Mixed Linkage β-1,3/1,4-Glucan Oligosaccharides Induce Defense Responses in Hordeum vulgare and Arabidopsis thaliana

Plants detect conserved microbe-associated molecular patterns (MAMPs) and modified “self” molecules produced during pathogen infection [danger associated molecular patterns (DAMPs)] with plasma membrane-resident pattern recognition receptors (PRRs). PRR-mediated MAMP and/or DAMP perception activates signal transduction cascades, transcriptional reprogramming and plant immune responses collectively referred to as pattern-triggered immunity (PTI). Potential sources for MAMPs and DAMPs are microbial and plant cell walls, which are complex extracellular matrices composed of different carbohydrates and glycoproteins. Mixed linkage β-1,3/1,4-glucan (β-1,3/1,4-MLG) oligosaccharides are abundant components of monocot plant cell walls and are present in symbiotic, pathogenic and apathogenic fungi, oomycetes and bacteria, but have not been detected in the cell walls of dicot plant species so far. Here, we provide evidence that the monocot crop plant H. vulgare and the dicot A. thaliana can perceive β-1,3/1,4-MLG oligosaccharides and react with prototypical PTI responses. A collection of Arabidopsis innate immunity signaling mutants and >100 Arabidopsis ecotypes showed unaltered responses upon treatment with β-1,3/1,4-MLG oligosaccharides suggesting the employment of a so far unknown and highly conserved perception machinery. In conclusion, we postulate that β-1,3/1,4-MLG oligosaccharides have the dual capacity to act as immune-active DAMPs and/or MAMPs in monocot and dicot plant species.


INTRODUCTION
Plants are constantly exposed to a variety of potential pathogens, including oomycetes, fungi and bacteria. To counteract potential intruders, plants have evolved a complex immune system. The presence of potentially harmful microbes can be recognized by plants through the perception of conserved microbe associated molecular patterns (MAMPs) that are absent from plants (Bigeard et al., 2015;Hückelhoven and Seidl, 2016). Besides the perception of these "non-self " molecules, plants are also able to sense altered plant-derived molecules that are only present upon pathogen attack or cell damage. These "self " molecules with eliciting activity are referred to as danger or damage associated molecular patterns (DAMPs) (Bigeard et al., 2015;Hou et al., 2019). The perception of MAMPs and DAMPs is mediated by plasma-membrane localized pattern recognition receptors (PRRs) that can be classified as receptor-like kinases (RLKs) or receptor-like proteins (RLPs) (Couto and Zipfel, 2016;Boutrot and Zipfel, 2017). Upon perception of MAMPs and DAMPs a signaling cascade is induced which results in the activation of pattern-triggered immunity (PTI). Typical PTI responses include intracellular Ca 2+ elevation, generation of reactive oxygen species (ROS), phosphorylation of mitogen activated protein kinases (MAPKs) and transcriptional reprogramming (Bigeard et al., 2015;Hückelhoven and Seidl, 2016;Hou et al., 2019).
As β-1,3/1,4-MLGs are present in the cell wall of monocotyledonous plants but are also abundant in bacterial and fungal species, they represent potential DAMPs and MAMPs in monocots and dicots, respectively. Thus, we analyzed the eliciting capacity of β-1,3/1,4-MLG oligosaccharides in H. vulgare and A. thaliana in this study. We found that β-1,3/1,4-MLG oligosaccharides derived from the hydrolysis of the H. vulgare β-1,3/1,4-MLG polysaccharide trigger canonical PTI responses in both, the monocot crop plant H. vulgare as well as the model dicot A. thaliana, suggesting a potential dual function as both DAMP and/or MAMP in a plant lineage-dependent manner. Reverse genetics and an accession screen in Arabidopsis revealed that known receptors and co-receptors of PTI are not involved in β-1,3/1,4-MLG oligosaccharide perception and that yet to be identified conserved molecular components mediate β-1,3/1,4-MLG oligosaccharide-induced signaling.

Plant Material and Growth Conditions
The Arabidopsis accession Col-0 was the background for all transgenic and mutant lines used in this study. Further Arabidopsis accessions that were used are listed in Supplementary Table 1. Seeds were surface sterilized by washing three times for 2 min with 70% EtOH and 0.05% Tween-20 with agitation. Seeds were afterward washed two times for 1 min with 100% EtOH and dried. The dry seeds were either sown on soil or grown on aqueous 1 /2 Murashige and Skoog (MS) medium. For RNA extraction and MAPK experiments, 7-days old seedlings were transferred into individual wells (two seedlings per well) of a transparent 24-well plate and grown for seven further days. Plants were grown in a growth cabinet (CLF Plant Climatics, Wertingen, Germany) under short day conditions (12 h light, 12 h darkness).

