XA21-specific induction of stress-related genes following Xanthomonas infection of detached rice leaves

Plant growth, peptide and bacterial treatments of detached rice leaves

For peptide treatments, wild type (WT) Kitaake and progeny from line EFR:XA21:GFP-3-8 (EFR residues 1-649 and XA21 residues 651-1025 called EFR:XA21:GFP in this study) Kitaake rice leaves were harvested from plants grown in the greenhouse for 4.5 weeks (Schwessinger et al., 2015a). 1.5–2 cm leaf sections were collected from expanded adult leaves using surgical grade scissors. Tissue from the leaf base and leaf tip was discarded. Detached leaves were equilibrated overnight in 6-well Costar cell culture plates under constitutive light (between 5–10 µmol/(m²*s)) (Fig. S1). Peptide treatments were performed in the morning and collected at the indicated times.

For bacterial inoculations, we used detached rice leaves harvested from 4-week old plants grown using a hydroponic growth system as described previously (Pruitt et al., 2015) under a light intensity of 280 µmol/(m²*s). Freshly harvested leaves from Kitaake and Ubi-Myc:XA21 Kitaake rice (called Myc:XA21 rice in this study) (Park et al., 2010) were cut into 1.5–2 cm pieces and immediately (without overnight equilibration) floated on 10 mM MgCl₂ solution for mock treatments or 10 mM MgCl₂ containing fresh Xoo cell suspensions at O.D.₆₀₀ of 0.1 (approximately 1 × 10⁷ cells mL⁻¹). The samples were left overnight under constitutive light (between 5–10 µmol/(m²*s)) and collected 24 h post-inoculation (hpi). Leaves were floated on approximately 1.5 mL Xoo cell suspension media in 6-well Corning Costar cell culture plates (Fig. S1). The detached leaf infection assay allows a more uniform distribution, compared to the scissor inoculation method (Kauffman et al., 1973), of Xoo inoculum by floating leaves on bacterial suspensions.

RNA isolation and qPCR gene expression analysis for peptide treated and bacterial infected leaf samples

Detached leaves were frozen in liquid nitrogen and powdered using a Qiagen tissuelyser. For tissue from greenhouse grown plants, RNA was extracted from powdered tissue using TRI Reagent and precipitated with isopropanol. For tissue from hydroponically grown plants, RNA was extracted using the Spectrum Plant Total RNA Kit from Sigma-Aldrich. RNA was DNase treated using the TURBO DNase kit from Life Technologies. RNA concentrations were normalized to the lowest sample concentration in each experiment. cDNA was synthesized from 2 µg of total RNA using the High Capacity cDNA Reverse Transcription Kit by Life Technologies. Gene expression changes were determined by ΔΔCt method (Schmittgen & Livak, 2008) normalized to Actin (LOC_Os03g50885) and compared to mock treated samples.

Identification of genes differentially regulated in elf18 treated EFR:XA21:GFP rice using RNA sequencing

Plant growth, leaf tissue isolation, and treatments were performed as described above. RNA was isolated from untreated Kitaake as well as untreated and elf18 treated EFR:XA21:GFP leaf tissue using the Spectrum Plant Total RNA Kit from Sigma-Aldrich and on-column DNase treated to remove genomic DNA contamination following the manufacturer’s instructions. RNA was quantified using the Quant-IT Ribogreen RNA Assay Kit. Sequences were deposited to the NCBI Sequence Read Archive (BioProject ID PRJNA250865).

RNAseq libraries, sequencing, and reference alignment were performed as described previously (Schwessinger et al., 2015a). Sample correlation between Kitaake and EFR:XA21:GFP replicates and differential gene expression analysis was performed using the Bioconductor ‘edgeR’ package for R (Robinson, McCarthy & Smyth, 2010; McCarthy, Chen & Smyth, 2012).

