A point mutation in the kinase domain of CRK10 leads to xylem vessel collapse and activation of defence responses in Arabidopsis

Abstract

Cysteine-rich receptor-like kinases (CRKs) are a large family of plasma membrane-bound receptors ubiquitous in higher plants. However, despite their prominence, their biological roles have remained largely elusive so far. In this study we report the characterization of an Arabidopsis mutant named crk10-A397T in which alanine 397 has been replaced by a threonine in the αC helix of the kinase domain of CRK10, known to be a crucial regulatory module in mammalian kinases. The crk10-A397T mutant is a dwarf that displays collapsed xylem vessels in the root and hypocotyl, whereas the vasculature of the inflorescence develops normally. In situ phosphorylation assays with His-tagged wild type and crk10-A397T versions of the CRK10 kinase domain revealed that both alleles are active kinases capable of autophosphorylation, with the newly introduced threonine acting as an additional phosphorylation site in crk10-A397T. Transcriptomic analysis of wild type and crk10-A397T mutant hypocotyls revealed that biotic and abiotic stress-responsive genes are constitutively up-regulated in the mutant, and a root-infection assay with the vascular pathogen Fusarium oxysporum demonstrated that the mutant has enhanced resistance to this pathogen compared with wild type plants. Taken together our results suggest that crk10-A397T is a gain-of-function allele of CRK10, the first such mutant to have been identified for a CRK in Arabidopsis.

Keywords: Fusarium oxysporum; Arabidopsis; cysteine-rich receptor-like kinase CRK10; defence response; gain-of-function mutation; kinase regulation; transcriptomics; xylem vessel collapse; αC helix.

Fig. 1.

The crk10-A397T mutant is a dwarf. (A–F) Rosette morphology of WT (A–C) and crk10-A397T (D–F) plants at 2 (A, D), 3 (B, E), and 4 (C, F) weeks after sowing. Scale bar: 1 cm. (G) Leaf series of 4-week-old WT (top) and crk10-A397T (bottom) plants. Scale bar: 1 cm. (H) Main inflorescence stem of 5-week-old WT (left) and crk10-A397T (right) plants. Scale bar: 1 cm. (I) Ten-week-old WT (left) and crk10-A397T (right) plants. Scale bar: 2 cm. (J) Siliques of WT (left) and crk10-A397T (right) plants. Scale bar: 1 cm. (K) Seeds of WT (left) and crk10-A397T (right) plants. Scale bar: 500 µm.

Fig. 2.

Xylem vessels collapse in the root and hypocotyl of crk10-A397T plants, but not in the stem. (A–F) Transverse cross sections of the base of stem (A, B), hypocotyl (C, D), and roots (E, F) of 5-week-old WT (A, C, E) and crk10-A397T (B, D, F) plants. Stain: potassium permanganate. Scale bars: 100 µm (A, B, C, E, F) and 50 µm (D). (G, H) Detail of xylem vessels in hypocotyls of 4-week-old WT (G) and crk10-A397T (H) mutant plants. Stain: potassium permanganate. Insert in top left corner of image shows original micrographs. Sale bars:, 50 µm (inset G) and 25 µm (inset H). (I, J) Detection of autofluorescence of lignin on resin-embedded cross sections of 4-week-old hypocotyls of WT (I) and crk10-A397T (J) plants. Scale bars: 10 µm. (K–T) Transverse cross sections of resin-embedded hypocotyls of WT (K, M, O, Q, S) and crk10-A397T (L, N, P, R, T) plants at 1, 2, 3, 4, and 5 weeks after sowing. Stain: potassium permanganate. Scale bars: 50 µm (K, L, M, Q, R), 100 µm (N, S, T), and 25 µm (O, P). (U, V) Transverse cross sections of resin-embedded hypocotyls of 5-week-old WT (U) and crk10-A397T (V) plants; arrows indicate span of xylem I (XI) and xylem II (XII) areas. Stain: potassium permanganate. Scale bars, 100 µm (U) and 50 µm (V).

Fig. 3.

