Reviewers: Matthew Helm1, Sara Allen1, Ariana Myers2, Hana Zand Karimi3, Carolina Mazo-Molina4, and Steven R. Scofield1, 2USDA-ARS, Crop Protection and Pest Control Research Unit, West Lafayette, INDepartment of Agronomy, Purdue University, West Lafayette, INDepartment of Biology, Indiana University, Bloomington, INBoyce Thompson Institute for Plant Science Research, Ithaca, NY
Reviewers: Matthew Helm1 and Morgan E. Carter21. USDA-ARS, Crop Protection and Pest Control Research Unit, West Lafayette, IN2. Plant Pathology and Plant-Microbe Biology Section, School of Integrative Plant Science, Cornell University, Ithaca, NY SummaryThis is a review of Mazo-Molina and Mainiero, et al. bioRxiv (doi: https://doi.org/10.1101/2020.03.05.979484) posted on March 5, 2020. This study builds upon previous work that described the serendipitous discovery of a locus in the distantly related relative of tomato Solanum lycopersicoides, known as Pseudomonas tomato race 1 (Ptr1) (Mazo-Molina et al., (2019). Mazo-Molina and colleagues (2019) previously showed the Ptr1 locus confers resistance to several Pseudomonas syringae pv. tomato race 1 strains, but not the race 0 strain DC3000. Comparative genomic analysis of the effector repertoire between a race 1 strain of P. syringae pv. tomato, T1, and DC3000 identified AvrRpt2 as the cognate effector recognized by the Ptr1 locus. In Arabidopsis, AvrRpt2 is a cysteine protease that cleaves RIN4 and, upon cleavage, leads to activation of the Arabidopsis NLR protein RPS2. In this paper, Mazo-Molina and Mainiero et al. describe the identification and functional characterization of the Ptr1 gene. A single recombinant among 585 F2 S. lycopersicoides individuals segregating for the Ptr1 locus was identified that narrowed the Ptr1 candidate to eight NLR genes. By employing currently available gene models for S. lycopersicoides and 3’ RNA-seq data, Mazo-Molina and Mainiero et al. narrowed the list of Ptr1 candidates to three genes; A, B, and D. Candidate D was shown to be a pseudogene as it contained multiple mutations that disrupted the reading frame and was, therefore, not included in their functional analyses. Co-expression of candidate A with either AvrRpt2 or RipBN in N. glutinosa induced an HR-like cell death response. Candidate B, however, did not induce cell death by itself or when co-expressed with either AvrRpt2 or RipBN, and was thus not considered to be the Ptr1 gene. Taken together these experiments demonstrated that candidate A is the Ptr1 gene. Moreover, they show Ptr1 is conserved in many species within the Solanaceae family and that the Ptr1 ortholog from N. benthamiana and potato also mediate recognition of AvrRpt2 and RipBN. Lastly, phylogenetic analyses indicate that Ptr1, RPS2, and Mr5 (an NLR gene from apple whose protein product recognizes AvrRpt2) are not orthologous. The authors thus conclude the ability to recognize AvrRpt2 protease activity evolved convergently. CommentsResults section – LA4277 genome. In the first paragraph of the results section, the authors discuss NLR-encoding genes in the S. lycopersicoides genome sequence as well as in LA4277-R. Later, it was mentioned that the LA4277 genome was assembled as part of this paper. We suggest clarifying at the first introduction of the LA4277 genome features that this genome was sequenced in this study, for readers unfamiliar with available genomes. Results section—cell death assays. To test whether AvrRpt2 causes degradation of SlRin4 proteins resulting in activation of candidate A, the authors carried out cell death assays (Figure 2) by transiently co-expressing candidate A along with SlRin4-3 and AvrRpt2 or RipBN in N. glutinosa. However, in Figure 2A, they show transient co-expression of candidate A with either AvrRpt2 or RipBN is sufficient to induce an HR like cell death response, suggesting candidate A is guarding an endogenous RIN4-like protein in N. glutinosa. Did the authors try silencing the endogenous RIN4 protein in N. glutinosa and then co-express AvrRpt2/RipBN, candidate A, and SlRin4-3 to test whether SlRin4-3 has a role in AvrRpt2 recognition? We would also suggest performing an electrolyte leakage assay to quantify the cell death responses in Figure 2C. Results section—Figure 2C. In the experiment depicted in Figure 2C, what was the rationale behind including only SlRin4-3 and not the other additional Rin4 proteins? Candidate B. Based on the immunoblot analysis in supplemental figure S2, candidate B protein expression is considerably weaker than that of candidate A protein expression (Figure 2D). Could it be a formal possibility that candidate B recognizes AvrRpt2 and RipBN, but because of the weak protein accumulation, there is no observable cell death in N. glutinosa? Discussion section. As part of a very thorough discussion section, the second paragraph on page 10 covering RIN4 structure did not seem as relevant to the findings in this paper, as detailed analysis of Ptr1 detection of RIN4 cleavage was not experimentally investigated.
