PREreview of Two NLR immune receptors acquired high-affinity binding to a fungal effector through convergent evolution of their integrated domain

Abstract

This is a review of Białas et al. bioRxiv doi: https://doi.org/10.1101/2021.01.26.428286 posted on January 27, 2021. In this paper, the authors study the HMA domain from two NLR immune receptors. The authors identify a common ancestral version for these domains and highlight a convergent evolution that allows the interaction with a common effector.

Summary

HMA domains are integrated in some NLRs where they confer novel recognition specificities. The rice Pik-1 receptors, are such a NLR family, that recognize Avr-Pik effectors from Magnaporthe oryzae. Using novel Pik-1 orthologs, phylogenetic analyses, algorithms for ancestral sequence reconstruction and functional characterization of chimeric HMAs, this study examines the HMA molecular evolution of the allelic variants Pikp-1 and Pikm-1 to demonstrate that they convergently evolved to recognize Avr-PikD.

Comments

The immune profile in response to Avr-PikD for the majority of the novel Pik orthologs described in this manuscript is not known. However, I would suggest to indicate that Avr-PikD is recognized by rice Pik alleles Pikh, Pikp, Pikm, Piks and Pik* (Kanzaki et al., 2012; De la Conception et al., 2020). Hence, it could be discussed that, while Pikm and Pik* display the typical interface EMVKE, Piks, the closest ortholog of Pikm, is EMAKE and also recognize Avr-PikD.

In this Figure 5A, the legend indicates "Historical mutations […] are shown next to the appropriate nodes." However, the node indicating the L221 mutation (IAQVV to LAQVV) is not at the correct position as both branches deriving from this node harbour the L221 mutation. The LAQVV indication rather describes (i) the branch before this node or (ii) the mutation that appeared at the previous ancient node.

Also, while the historical order of the substitutions places Q228K before V229I, both substitutions are found in the 3 sequences at the bottom of the tree (O. rufipogon W2003, OsPikh-1 and OsPikp-1). Therefore, it appears incorrect to place the two individual substitutions on different nodes that discriminate O. rufipogon W2003 vs the two others.

In Figure 9A, in MKANK-EMVKE (and for the 3 others combinations), it may be more judicious to replace the dash with a slash (as in the text), or maybe even better with an arrow to show the evolution, otherwise it could read like a first domain fused to a second one. The same could be applied to ANK-VKE and AV-VE. Also, as the reader constantly swap between Pikm-1 vs Pikp-1, and between interface 1 vs 2, and between ancHMA vs newHMAs, adding the legend "interface 1" or "interface 2" (as in Fig S26) on top of the 3D structures may also help a reader. Finally, I would also suggest adding, at least, the positions A260/N261/K262, E230, V261/K262/E263 and V231 on the protein models (as it is not so easy to find where are the two K262).

Regarding the amino acid position, your text and description are right, but I had trouble to understand "In both cases, Lys-262 (K-262)" when you described "EMVKE and LKANK" since (i) the lysine residues are in different positions in these two domains and (ii) because the NK part of LKANK is not presented in Fig4A so I presumed that this last lysine would be K-263. I then noticed that the numbers above the ancHMA in Fig4A and Fig7A are different because they each correspond to the presented Pik-HMA (and Pikp and Pikm have a one residue shift (de la Conception et al., 2018)). Therefore, in Fig4A and Fig7A, I think that it would appear judicious to place the numbering below their respective Pik-HMA. In addition, while reporting "Lys-262 is structurally shifted" is correct, I think that it may be pertinent to present an alignment to support this description which explains the ‘looping out’ of Pikp-HMA.

In Figure 8A, EMANK should be presented in purple for "EM" and green for "ANK" (as in 8B and 8D). As discussed for Figure 5A, EMANK describes the branch before the node. As for EMVKE, the position of the node is correct but few members are not EMVKE (likely because the phylogenetic tree is computed using the complete HMA domains, and not only these five mutations).

In Figure S7, I am not sure if you performed some manual adjustments but I would suggest adjusting TraesCS7D02G007700.1 on the 5' HMA integration site. Some manual adjustments would definitely strengthen the percentage identity.

For this same figure S7, the text refers to LpPik-1 but the figure is LPERR11G24730. It may be a good idea to indicate LpPik-1 (LPERR11G24730) in the text.

I understand that the phylogenetic tree presented in Fig S9A is the annotated version used for Fig 5A and Fig 8A. While the Pik-1 nomenclature is presented in Table S7, I find difficult to find OsPik* in this Fig S9A which is quite important to navigate the Pik-1 phylogeny. Therefore, I would suggest to present OsPik* as Pik*_1_HM048900_1 rather than Pik_1_HM048900_1.

The authors should pay attention to the OsPik* nomenclature which is presented with too many variations in the different Tables and Figures:

It reads Pik* in TabS10

It reads Pik*-1 in Fig2

It reads Pik-1* in FigS26

It reads OsPik* in Fig S6

It reads OsPik-1 in Fig S11

It reads OsPik*-1 in Fig S3

It reads Pik_1_HM048900_1 in Fig S9

It reads OsPik_1* HM048900_1 in Fig S12

In Figure S9, ancHMA I-N2 appears to be the reference sequence. Therefore, in the antepenultimate position, OBART11G23150 should not present an N but a dot.

In Figure S12B, I would suggest to also present the Pikm-1 HMA sequence as a last line in the alignment (even without its own data on probability for marginal reconstruction).

In Figure S26, the top line presents the Pikm-1 EM and the second line the ancHMA MK, but going to the right ancHMA is now on line 1 with IA. Is there a reason to not present ancHMA on line 1?

For homology modelling in Figure S33, which "failed" for MKANK, the authors present no obvious reason for using only Pikm-HMA–AVR-PikD as a model. It may be a good idea to consider using both Pikm-HMA–AVR-PikD (pdb:6FU9) and Pikp-HMA–AVR-PikD (pdb:6G10) as templates. And maybe even other Pik-HMA–AVR-Pik templates.

Reviewer

Freddy Boutrot, Anova-Plus, Evry, France.

References

De la Conception et al., 2018. Polymorphic residues in rice NLRs expand binding and response to effectors of the blast pathogen. Nature Plants. 4:576-585. https://doi.org/10.1038/s41477-018-0194-x

De la Conception et al., 2020. The allelic rice immune receptor Pikh confers extended resistance to strains of the blast fungus through a single polymorphism in the effector binding interface. BioRxiv https://www.biorxiv.org/content/10.1101/2020.09.05.284240v1

Kanzaki et al., 2012. Arms race co-evolution of Magnaporthe oryzae AVR-Pik and rice Pik genes driven by their physical interactions. The Plant Journal. 72:894-907.  https://doi.org/10.1111/j.1365-313X.2012.05110.x