rs K60 + chemical. Additionally, could the legend clarify if the control treatment is Rs K60 + DMSO?

Figure 6:

A chemical disruptor would be an amazing positive control in these experiments.

Figure 7:

-Similar to Figures 4 and 5, it would be great if the authors make use of colors to better distinguish different treatments and/or use boxplots instead. The data points could be rendered in a solid color to facilitate visualization.

Discussion:

 -While the discussion is well written, it reads a bit like a review in some parts. Particularly the section on “What is the function of RipU^K60-induced cytoskeletal changes”. The section “RipU^K60 is unusual among T3Es that impact the cytoskeleton” does not support this claim in that RipU was not discussed much at all in the context of other bacterial T3Es. The authors spend most of their time describing the mechanisms of other effectors. We thought the discussion could benefit from some shortening/sharpening.

General thoughts and comments that are beyond the scope of this paper:

-Could a comparison be made between RipU and other cytoskeleton perturbing effectors mentioned in the discussion? Is there a common theme? Perhaps in terms of fold, etc.? Alphafold could be used to predict their structure and compare. As Rs is a xylem-colonizing pathogen, does cytoskeletal disruption occur in these tissues and is the immune suppression activity of RipU tissue-specifc?

Competing interests

The author declares that they have no competing interests.