Bacterial strains and generation of mutants

We generated a PXO99AΔhrpA1 mutant in Philippine race 6 strain PXO99Az, a derivative of strain PXO99 (referred to as PXO99A in this study) (Salzberg et al., 2008). Xoo was grown in 10 g PSB (10 g Peptone (bacto-Peptone), 10 g Sucrose, 1 g sodium glutamate (glutamic acid, monosodium salt), final volume 1L, pH 7.0) or on PSA plates (PSB with 16 g/L bacto-agar) at 28°C. PXO99AΔhrpA1 was generated by single crossover mutagenesis using the suicide vector pJP5603 (Penfold & Pemberton, 1992). An approximately 500 base pair sequences within hrpA1 was amplified using forward primer 5′-CGGGGTACCGTGCTGCGTGATTTGTCCG-3′and reverse primer 5′-CGCGGATCCTGACTTGGTCGATGCAGTCC-3′and cloned into the multiple cloning site of pJP5603 using the restriction enzyme sites KpnI and BamHI. PXO99A-competent cells were transformed with the suicide plasmids by electroporation and plated to PSA with kanamycin (50 µg/ml). PXO99AΔhrpA1 colonies with kanamycin resistance were screened by PCR for colonies with single crossover events, which contain the vector disrupting the target gene. PXO99AΔraxST and PXO99AΔraxST(raxST) complemented strains used in this study were described previously (Pruitt et al., 2015). PXO99AΔraxST evades XA21-mediated immunity while the complemented PXO99AΔraxST(raxST) strain does not.

RNAseq analysis identifies 2212 genes that are differentially regulated in elf18 treated EFR:XA21:GFP rice leaves

We analyzed the transcriptomic profile of EFR:XA21:GFP rice lines treated with elf18 to identify genes differentially regulated during this response. We sequenced cDNA from EFR:XA21:GFP leaves treated with 500 nM elf18 for 0.5, 1, 3, 6, and 12 h. We also included untreated EFR:XA21:GFP and Kitaake as controls (Table S1). Multidimensional scaling of pairwise biological coefficient of variance comparisons for each sample revealed that replicate samples group together (Fig. 1A). This grouping of biological replicates demonstrates the overall transcriptional similarity between each sample (Robinson, McCarthy & Smyth, 2010).

We identified 2,212 genes that were differentially regulated in EFR:XA21:GFP rice treated with elf18 compared with untreated (0 h) samples. Using a false discovery rate (FDR) cutoff of 0.05 and absolute expression log fold change (logFC) of 2 or greater, we previously reported that the transcriptomic profile of untreated Kitaake compared to untreated EFR:XA21:GFP did not differ significantly (Schwessinger et al., 2015a). Over the treatment time course, we identified 2,212 DRGs (FDR <0.05, absolute logFC >2) using untreated EFR:XA21:GFP at 0 h as a reference. The number of DRGs that overlap between the elf18 treatment time points are summarized in Fig. 1B and File S1. Of the 2,212 differentially regulated genes, there were 1,420 up-regulated and 792 down-regulated genes. The highest number of DRGs (1,504) was observed 6 h post elf18 treatment. These results show that elf18 treated EFR:XA21:GFP rice express a substantially different set of genes over time compared to untreated (0 h) samples.

Stress response related genes are up-regulated while photosynthesis related genes are down-regulated in EFR:XA21:GFP rice treated with elf18

To examine the types of biological processes affected in elf18 treated EFR:XA21:GFP rice, we analyzed GO term enrichment of DRGs using the AgriGo analysis tool (Du et al., 2010). A total of 1,204 out of 1,420 of the up-regulated DRGs and 682 of the 806 down-regulated DRGs had GO annotations. An FDR of 0.05 or less was used to define significantly enriched terms compared to the Michigan State University annotation reference as calculated by the AgriGo tool (Du et al., 2010; Kawahara et al., 2013). Fig. 1C and File S2 summarize the most enriched GO terms in each of the three major DRG clades. Clade 1 contains 1,333 genes that are mostly up-regulated over time. Genes from clade 1 are enriched for metabolic process (GO:0008152), response to stimulus (GO:0050896) and response to stress (GO:0006950) GO terms (Fig. 1C). Clade 2 genes (122) are up-regulated across all time points and are enriched for secondary metabolic process (GO:0019748), metabolic process (GO:0008152) and response to stress (GO:0006950) GO terms (Fig. 1C). Clade 3 consists of 757 genes that are mostly down-regulated in all timepoints. Photosynthesis (GO:0015979) and response to abiotic stimulus (GO:0009628) are the most enriched GO terms associated with clade 3 genes (Fig. 1C).

qPCR validation of genes up-regulated in elf18 treated EFR:XA21:GFP plants

We chose 23 DRGs from the elf18 treated EFR:XA21:GFP rice RNAseq dataset with relatively high logFC and low FDR values after 3, 6, and 12 h for detailed analysis. We first assessed if the 23 genes up-regulated in elf18 treated EFR:XA21:GFP could be validated by qPCR analysis. Eleven out of 23 DRGs were up-regulated in EFR:XA21:GFP rice leaves after elf18 treatment. Transcripts of the remaining 12 candidate genes were detectable by qPCR amplification but were not up-regulated in elf18 treated EFR:XA21:GFP leaves (File S3).

