Differential regulations of abscisic acid-induced desiccation tolerance and vegetative dormancy by group B3 Raf kinases in liverworts

Abstract

Phytohormone abscisic acid (ABA) plays a key role in stomata closure, osmostress acclimation, and vegetative and embryonic dormancy. Group B3 Raf protein kinases (B3-Rafs) serve as positive regulators of ABA and osmostress signaling in the moss Physcomitrium patens and the angiosperm Arabidopsis thaliana. While P. patens has a single B3-Raf called ARK, specific members of B3-Rafs among six paralogs regulate ABA and osmostress signaling in A. thaliana, indicating functional diversification of B3-Rafs in angiosperms. However, we found that the liverwort Marchantia polymorpha, belonging to another class of bryophytes, has three paralogs of B3-Rafs, MpARK1, MpARK2, and MpARK3, with structural variations in the regulatory domains of the polypeptides. By reporter assays of the P. patens ark line and analysis of genome-editing lines of M. polymorpha, we found that these B3-Rafs are functionally redundant in ABA response, with respect to inhibition of growth, tolerance to desiccation and expression of stress-associated transcripts, the majority of which are under the control of the PYR/PYL/RCAR-like receptor MpPYL1. Interestingly, gemmae in gemma cups were germinating only in mutant lines associated with MpARK1, indicating that dormancy in the gametophyte is controlled by a specific B3-Raf paralog. These results indicated not only conservation of the role of B3-Rafs in ABA and osmostress response in liverworts but also functional diversification of B3-Rafs, which is likely to have occurred in the early stages of land plant evolution.

Keywords: Marchantia polymorpha; RNA-seq; Raf kinase; abscisic acid; genome editing; liverworts; vegetative dormancy.

Figure 1

(A) Phylogenetic relationships and polypeptide structures of group B2 and B3 Raf kinases. The phylogenetic analysis was conducted by the neighbor-joining method using CLASTALW. Numbers on the branches indicate bootstrap values (100 replicates) and the bar represents the number of amino acid changes per branch length. (B) Reporter assays of ark mutant of P. patens using cDNA constructs of group B2 and B3 Raf kinases of M. polymorpha. The constructs of cDNA fused to the rice actin promoter, proEm-GUS and proUbi-LUC were introduced into protonema cells by particle bombardment. The vector without cDNA was used as a control. The bombarded cells were cultured for 1 day with or without 10-μM ABA before GUS and LUC assays. Levels of gene expression are represented by GUS/LUC ratio. Error bars indicate standard error of the mean. *p < 0.05 and **p < 0.01 in the t-test (n = 3) compared with WT of the same treatment.

Figure 2

Effect of ABA on growth of gametophytes of M. polymorpha. Gemmae of wild type (WT) and three each of double mutant lines of MpARK1 and MpARK2 (Mpark1/2^gea to c) and MpARK1 and MpARK3 (Mpark1/3^gea to c) were grown on medium without ABA (control) (A) or with 1-μM (B) or 10-μM (C) ABA for 14 days. (D) Fresh weight of the gemmalings was plotted in histograms. Error bars indicate the standard error (n = 3). *p < 0.05 and **p < 0.01 in the t-test compared with WT of the same treatment.

Figure 3

Comparison of dormancy of gemmae in gemma cups. (A) Appearance of gemmae in a gemma cup of WT and the B3-Raf genome-editing lines of M. polymorpha. Germination of rhizoids was frequently observed in gemmae of Mpark1^ge, Mpark1/2^ge, and Mpark1/3^ge lines while gemmae of WT and Mpark2^ge and Mpark3^ge remained dormant within the gemma cup. (B) Germination rate of gemmae in gemma cup of WT and genome editing lines. Gemmae isolated from a gemma cup located at the most basal position in the thallus were used for analysis. Germination of rhizoids from gemmae was analyzed by staining with propidium iodide, and the percentages of gemmae growing rhizoids were plotted in histograms. The experiment was carried out on a date different from that shown in Supplementary Figure S4. Error bars indicate +/− SE of the mean (n = 3).

Figure 4

Effect of ABA on desiccation tolerance in M. polymorpha. Gemmae of wild type (WT) and the B3-Raf genome-editing lines were cultured for 1 day with or without 1 or 10 μM ABA in liquid medium. The gemmae were transferred to a container containing silica gel and dried for 2 days. After rehydration, the gemmae were transferred onto a fresh agar medium and cultured for 10 days to determine survival. (A) Growth appearance of gemmalings after culture for 10 days. (B) Survival rate was plotted in histograms. Error bars indicate +/− SE. *p < 0.05 by t-test (n = 3).

Figure 5

Profiles of ABA-responsive gene expression in wild type (WT) and the B3-Raf genome-editing lines. (A) Percentages of ABA-induced genes (>2) for which expression was reduced (<2) in the genome-editing lines. (B) Heat map showing ABA-induced and ABA-reduced genes in WT and the genome editing lines. The fold change used for the color key was calculated by comparing values of ABA-treated and control samples.

Figure 6

Relationships between ABA-induced genes mediated by the ABA receptor MpPYL1 and B3-Rafs. Using ABA-induced 355 genes commonly identified in this study and our previous study (Jahan et al., 2019), percentages of genes under the control of MpPYL1, B3-Rafs and both are analyzed. Genes for which expression in Mppyl1^ge, Mpark1^ge, Mpark1/2^ge, and Mpark1/3^ge is less than half of WT are represented as MpPYL1-, MpARK1-, MpARK1/2-, and MpARK1/3-regulated genes (A–C). The result of analysis against the merge of Mpark1^ge, Mpark1/2^ge, and Mpark1/3^ge (MpARK₁₂₃ merge) is also shown (D).

Figure 7

ABA-induced expression of LEA-like genes in wild type (WT) and B3-Raf genome editing lines. Quantitative RT-PCR analysis was carried out using RNA isolated from gemmalings treated with or without 10 μM ABA. Transcripts for MpLEAL1 (Mapoly0112s0030) (A), MpLEAL3 (Mapoly0035s0082) (B), MpLEAL5 (Mapoly0087s0015) (C), and MpLEAL6 (Mapoly0027s0114) (D) were analyzed. Relative values were determined by SYBR green-based quantitative PCR analysis using MpEF1 (Mapoly0024s0116) unaffected by ABA as a reference. Error bars indicate +/− SE. *p < 0.05, **p < 0.01 by t-test (n = 3) compared with WT of the same treatment.