Comparing Arabidopsis receptor kinase and receptor protein‐mediated immune signaling reveals BIK1‐dependent differences

Summary Pattern recognition receptors (PRRs) sense microbial patterns and activate innate immunity against attempted microbial invasions. The leucine‐rich repeat receptor kinases (LRR‐RK) FLS2 and EFR, and the LRR receptor protein (LRR‐RP) receptors RLP23 and RLP42, respectively, represent prototypical members of these two prominent and closely related PRR families. We conducted a survey of Arabidopsis thaliana immune signaling mediated by these receptors to address the question of commonalities and differences between LRR‐RK and LRR‐RP signaling. Quantitative differences in timing and amplitude were observed for several early immune responses, with RP‐mediated responses typically being slower and more prolonged than those mediated by RKs. Activation of RLP23, but not FLS2, induced the production of camalexin. Transcriptomic analysis revealed that RLP23‐regulated genes represent only a fraction of those genes differentially expressed upon FLS2 activation. Several positive and negative regulators of FLS2‐signaling play similar roles in RLP23 signaling. Intriguingly, the cytoplasmic receptor kinase BIK1, a positive regulator of RK signaling, acts as a negative regulator of RP‐type immune receptors in a manner dependent on BIK1 kinase activity. Our study unveiled unexpected differences in two closely related receptor systems and reports a new negative role of BIK1 in plant immunity.


Table S1
Arabidopsis thaliana mutant and transgenic lines used in this study. Table S2 GO term list of RNA-seq data done in Arabidopsis thaliana under elicitor treatment.

Table S3
Examples of genes specifically upregulated by flg22 or nlp20 categorized by GO terms.
Video/Movie S1 Time-lapse recording of cytoplasmic Ca 2+ elevations in an R-GECO1-expressing root treated with flg22.
Video/Movie S2 Time-lapse recording of cytoplasmic Ca 2+ elevations in an R-GECO1-expressing root treated with nlp20. Leaves of Arabidopsis Col-0 wild type plants were treated with 100 nM nlp20 or 10 nM flg22, and changes in membrane depolarization (∆V) were monitored continuously. The response to flg22 was generally higher than nlp20, but the difference was not statistically significant.

Fig. S2
Spatiotemporal analysis of calcium responses to nlp20 and flg22. Kymograph analysis of R-GECO1 signal intensities upon root treatment with 10 μM nlp20 (a) or 10 μM flg22 (b). Signal intensities are normalized for the average signal intensity of the baseline (25 frames). The cartoon above illustrates the different developmental zones of the root along the spatial (horizontal) axes of the kymographs. M, meristem; TZ, transition zone; EZ, elongation zone; DZ, differentiation zone. The calibration bar on the right indicates signal intensities with blue color indicating low and red color indicating high intensity.

Fig. S3
Time course of flg22 and nlp20-triggered ROS production and MAPK activation.
Arabidopsis leaf discs were treated with flg22 or nlp20 (100 nM or 1 µM), or water as control (mock), and ROS production (a) or MAPK activation (b) was monitored over time as described in Fig. 1. Bars in (a) present means ± SD (n≥6) of relative fluorescence units (RLU). (c) Arabidopsis wild-type seedlings were treated with water or 0.5 µM nlp20 or flg22 for 1 and 6 hours, and isolated RNA was subjected to RNA sequencing as described in Fig. 2. Given are the fold changes (Log2) of MLO12 transcript levels compared the water control. For qRT-PCR, leaves of Arabidopsis wild type plants were infiltrated with water (mock), 0.5 µM flg22 or 5 µM nlp20, and collected 6 hours after treatment. Relative expression of the MLO12 gene was normalized to the levels of EF-1α transcript and calibrated to the levels of mock treatment.
(a) Arabidopsis leaf discs were treated with 100 nM flg22, elf18, nlp20, or PG3, or water as control (mock), and ROS production was monitored over time. Bars present means ± SD (n=6) of relative luminescence units (RLU). (b) Arabidopsis leaf discs were treated with flg22, elf18, nlp20, or PG3, or water as control (mock), and ethylene production was measured at 3 and 6 hours post incubation. Bars present means ± SD (n=3).  Arabidopsis leaf discs of the indicated bak1 (a, b) or bik1 (b, c) mutant lines were treated with water (mock), 500 nM flg22 (a, c) or 500 nM nlp20 (b, d) and ROS production was monitored over time as described in Fig. 1. Data present means ± SD (n≥6) of relative fluorescence units (RLU).
Leaf pieces of wild-type, bik1, sid2, or bik1 sid2 plants were treated with water, 500 nM flg22, or 500 nM nlp20, and ROS accumulation was determined as in Fig  (a) Levels of the indole glycosinolate indol-3-ylmethyl glucosinolate (I3M) were determined in leaves infiltrated with 1 µM flg22 or nlp20 (also 0.1 µM for 48 hours), or water (mock) and harvested after 12 and 48 hours. Bars (nmol I3M/g fresh weight) present average values ± SD (n = 2). (b) Arabidopsis wild-type seedlings were treated with water or 0.5 µM nlp20 or flg22 for 1 and 6 hours, and isolated RNA was subjected to RNA sequencing. Given are the fold changes (log2) of CYP81F2 transcript levels compared the water control.   file uploaded as separate .xls file Table S3 Examples of genes specifically upregulated by flg22 or nlp20 categorized by GO terms.
Video/Movie S1 Time-lapse recording of cytoplasmic Ca 2+ elevations in an R-GECO1-expressing root treated with flg22.
Video/Movie S2 Time-lapse recording of cytoplasmic Ca 2+ elevations in an R-GECO1-expressing root treated with nlp20.