Establishment of bacterial infection assay of detached rice leaves

We established a detached leaf infection assay to test if genes identified in the EFR:XA21:GFP experiments are representative of genes differentially regulated in Xoo infected Myc:XA21 rice. We observed bacterial ooze from the detached rice leaves three days after inoculation with Xoo strain PXO99A (Fig. 2). To further assess if Xoo infects rice leaves in our detached leaf infection assay, we measured the expression level of Os8N3 (LOC_Os08g42350), which was previously shown to be up-regulated in rice upon Xoo infection and is thus a useful marker of successful infection (Yang, Sugio & White, 2006). For these experiments, we also included a mutant PXO99A strain (PXO99AΔhrpA1) that is unable to infect rice as a control. The hrpA1 gene encodes a pilus protein that is essential for type III-secretion of effectors required for host infection (Wengelnik et al., 1996). We observed that the PXO99AΔhrpA1 Xoo mutant is unable to infect Kitaake and Myc:XA21 rice plants (Fig. S2). Both WT Kitaake and Myc:XA21 detached leaves express Os8N3 at higher levels compared to mock treatments 24 hpi with WT PXO99A, but not with PXO99AΔhrpA1 (Fig. 3). These results demonstrate that Xoo infects detached rice leaves.

We next employed the detached leaf infection assay to examine the expression of the stress-related marker PR10b in Xoo infected Myc:XA21 rice leaves. Compared with mock treated controls, PR10b is up-regulated in flg22 treated rice, elf18 treated EFR:XA21:GFP rice and Myc:XA21 rice treated with the RaxX21-sY (Chen et al., 2014; Schwessinger et al., 2015a; Pruitt et al., 2015). Using qPCR, we detected significant up-regulation of PR10b expression in Myc:XA21 rice leaves 24 hpi with PXO99A and PXO99AΔhrpA1. PR10b up-regulation was not observed in infected Kitaake leaves (Fig. 4). These results show that the detached leaf infection assay can be used to assess XA21-mediated marker gene expression and also indicate that RaxX expression or secretion is not affected by the ΔhrpA1 mutation.

Eight out of 11 genes induced in efl18 treated EFR:XA21:GFP rice are up-regulated in Xoo infected XA21 rice

We next employed the detached leaf infection assay to identify genes up-regulated upon Xoo infection of Myc:XA21 rice leaves. For these assays, we examined the gene expression of the 11 genes validated by qPCR analysis of elf18 treated EFR:XA21:GFP rice described above. We inoculated Kitaake and Myc:XA21 rice with WT PXO99A, PXO99AΔraxST mutants, and PXO99AΔraxST complemented with raxST (PXO99AΔraxST(raxST)). Xoo strains carrying mutations in raxST do not activate XA21-mediated immunity (Da Silva et al., 2004; Pruitt et al., 2015). The expression of 8 of 11 genes was specifically up-regulated in detached Myc:XA21 rice leaves 24 hpi with PXO99A and PXO99AΔraxST (raxST) but not in Myc:XA21 rice leaves infected with PXO99AΔraxST (Fig. 5 and File S3). The 8 validated marker genes encode a putative subtilisin-like protein (LOC_Os04g03100), a reticuline oxidase-like protein precursor (LOC_Os06g35700), the decarboxylase OsTDC1 (LOC_Os08g04540) (Kang et al., 2007), a peroxidase precursor (LOC_Os11g02100), RsOsPR10a (LOC_Os12g36830) (Hashimoto et al., 2004; Takeuchi et al., 2011), CYP71Z7 (LOC_Os02g36190) (Li et al., 2013), OsKO5 (LOC_Os06g37224) (Itoh et al., 2004), and one protein without a putative function (LOC_Os10g28299). The 3 remaining genes that were up-regulated in elf18 EFR:XA21:GFP rice but not in Myc:XA21 rice leaves encode a isoflavone reductase (LOC_Os01g13610), a subtilisin-like protein (LOC_Os04g03100), and a reticuline oxidase-like protein precursor (LOC_Os06g35700).