CRK10 is a plasma membrane-localized protein expressed in association with the vasculature in the hypocotyl. (A, B) Histochemical staining of reporter lines expressing the CRK10_Pro:GUS construct showed expression of the reporter gene in the vasculature of stem and hypocotyls, as shown by free-hand cross section of 8-week-old inflorescence stem (A) and cross section of 2-week-old hypocotyl embedded in resin (B). Detail of cross section in (B) shows presence of histochemical staining in xylem parenchyma cells (asterisks) and differentiating xylem vessels (arrows) in the hypocotyl. Scale bars: 50 µm (A) and 25 µm (B). (C, D) Hypocotyl of 4-day-old seedling of transgenic Arabidopsis plant expressing 35S:CRK10-mCherry before (C) and after (D) plasmolysis. Representative Hechtian strands are indicated by white arrows. Scale bars: 20 µm.

Fig. 4.

The root–hypocotyl system is responsible for the dwarf phenotype of crk10-A397T mutant plants. Images of non-grafted plants (A), self-graft controls (B), and graft combinations (C) of WT and crk10-A397T mutant. Plants were imaged 3 weeks after micrografting was performed. The phenotype observed for the reciprocal grafting combinations was consistently observed in two independent repetitions of the experiment. An average number of 10 grafts per combination was recovered each time. Annotation: scion/rootstock.

Fig. 5.

His-CRK10kd^WT and His-CRK10kd^A397T are active kinases and the A397T substitution introduces an additional autophosphorylation site in the kinase domain of the protein. (A) Amino acid sequence of the cytoplasmic kinase domain of CRK10 used for in situ phosphorylation studies (CRK10 amino acid residue numbering shown on the left). Subdomains I–XI are indicated by roman numerals. Ala/Thr397 site is highlighted in red, and the RD motif is underlined and highlighted in bold. The conserved glycine-rich loop, catalytic loop and activation segment motifs are also shown. (B) Structure of the CRK10 kinase domain generated by homology modelling to the active kinase domain of BRI1. A molecule of an ATP analogue, coloured blue, occupies the active site. The αC helix is coloured red, and the atoms in the side chain of the mutated threonine residue are depicted as red spheres. (C) Recombinant His-CRK10kd^WT and His-CRK10kd^A397T extracts and their respective dead kinase controls (His-CRK10kd^WT-D473N and His-CRK10kd^A397T-D473N) resolved by SDS-PAGE; detection by western blot with horseradish peroxidase-conjugated anti-His antibody is shown. (D) SDS-PAGE and western blot following treatment of purified His-tagged proteins with λ-phosphatase (PPase); control: untreated protein. (E) In situ autophosphorylation sites in His-CRK10kd^WT and His-CRK10kd^A397T identified by LC-MS/MS analysis of the recombinant protein kinase domain. Threonine 397 is highlighted in red. C-term tail, C-terminal tail; His, 6× His-tag; JM, juxta-membrane domain; KD, kinase domain. (F) MS/MS spectrum of the doubly charged (m/z 758.4) tryptic phosphopeptide NEVVLVTKLQHR in which the threonine residue is phosphorylated. Neutral losses of phosphoric acid from both the precursor ion and the C-terminal y-ions are observed.

Fig. 6.

Transcriptional reprogramming in the crk10-A397T mutant shows activation of defence responses to biotic and abiotic stresses. (A) Venn diagram displaying the number of differentially expressed genes (DEGs) in the crk10-A397T hypocotyls compared with the WT at each developmental time point. (B) Top 15 enriched GO terms (Biological Process) for the up-regulated core DEGs in the crk10-A397T mutant plotted against their respective –log₁₀ false discovery rate (FDR). (C) Top 15 perturbations showing highest overall similarity to crk10-A397T mutant expression signature (analysis performed using the GENEVESTIGATOR Signature tool; log₂ fold change values of time point at 3 weeks after sowing was used as input for 274 core genes).

Fig. 7.

crk10-A397T mutant plants are more resistant to infection with Fusarium oxysporum f. sp. conglutinans 699. (A–D) Representative images of WT (A), crk10-2 (B), CRK10 OE-1 (C), and crk10-A397T mutant (D) plants at 11 d post-inoculation with F. oxysporum. (E) Mortality curve of WT, crk10-A397T, CRK10 OE-1, and crk10-2 plants from day 7 to 20 post-inoculation with F. oxysporum. Mortality is shown as a percentage. The experiment was repeated twice with similar results. (F) Probability of survival of each genotype following inoculation with the pathogen F. oxysporum. Associated 95% confidence intervals are shown. Deviance test, χ₃²=19.68, P<0.001. LSD, least significant difference.