Plant Material
Arabidopsis plants were grown on soil or half-strength Murashige and Skoog (MS) medium as described (Brock et al., 2010). Plants were grown in climate chambers under short-day conditions (8 h : 16 h, light : darkness, 150 μmol/cm 2 s white fluorescent light, 40-60 % humidity, 22 °C). All mutants used in this study are in Arabidopsis thaliana accession Col-0 background (listed in Table S1).

Ion Flux Measurements
Membrane potential recordings were performed in 5-7-week-old plants. One day before measurements leaves were detached, glued to the base of a chamber (adaxial site), peeled off (abaxial epidermis) and left for recovery overnight in a standard solution containing 0.1 mM KCl, 1 mM CaCl2 and 5 mM MES adjusted to pH 5.5-5.7 with Tris. During experiments, exposed tissue was constantly perfused with the standard solution (2 ml/min); elicitors were applied for 2 min. For impalements, microelectrodes from borosilicate glass capillaries with filament (Hilgenberg, Malsfeld, Germany) were pulled on a horizontal laser puller (P2000, Sutter Instruments Co, Novato, CA, USA). They were filled with 300 mM KCl and connected via a Ag/AgCl half-cell to a headstage (Axon Inst., Union City, CA, USA) (Scherzer et al., 2015).

Callose Deposition
To visualize callose apposition, leaves of 5-week-old Arabidopsis plants were infiltrated with water or the indicated peptides and stained with aniline blue after 24 hours as described (Wang et al., 2009). Pictures were taken with an inverted microscope (Nikon ECLIPSE 80i), then adjusted and analyzed using Photoshop CS5. Quantification of callose was performed by counting dark pixels selected by Magic tool and calculated in % relative to the total pixels of the image.

Ion Leakage Assay
Ion leakage assays were performed as described (Lenarčič et al., 2017). Leaves of 5 to 6-week-old Arabidopsis plants were infiltrated with ddH2O or 500 nM PG3. After 10 min of infiltration, leaf discs (Ø 7 mm) were punched out and transferred into a 24-well plate. Two leaf pieces per well were floated on 1 mL ddH2O and shaken at 50 rpm. After 30 min of incubation, leaf pieces were transferred to fresh ddH2O and conductivity was measured at the times indicated using a conductivity meter (QCond2200).

Indole glucosinolate glucobrassicin determination
Leaves of 5-week-old Arabidopsis plants were infiltrated with peptide solution or ddH2O. For analysis of I3M (glucobrassicin) 200 mg of fresh plant leaves were harvested and homogenized in liquid nitrogen.
Extraction of the analytes was carried out with 500 µl 80 % methanol containing 0.1 % formic acid, followed by a second extraction with 500 µl 20 % methanol containing 0.1 % formic acid. Both supernatants were combined and dried in a speed vac. For analysis with a Waters Acquity UPLC -SynaptG2 LC/MS system the samples were redissolved in 100 µl 20 % methanol containing 0.1 % formic acid. 5 µl were injected onto a Water Acquity HSS T3 reverse phase column. Separation was carried out with a linear 10 min 99 % water to 99% methanol (both solvents containing 0.1 % formic acid) gradient. For detection, the mass spectrometer was operated in negative ESI mode. For quantification of I3M, integrated extracted ion chromatograms were calculated into pmol with a calibration function between 1 nM and 1 mM. The obtained results were then normalized to the exact amount of fresh weight material used.

Western Blot Analyses
For MAPK activity assays, Arabidopsis leaves were infiltrated with ddH2O or peptide solution and harvested at the indicated time points. Protein extraction and immunoblot analyses using the anti-phospho p44/42 MAP kinase antibody (Cell Signaling Technology) were performed as described (Brock et al., 2010).
Protoplasts were isolated using the protocol as described (Lu et al., 2011). For BIK1 phosphorylation assays, 0.1 mL protoplasts at a density of 2 × 10 5 /ml were transfected with 20 μg of plasmid DNAs carrying BIK1-HA as described (Lu et al., 2010), then treated with flg22 or nlp20. Anti-HA antibody was used for immunoblot analyses.

Statistical analysis
Data sets were analyzed using Microsoft Office Excel or JMP® 12.2.0. Comparisons between two groups were made using Student's t-test. Multiple groups were compared using ANOVA followed by Student's ttest for all possible individual comparisons.