NLR immune receptor-nanobody fusions confer plant disease resistance

Jiorgos Kourelis, Clémence Marchal, Sophien Kamoun

Posted on: 4 November 2021 , updated on: 20 April 2023

Preprint posted on 24 October 2021

Article now published in Science at http://dx.doi.org/10.1126/science.abn4116

Pikobodies with huge potential: One giant leap toward ready-to-order plant resistance genes

Selected by Marc Somssich

Categories: bioengineering, cell biology, immunology, molecular biology, pathology, plant biology

Updated 19 April 2023 with a postLight by Marc Somssich

The bioRxiv preprint highlighted here has now been republished in Science, where it has already attracted some attention from the mainstream media. My favorite headline comes from German magazine Spektrum, which writes “This is how Alpacas can protect plants from disease”.

And while we are on the topic of Alpacas, this is also where my main criticism of the republished and peer reviewed version of this article comes in: The original preprint ended with a figure (Fig. 4), illustrating the pipeline of how to generate the synthetic Pikobodies that can provide plant disease-resistance. This figure, anchored by a mildly annoyed Llama (or Alpaca), and adopted with minimal (but pointed) change for Fig. 2 of our preLight, was not just beautiful and eye-catching, but also a very well executed and easy-to-grasp summary of the manuscript. Such a figure is not only informative, it is also useful in illustrating and communicating scientific advances to the scientific community, and, maybe even more important, to the general public. It is clear that the figure was moved to the supplementary material because of the word/figure/page restrictions imposed by the journal, but given that these restrictions are an outdated remnant from the time of paper-based journals (and it would be possible to offer online-versions of papers with additional figures), it is a real shame that this figure is now (pretty much) hidden from science communicators and the public.

This criticism aside, it is important to also highlight the work that has replaced our beloved Alpaca as Figure 4 in the journal-version of the paper. One of the main points I raised as “next steps” in my preLight was the necessity to test the functionality of the Pikobodies in stable transgenic plant lines, rather than transiently transformed leaves of N. benthamiana only. I suggested to do this in Arabidopsis thaliana, since the generation of stable plant lines is work and time consuming. However, for their Science paper, Kourelis and Marchal et al. (2023) took this a step further and produced stable transgenic N. benthamiana lines. These transgenic plant lines expressed their GFP-targeting Pikobody^Enhancer, and when challenged with GFP– or mCherry-expressing Potato virus X indeed conferred resistance against the GFP-expressing variant of the virus, but not the mCherry-version. This acquired resistance was independent of the infection method (lab-based infiltration, or, closer to a real-life situation, leaf wounding).

Given that the Pikobodies need to work in stable transgenic plants, should they be used in the field to confer resistance to crops, this additional experiment is extremely valuable and has improved this paper significantly. It will be very interesting to see the further developments around Pikobodies, particularly in crops and in the field.

Background

In his talk ‘Beyond single genes: receptor networks underpin plant immunity’ Prof. Sophien Kamoun points out one of the main differences between the immune systems of plants and animals (to any animal researcher reading this: Yes, plants really have an immune system): While animals have an adaptive immune system, enabling them to acquire immunity against new pathogens and creating an immunological memory through the production of antibodies, plants find themselves in a continuous evolutionary arms race with their pathogens, having to evolve new receptors to sense newly evolved pathogen effectors (1,2).

The plant immune system is assumed to be a two-tiered system, with the two tiers, pattern- and effector-triggered immunity (PTI & ETI), being tightly interconnected (1,3). PTI is centred around plasma membrane-localized pattern recognition receptors (PRRs), which sense extracellular pathogens trying to breach the membrane and enter the cell. On tier 2, ETI is an intracellular surveillance system to detect avirulence (AVR) effector proteins that were injected into the cell by pathogens that successfully escaped PRR-detection (1,2). Both detection systems launch the plant’s defence response, with activated ETI leading to an amplification of the PTI-triggered response, that can eventually result in controlled death of the infected cell in order to protect the cells around it (hypersensitive cell death response (HR), which is used as the main readout for an immune reaction in the new preprint discussed here (4)) (1,2).

The receptors involved in ETI are mainly nucleotide binding leucine-rich repeat (NLR) proteins, which are generally composed of three domains (Fig. 1): on their N-terminus there can be a coiled-coil (CC) or Toll/interleukin-1 receptor/R protein (TIR) domain, necessary for the NLR to oligomerise and activate an immune response (5). In the centre is a nucleotide-binding (NB) domain involved in effector-binding, and located on the C-terminus is an LRR domain with an auto-inhibitory function, and involvement in effector-binding (5). There are exceptions to this canonical structure, of course, and some NLRs sense the pathogen effectors via an additional integrated domain (ID) (Fig. 1) (5). While such IDs generally are highly specific to a certain effector, recent work led by Juan Carlos de la Concepcion has provided proof of concept that their specificity can also be manipulated (6). In their study, de la Concepcion et al. (2019) use the ID of the NLR Pikp and engineer it to bind several AVR-Pik effector variants, rather than just its specific AVR-Pik effector (6). Similarly, Liu et al. (2021) have engineered the ID of the rice NLR RGA5 to recognize related variants of its specific effector AVR-Pib (7).

Fig. 1: Structure of plant NLR effector receptors. Canonical NLRs carry a CC (or TIR), NB and LRR domain, and form a heterodimer made up of a sensor and a helper NLR. Some atypical NLRs have an additional ID to bind the cognate pathogen effector protein. From ref. 4.

