Rice cellulose synthase-like protein OsCSLD4 coordinates the trade-off between plant growth and defense

Abstract

Plant cell wall is a complex and changeable structure, which is very important for plant growth and development. It is clear that cell wall polysaccharide synthases have critical functions in rice growth and abiotic stress, yet their role in plant response to pathogen invasion is poorly understood. Here, we describe a dwarf and narrowed leaf in Hejiang 19 (dnl19) mutant in rice, which shows multiple growth defects such as reduced plant height, enlarged lamina joint angle, curled leaf morphology, and a decrease in panicle length and seed setting. MutMap analysis, genetic complementation and gene knockout mutant show that cellulose synthase-like D4 (OsCSLD4) is the causal gene for DNL19. Loss function of OsCSLD4 leads to a constitutive activation of defense response in rice. After inoculation with rice blast and bacterial blight, dnl19 displays an enhanced disease resistance. Widely targeted metabolomics analysis reveals that disruption of OsCSLD4 in dnl19 resulted in significant increase of L-valine, L-asparagine, L-histidine, L-alanine, gentisic acid, but significant decrease of L-aspartic acid, malic acid, 6-phosphogluconic acid, glucose 6-phosphate, galactose 1-phosphate, gluconic acid, D-aspartic acid. Collectively, our data reveals the importance of OsCSLD4 in balancing the trade-off between rice growth and defense.

Keywords: cellulose synthase-like D; disease resistance; rice; trade-off; widely targeted metabolomics.

Figure 1

Characteristics of dnl19 mutant. (A) Dwarfism was observed in dnl19 plants at mature stage. Bar = 10 cm. (B) Flag leave of dnl19 was slenderer than that of wild type (WT). Bar = 1 cm. (C) Panicle morphology was affected in dnl19. Bar = 1 cm. (D) The culm of dnl19 was shorter than that of WT. Bar = 1 cm. (E) The grain shape was affected in dnl19 plants. Bar = 1 cm. (F) Both the flag leaf length and width were reduced in dnl19 mutant. Data are means ± SD (n ≥13, **P < 0.01, Student’s t-test). (G) The dnl19 plants displayed a decrease in panicle length and seed setting rate. Data are means ± SD (n≥10, **P < 0.01, Student’s t-test). (H) The 1st, 2nd and 3rd internodes of dnl19 were shortened. Data are means ± SD (n≥29, **P < 0.01, Student’s t-test). (I) Grain length, grain width and 1000-grain weight of WT and dnl19 plants were statistically analyzed. Data are means ± SD (**P < 0.01, Student’s t-test).

Figure 2

Effects of the dnl19 mutation on leaf and culm structure. (A) Cross section of flag leaf blades in WT and dnl19 plants. SV: small vein, LV: large vein, MR: midrib. Bar = 1 mm. (B, C) Magnified images of the leaf vein shown in (A). CC: clear cell, Arrow: bulliform cell. Bar = 0.5 mm in (B) and 0.2 mm in (C). (D) Number of large and small veins, and midrib thickness in leaf blades of WT and dnl19 plants. Data are means ± SD (n=10, **P < 0.01, Student’s t-test). (E) Longitudinal section of the 2nd internodes at the mature stage from WT and dnl19. Bar = 0.1 mm.

Figure 3

OsCSLD4 is the causal gene of dnl19 mutant. (A) Schematic diagram (not in scale) illustrates the InDel site in OsCSLD4. The arrow indicates the position of the dnl19 mutation, and the red dash indicates a C base deletion. (B) Sequencing results of the OsCSLD4 in wild type and dnl19 mutant. (C) Leaf morphology of the wild plant, dnl19 and functionally complemented plants (Com-1, Com-2). Bar = 1 cm. (D) Gross morphology of WT, dnl19, Com-1 and Com-2 at the maturing stage. Bar = 10 cm. (E) Gross morphology of Nipponbare (Nip) and oscsld4 at the tillering stage. Bar = 10 cm.

Figure 4

Global transcriptome analysis of dnl19 mutant. (A) Venn diagram of common genes in dnl19 and WT. (B) Volcano map of differentially expressed genes (DEGs) between dnl19 and WT. (C) Significantly enriched GO terms of DEGs between dnl19 and WT. GO terms belong to biological processes. (D) Heatmap of pathogen-associated genes transcription in dnl19 and WT. (E) Investigation of the relative expression of pathogen-associated genes by qRT-PCR. The expression of OsActin1 was used as an internal control.

Figure 5

Phenotypical characterization of dnl19 and wild type against M. oryzae and Xoo. (A) The dnl19 mutant was inoculated with the M. oryzae isolate. Blast resistance was evaluated by punch inoculation. Lesion length were determined on leaves at 7 days after inoculation. Bar = 1 cm. Data are means ± SD (n=6, **P < 0.01, Student’s t-test). (B) Fungal growth was quantified by qPCR. Data are means ± SD (n=6, **P < 0.01, Student’s t-test). (C) The change of relative expression level of OsCSLD4 in response to Xoo strain PXO99 treatment. The leaves of Nipponbare plants at tillering stage were used for inoculation. The expression of OsActin1 was used as an internal control. (D) Phenotypes of disease reactions in dnl19 flag leaves after inoculation with Xoo strain PXO99. Bar = 1 cm. Disease lesion length of wild type and mutant was measured at 2 weeks after inoculation. Data are means ± SD (n≥10, **P < 0.01, Student’s t-test).

Figure 6

Widely targeted metabolomics analysis of dnl19 mutant. (A) Scatter plot of OPLS-DA scores for WT group versus dnl19 group. (B) Permutation tests of OPLS-DA model. The X-axis indicates the accuracy of the OPLS-DA model, and the Y-axis indicates the accuracy frequency of 200 models in 200 permutation tests. R²(Y) = 0.999, Q^{2 =} 0.538. (C) Volcano map of differential metabolites with VIP value ≥1 and P-value<0.05. The abscissa represents the fold change of each metabolite, and the ordinate represents the P-value by student’s t-test. (D) Analysis of overall changes of KEGG metabolic pathway. Each dot represents a metabolic pathway. The X-axis was the differential abundance (DA) score, and the Y-axis was the ID number of KEGG metabolic pathway. * represents significance. * P < 0.05, ** P < 0.01, *** P < 0.001.