Write a comment

In this study, Demont et al. looked into the effect of temperature on plant immune responses, particularly those mediated by nucleotide-binding leucine-rich repeat receptors (NLRs) with an N-terminal Toll/interleukin-1 receptor (TIR) domain. The authors employed Arabidopsis inducible transgenic lines expressing TIR-containing NLRs (TNLs) and isolated TIR domains to assess how temperature influences immune signalling. Their findings reveal that while the immune responses triggered by full-length TNLs are compromised at elevated temperatures (30°C), the signalling pathways initiated by isolated TIR domains remain robust under the same conditions. This suggests that downstream immune responses are more resilient to temperature changes than the initial full length NLR stability and activation. This work highlights the need for further investigation into how temperature impacts stability and activation of NLRs, which is critical for improving plant disease resistance in the context of climate change. It also raises important questions about the molecular mechanisms that confer thermostability to isolated TIR domains and robustness to the TIR-dependent downstream immune signalling.

We thoroughly enjoyed reading the paper, as it provides very valuable insights on TNL immune signalling in the context of global warming and climate change. The story tells a clear message, and the authors took advantage of a nice experimental system with inducible transgenic lines to test relevant and challenging hypotheses.

General comments

• We think that, throughout the manuscript, the authors should define and be more careful about the use of certain terms such as “thermostable” or “thermotolerant”, when referring to proteins, their biochemical properties or signaling pathways. There is a distinction between thermostable, thermoresilient and thermotolerant. The authors should state the definition of the pertinent terms and use them when appropriate.

• Is it common in studies on the effects of temperature to use the terms “permissive” and “non-permissive”? What exactly are the temperatures chosen in the study “permitting” or “not permitting”? The authors should clarify based on why they chose this classification.

• At times, the description of results can be somewhat convoluted and repetitive. We think that a slightly clearer way of describing them might be starting with how the controls behave under the conditions tested, and then building up from there, adding the comparison of the different transgenic lines being tested.

• Overall, we consider that the figure titles could be more assertive to guide the interpretation of results (i.e. Describe the main point that a figure is trying to convey, such as “Expression of EDS1 and genes in the SA pathway are induced by isolated TIRs, even at 30^oC”, in Figure 2).

• One important disclaimer that the authors should consider when making conclusions is that MHV mutations that induce NLR autoactivity might also alter the stability of the activated receptor complex. The overall stability of the effector-activated resistosome might be higher than the stability of an autoactive complex. These differences in stability might confound, to some extent, the effect of temperature on the function of full length NLRs. 

Results

• In Figure 1B, wild type Col0 should be included as a control at the different temperatures tested, with and without Dex, to show that neither temperature nor Dex treatment induce any phenotypes in wild type leaves.

• In the figures, the use of DMSO as negative control should be clearly stated accompanying “- Dex” to make it clearer that an appropriate negative control for the inducible system was used.

• Why did the authors choose to use a different tag for the L6^MHV transgenic lines, compared to L6^TIR, RPS4^TIR, and RPS4?

• If out of the two inducible lines generated for full length RPS4, one shows responsiveness at 21^oC, but the other one does not, it is misleading to choose the one that is responding, as this result is not robust enough. A third line should have been tested to check which phenotype is more commonly observed.

• Based on the previous comment, in line 116: the use of “all generated constructs” can be misleading. Rather specify which constructs are competent for activation of immune response under permissive temperature.

• Transgenic lines expressing TIR-only proteins (TN2 and RBA1) showed autoimmune responses at 30^oC. Is this phenotype also observed at 21^oC?

• Throughout the study, the authors extensively show that the autoimmune phenotypes in the Arabidopsis inducible lines are derived from TIR immune signalling. We think that one key negative control to clearly show that this is indeed the case is to include inducible lines transformed with catalytically inactive mutants of the TIR-only proteins/domains. This would demonstrate that all the phenotypes observed throughout the study are attributable to robust TIR-immune signalling and no other unrelated background responses.