Calcium Measurements
Intracellular calcium was measured using an aequorin-based calcium assay (Ranf et al., 2012). Calcium responses in the absence of an elicitor was included as negative control. The Ca 2+ concentrations were calculated and normalized according to Rentel and Knight, 2004 and are depicted as L/Lmax with L representing the luminescence at any time point upon β-1,3/1,4-MLG oligosaccharide or MAMP treatment and Lmax representing the total remaining aequorin. To calculate Lmax, the luminescence obtained upon treatment with the discharge solution was integrated.

ROS Measurements
The generation of ROS was determined using a luminol-based assay. Leaf discs (4 mm diameter) of 5-7 weeks old Arabidopsis plants or 10-12 days old H. vulgare plants were incubated in water overnight in a flat-bottom 96-well plate. The water was replaced with a luminol solution [10 µg ml −1 Horseradish peroxidase (P6782, Sigma), 100 µM L-012 (120-04891, WAKO Chemicals)] containing no elicitor, elicitors at the indicated concentrations or a 1:10 dilution of H. vulgare β-1,3/1,4-MLG polysaccharide hydrolysis products. Luminescence was recorded with a TECAN infinite R M200 plate reader for 60 min in 1 min intervals with an integration time of 150 ms.

MAPK Assays
Arabidopsis seedlings were grown in vitro as described above. One day before the treatment, the medium was replaced with 500 µl 1 /2 fresh MS medium to ensure equal volumes. 14days old seedlings were treated with the elicitors for 12 min and directly frozen in liquid nitrogen. H. vulgare plants were grown on soil as described above, 12-14 leaf discs of 4 mm diameter were harvested from second leaves of 14days old plants and incubated for 16 h in 2 ml ultrapure water. The leaf discs were transferred to fresh ultrapure water and incubated for 30 min. Subsequently, elicitor solutions were added to the indicated final concentrations. Negative control samples were treated with an equivalent volume of water. The leaf discs were incubated for 12 min and then directly frozen in liquid nitrogen. The frozen seedlings or leaf discs were homogenized in 600 or 200 µl extraction buffer, respectively (Petutschnig et al., 2010). After centrifugation for 10 min at 4 • C at 13.000 rpm, the protein concentration was determined via Bradford Assay with BSA as standard and protein concentrations were equalized. Protein extracts were frozen at −20 • C. Samples were separated on a 10% SDS gel. Immunoblot analysis were performed using Phospho p44/42 (#9101, Cell Signaling Technology, 1:5000) as primary antibody and a goat anti-rabbit IgG (A3687, 1:5000, Sigma Aldrich) as secondary antibody.

RNA Isolation and qRT-PCR
Arabidopsis seedlings were grown in vitro as described above. One day before the treatment, the medium was replaced with 500 µl 1 /2 MS medium to ensure equal volumes. 14-dayold seedlings were treated with the elicitors for 30 min and directly frozen in liquid nitrogen. Total RNA was extracted from seedlings using Qiazol (Qiagen, Hilden, Germany) and digested with DNase (EN0521, Thermo Scientific). 1 µg RNA per 20 µl reaction were used to generate cDNA using RevertAid TM H Minus M-MulVRT (EP0451, Thermo Scientific). For qRT-PCRs, 3 µL of 1:500 diluted cDNA was used to analyze gene expression with SsoFast EvaGreen supermix (1725204, BioRad) using the following PCR conditions: 95 • C for 30 s, 45 cycles of 95 • C for 5 s, and 55 • C for 10 s, followed by fluorescence reading. For normalization, UBIQUITIN5 was amplified in parallel on each plate. Aliquots of cDNAs used within one experiment were pooled and a dilution series was prepared from the pool to calculate primer efficiencies. Primer pair efficiencies were determined to be 98.53% (WRKY33), 100.06% (WRKY53) and 103.58% (Ubiquitin5) for the experiment shown in Figures 1D

Seedling Growth Inhibition
Arabidopsis seedlings were grown for 5 days on 1 /2 MS medium and then transferred to 24-well plates (one seedling per well). Each well contained 500 µL 1 /2 MS medium with either no elicitor, or one of the following substances:1 µM flg22, 10 mM Sodium Phosphate buffer (pH = 6.5) or a 1:10 dilution of the enzymatically generated H. vulgare β-1,3/1,4-MLG oligosaccharides. Pictures were taken of 13-day-old seedlings. To determine the dry weight, seedlings were dried for 1 day at 65 • C and the total weight of all eight seedlings was determined.
Matrix Assisted Laser Desorption Ionization-Time of Flight (MALDI-TOF) analysis of mixed-linkage glucans was performed with a Bruker Autoflex system (Bruker Daltonics) operated in reflectron mode. 10 mg/ml of the oligosaccharide were mixed 1:5 with 2,5-dihiydroxybenzoic acid in 1:1 H 2 O:MeOH on a Bruker MTP 384 grounded steel MALDI plate. The samples were allowed to dry and directly analyzed.
H. vulgare and A. thaliana and induce canonical patterntriggered immune responses.
Together, these experiments corroborate the ability of β-1,3/1,4-MLG oligosaccharides to act as elicitors of immune responses in the monocot H. vulgare and the dicot A. thaliana. Notably, the analyzed immune responses were stronger in response to β-1,3/1,4-MLG tetrasaccharides in both H. vulgare and A. thaliana, possibly indicating that the corresponding receptor has a higher affinity to longer β-1,3/1,4-MLG oligosaccharides.