As a result of this work, the Pik NLR gained a wider effector recognition profile (6,7). But maybe more importantly, these studies demonstrated that the ID is tolerant of manipulation and therefore may function as a scaffold to engineer completely new designer effector-recognition domains. And by engineering NLR gene stacks that readily provide crop plants with resistance to several pests common to a geographical area or season, a pseudo-adaptive immune system for plants could potentially be created. And this is exactly the direction Jiorgos Kourelis and Clémence Marchal (equal contribution) from the lab of Sophien Kamoun are heading with their new preprint: They engineer a completely new and adaptable(!) effector-recognition domain for a plant NLR (4).

In this Preprint

The authors used nanobodies, the minimal antigen-binding fragment of camelid antibodies, raised against GFP or mCherry and inserted these in the position of the ID of the Pik-1 NLR. In case these NLR-nanobody fusions (named Pikobodies) are functional resistance proteins, the engineered GFP/mCherry-Pik-1 NLRs would trigger an immune reaction in the plant in response to the detection of GFP or mCherry. But first the authors checked for any autoimmune activities of the engineered Pikobodies, which manipulated NLRs often have, by expressing them in Nicotiana benthamiana leaf epidermal cells. Pikobodies that trigger an autoimmune response will cause necrotic lesions around the infiltration site, as a result of local cell death. Several of the designed Pikobodies did not show such an autoimmune response, and thus were selected for further analyses.

These potentially functional Pikobodies included Pikobody(Enhancer) recognizing GFP, and Pikobody(LaM-4) recognizing mCherry. The authors then co-expressed these with GFP and mCherry proteins, to see if the Pikobodies recognized their respective target. And indeed, co-expression of Pikobody(Enhancer) with GFP resulted in cell death, while co-expression with mCherry showed no effects. On the other hand, expression of Pikobody(LaM-4) only resulted in cell death when expressed with mCherry, but not GFP. To see if these Pikobodies would also react to actual infection by a pathogen, the authors infected N. benthamiana leaves expressing the different Pikobodies with a GFP– or mCherry-expressing potato virus X. Again, the Pikobodies appeared to recognize their intended target protein and seemed to provide protection against viral infection.

Finally, the authors checked if their Pikobodies were compatible with gene stacking. Should the Pikobodies be used to provide ready-to-order resistance (R) genes in the future, crop plants would ideally be protected against several pathogens, which would require the introduction of more than one R gene (i.e., a stack of genes) (8). To test the compatibility of the Pikobodies to R gene stacking, the authors co-transformed N. benthamiana leaf epidermal cells with their Pikobody(Enhancer) or (LaM-4), in combination with the wild type Pik NLRs, or the two Pikobodies(Enhancer) and (LaM-4) together, and showed that this co-expression indeed resulted in an immune response to the two different recognized effectors.

Perspectives

The work presented in this preprint is a remarkable proof-of-concept study demonstrating that Pikobodies hold the potential to provide ready-to-order R genes against virtually any pathogen. The pipeline, as outlined by the authors in their final figure (Fig. 2), is simple and straight forward: Pathogen effectors are isolated and used to immunize a camelid (llamas or alpacas are often used, Sophien Kamoun made it clear on social media that he prefers camels). Anti(nano)bodies produced by the camelid are then isolated and integrated into the ID of a Pik NLR-scaffold. Finally, the resulting Pikobody NLRs are introduced into plants to provide resistance against one or more pathogens

Fig. 2: Pipeline to produce Pikobodies (counterclockwise from top left). Adapted from ref. 4.

The next steps toward this aim are clear: First, the Pikobodies must be shown to work in stably transformed plants, first in Arabidopsis thaliana, then in an actual crop plant, and the resistance must persist over several generations. Then, of course, Pikobodies must be produced that contain nanobodies against actual pathogen effectors; effectors from wheat and rice blast or potato late blight are some prime candidates that will be tested first. Such Pikobodies would then be expressed as R gene stacks in crop plants and hopefully provide resistance against these pathogens.

Should this concept indeed prove working and efficient in crop plants, it could be a revolution for the agricultural sector.

References:

Jones JDG, Dangl JL. The plant immune system. Nature. 2006;444: 323–9. Available at doi:10.1038/nature05286
Kamoun S. NLR receptor networks: filling the gap between evolutionary and mechanistic studies. Zenodo. 2021;: 5504058. Available at doi:10.5281/zenodo.5504058
Ngou BPM, Ahn H, Ding P, Jones JDG. Mutual potentiation of plant immunity by cell-surface and intracellular receptors. Nature. Springer US; 2021; Available at doi:10.1038/s41586-021-03315-7
Kourelis J, Marchal C, Kamoun S. NLR immune receptor-nanobody fusions confer plant disease resistance. bioRxiv. 2021;: 465418. Available at doi:10.1101/2021.10.24.465418
Duxbury Z, Wu C, Ding P. A Comparative Overview of the Intracellular Guardians of Plants and Animals: NLRs in Innate Immunity and Beyond. Annu Rev Plant Biol. 2021;72: annurev-arplant-080620-104948. Available at doi:10.1146/annurev-arplant-080620-104948
De la Concepcion JC, Franceschetti M, MacLean D, Terauchi R, Kamoun S, Banfield MJ. Protein engineering expands the effector recognition profile of a rice NLR immune receptor. Elife. 2019;8: 1–19. Available at doi:10.7554/eLife.47713
Liu Y, Zhang X, Yuan G, Wang D, Zheng Y, Ma M, et al. A designer rice NLR immune receptor confers resistance to the rice blast fungus carrying noncorresponding avirulence effectors. Proc Natl Acad Sci U S A. 2021;118: e2110751118. Available at doi:10.1073/pnas.2110751118
Ghislain M, Byarugaba AA, Magembe E, Njoroge A, Rivera C, Román ML, et al. Stacking three late blight resistance genes from wild species directly into African highland potato varieties confers complete field resistance to local blight races. Plant Biotechnol J. 2018;: 1–11. Available at doi:10.1111/pbi.13042