SummaryThis is a review of Thomas et al. bioRxiv manuscript (DOI: https://doi.org/10.1101/813758) posted on October 21, 2019. In this manuscript, the authors present evidence demonstrating that transgenic tomato lines expressing the Nicotiana benthamiana NLR protein Recognition of XopQ 1 (Roq1) display immunity to three bacterial pathogens of tomato: Xanthomonas perforans 4B, Xanthomonas euvesicatoria 85-10, and Pseudomonas syringae DC3000. Importantly, Xanthomonas XopQ (DXopQ) and P. syringae HopQ1 (DHopQ1) deletion mutants were able to colonize leaf tissue of both wild-type and Roq1-expressing tomato, demonstrating that this resistance response is dependent upon the expression and recognition of XopQ and HopQ1. Furthermore, phylogenetic analysis revealed that Ralstonia solanacearum, the causative agent of bacterial wilt in tomato, also contains a functional homolog of XopQ and HopQ1, known as RipB. Using an Agrobacterium-based transient expression assay in Nicotiana tabacum, the authors confirmed that Roq1 mediates recognition of multiple R. solanacearum RipB alleles (RipBGMI1000 and RipBMolk2), and such Roq1-dependent recognition suppresses R. solanacearum growth in transgenic tomato. Collectively, these data suggest an NLR immune receptor isolated from N. benthamiana is capable of conferring resistance to three different bacterial pathogens. Comments Results section—tomatoes expressing Roq1 are resistant to Xanthomonas and P. syringae. In this paper, the authors generated homozygous transgenic tomatoes that are reported to express Roq1. It is unclear, however, whether the transgenic tomato lines are indeed expressing Roq1, and whether expression of Roq1 is correlated with the observed immune response to Xanthomonas and Pseudomonas (Figures 1 and 2) and Ralstonia (Figure 5). I suggest including data confirming that Roq1 is indeed expressed in the transgenic lines and also whether Roq1 expression is correlated with resistance. Results section—Expression of Ro1 confers resistance to Xanthomonas perforans in the field. Though there was no significant increase in total marketable yield of Roq1-expressing plants, were there any additional adverse effects of Roq1 expression on plant development or any other agronomic traits? Results section—The R. solanacearum RipB effector, a homolog of XopQ/HopQ1, is recognized by Roq1. Here, the authors show that transient co-expression of Roq1 with either XopQ, RipBGMI1000, or RipBMolK2, induces a cell death response in N. tabacum. · Why was HopQ1 not included in the transient expression assay shown in Figure 4? · Does Roq1 preferentially associate with and induce a stronger HR when expressed with a particular effector? In other words, does transient co-expression of Roq1 with XopQ/HopQ1 induce a greater cell death response than Roq1 co-expressed with the RipB alleles? (An electrolyte leakage assay and co-IP could be used to quantify the cell death response between the various effector proteins and test for physical association, respectively).· Do the authors have any hints as to how Roq1 is able to mediate recognition of multiple bacterial effector proteins? · Do the authors have any hints as to the subcellular localization of Roq1 and which Roq1 domains (or domain fusions) have a functional role in recognizing these effectors? Results section—Roq1 tomatoes are resistant to R. solanacearum containing RipB. · It is unclear why the R. solanacearum DripB mutant was not used as a control in Figure 5B. Including this control would demonstrate that Roq1-mediated recognition of R. solanacearum is indeed dependent upon the expression of RipB in tomato· I would also suggest that the authors confirm the transgenic tomato lines are indeed expressing Roq1 and also test whether expression correlates with resistance as mentioned previously above. ReviewerMatthew Helm (USDA-ARS, Crop Protection and Pest Control Research Unit, West Lafayette, IN)