• In Figure 1C, EDS1 expression levels of wild type Col0 under all conditions tested should be included, as well as all the inducible lines without Dex treatment to show that changes in protein levels are attributable to the inducible system.

• In Figure 2, we suggest to also check the expression of PR1 as a classic marker of the SA pathway. Have the authors considered directly measuring SA levels under the conditions tested?

• Some sentences would benefit from being shorter to make a clearer point. For example, line 174, “… at 30 ^oC, EDS1 increased expression was abolished or significantly reduced in seedlings expressing full-length TNLs L6MHV and RPS4, while it was maintained in those expressing L6^TIR and RPS4^TIR” could be rephrased as “EDS1 expression at 30^oC was only enhanced in seedlings expressing TIR-only domains but not full length TNLs”.

• Line 196-197: Please specify this conclusion “Altogether, our results further support that elevated temperature affects NLRs function but not downstream signalling pathways”. In this particular case, the authors are only examining EDS1 expression, and two other SA markers.

• In Figure 3, why did the authors suddenly added SNC1^TIR but stopped using full length RPS4? It is also not clear in this figure the temperature at which the experiment was performed (21^oC or 30^oC ?).

• In line 218, the authors mention that the immunoblot analyses showed that RPS4^TIR generally accumulated at slightly lower levels in the adr1 triple and helperless mutants. However, only the protein accumulation in the helperless mutant under +Dex treatment was visibly lower.

• It is unclear why did the authors switched between L6^MHV-myc and L6^MHV-FSBP in the experiments of Figures 3 and 4.

• In Figure 4, a panel without Dex (- Dex) should be included as negative control, or at least be included as a supplemental figure. Full length L6^MHV in the nrg1 double and adr1 triple mutant backgrounds should also be included in the experiment as negative controls.

• Expression or protein accumulation of NRG1 and ADR1 in the transgenic inducible lines in the different mutant backgrounds could be recorded as supplemental data to confirm that e.g. ADR1 expression is not affected in the nrg1 double mutant transgenic inducible lines.

• In Figure 4, only the ADR1-L2^DV mutant allele was tested directly. Therefore, there is no direct evidence fully supporting the title of this results section, as no other ADR1 alleles were tested. In addition, the fact that it did not result in an autoimmune phenotype at 30^oC might be due to a background effect of the DV mutation (e.g. by destabilising the activated oligomer at this temperature), rather than lack of thermostability of the ADR1-L2 activated signalling complex. We understand that the use of the DV mutation is a proxy to test immune signalling by NLR activation. We consider that the authors should consider and state the limitations of the study and avoid absolute claims such as “Differential temperature sensitivity of ADR1s-mediated signalling”.

• Wild type Col0 should be used as control in Figure 4C.

• In Figure 5, Col0 should be used as control in all conditions tested.

• For the last section of the results, we think that the authors should not make the claim in the title that flg22 treatment enhances TIR-mediated immune signalling. This is misleading, as shown by the authors when testing expression of GFP-MiniTurbo-Flag upon +Dex/+flg22 treatment. Activation of PTI responses does not specifically enhance TIR-mediated signalling, but rather overall protein accumulation. We suggest changing the wording of this results section to avoid making such a bold claim.

We appreciate that the authors considered the limitations of the study at the end of the manuscript. We suggest to include some others highlighted here, and to describe the results of each section taking into account such limitations.

Concluding remarks

We would like to thank the authors for their significant efforts in studying this highly relevant and challenging topic. Investigating the impact of temperature on plant immunity, especially in the context of climate change, is critical for the future of agricultural sustainability. The work presented in this manuscript provides valuable insights into the mechanisms of immune signalling under temperature stress, offering potential pathways for developing more resilient crops. We hope that the suggested revisions will further strengthen the manuscript and help to enhance its clarity and impact. The authors’ dedication to advancing our understanding in this field is greatly appreciated, and we look forward to seeing the final version of this important work.

Competing interests

The authors declare that they have no competing interests.