DISCUSSION
During plant-microbe interactions, microbial and plant CWDEs are secreted into the extracellular space, where they act on their opponents' cell walls and release small oligosaccharides . Thus, the carbohydrate-rich cell walls of plants and their pathogens represent a source for potential DAMPs and MAMPs. In the last years, several cell wall-derived DAMPs and MAMPs have been identified (Saijo et al., 2018;Hou et al., 2019;Pontiggia et al., 2020), but it is conceivable that a substantially larger number of cell wall-derived ligands can be perceived by plants. Here we provide evidence that β-1,3/1,4-MLG oligosaccharides activate immune responses in more than 100 accessions of the dicot model plant A. thaliana and the monocot crop plant H. vulgare.
(D) Western Blot shows phosphorylated MAPK at different time points (0, 5, and 15 min). As loading control, the PVDF membrane was stained with Coomassie Brilliant Blue (CBB). Data show the result from one biological replicate. The experiments were repeated twice with similar results.
cell wall (Hacquard et al., 2013). The analysis of plant pathogen secretomes and the biochemical characterization of identified CWDEs have the potential to reveal whether CWDEs acting on the β-1,3/1,4-MLG polysaccharide are present in a given monocot plant pathogen species.
Besides A. thaliana, also the crop plants pepper and tomato can detect β-1,3/1,4-MLG oligosaccharides (Rebaque et al., 2021), which suggests that the responsible perception machinery is evolutionary highly conserved among dicot plants. Our study supports this conclusion and goes beyond in demonstrating an invariable intraspecific conservation in the dicot species Arabidopsis as well as potential evolutionary maintenance in the monocot lineage.
We also tested mutants of the prototypical protein MAMP receptors FLS2 and EFR (Gómez-Gómez and Boller, 2000;Zipfel et al., 2006), as well as the promiscuous co-receptor BAK1 and adaptor kinase SOBIR1, which are involved in the perception of many proteinaceous ligands Heese et al., 2007;Roux et al., 2011;Schwessinger et al., 2011). In support of a role in protein-derived MAMP/DAMP receptor complex formation, these proteins are dispensable for β-1,3/1,4-MLG oligosaccharide recognition. The observation that known components of the immune system are not implicated in β-1,3/1,4-MLG oligosaccharide perception implies that yet unknown molecular components mediate immune activation in response to β-1,3/1,4-MLG oligosaccharides. In previous studies, elicitor-insensitive Arabidopsis accessions have been used to identify immune receptors and signaling components (Gómez-Gómez et al., 1999;Gómez-Gómez and Boller, 2000;Jehle et al., 2013;Zhang et al., 2013). However, more than 100 A. thaliana accessions tested in this study were sensitive to β-1,3/1,4-MLG oligosaccharides indicating that the sensing and signaling machinery is highly conserved. We look forward to the identification of the molecular components required for β-1,3/1,4-MLG oligosaccharide perception by forward genetics in the future.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
VL, HB, EP, and SB conceived and designed the experiments. SB conducted all experiments, except for HPAEC-PAD, and MALDI-TOF analyses that were performed by GA and NJ. SB, EP, HB, and VL analyzed and discussed the data. SB, EP, and VL wrote the manuscript. All authors contributed to the article and approved the submitted version.

FUNDING
VL and SB were supported by the Deutsche Forschungsgemeins chaft (DFG project number 273134146) within the frame of the joint international DFG-IRTG/NSERC-CREATE training program IRTG 2172 "PRoTECT: Plant Responses To Eliminate Critical Threats" (https://www.uni-goettingen.de/de/protect/ 529150.html;2016projectB2) at the Göttingen Graduate Center of Neurosciences, Biophysics, and Molecular Biosciences. Work at the University of British Columbia was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC, Discovery Grants RGPIN 435223-13 and RGPIN-2018-03892).

ACKNOWLEDGMENTS
We thank Simone Ferrari (Sapienza University of Rome) for kindly providing oligogalacturonide preparations and Sabine Wolfarth for technical assistance.