Symptoms and Physiological Changes in Leaves After Alternaria alternata Infection
The progression of disease on inoculated leaves was observed and the percentage disease index were assigned and calculated (Fig. 1A, C). The inoculation areas appeared to be water soaked by 2 DPI, which were hardly obvious. And the blotches comprising damaged tissue appeared to be weakly brown and the blotch size were increased by 3 DPI (Fig. 1B). Then, over the subsequent 24 h, observations confirmed that the blotch size was no further increased, but the color deepened further. To determine whether the hormones and ROS in the leaves of P. davidiana × P. bollena responded to A. alternata, the contents of SA, JA and hydrogen peroxide (H2O2) were determined. The results showed that the contents of SA, JA and H2O2 in inoculated leaves were increased, among which the JA contents were significantly increased compared to the controls during infection (Fig. 1D-F). Additionally, assays of enzymes involved in plant resistance to pathogen infection, such as POD, SOD, PPO, PAL and CAT, were also performed, and the activities of all these enzymes were significantly higher than in the control (Fig. 1G-K). These data indicate that the defense system in P. davidiana × P. bollena is responsive to A. alternata infection.
RNA-Seq and identification of differentially expressed genes (DEGs) during disease progression
Based on macroscopic observations, four time points, 0 d, 2 d, 3 d and 4 d, were chosen, and 3 biological replicates from those points were then analyzed on an Illumina HiSeq 2500 platform. All together, 12 cDNA libraries were constructed and a total of 58.1 GB of clean reads were generated. The average Q20 and Q30 values of the raw reads were 97.17% and 92.23%, respectively. Approximately 72.13% of the reads were mapped to the reference genome sequences (Table S1).
To investigate the DEGs during pathogenic infection in poplar, the clustering method was used to analyze and identify genes with expression changes of more than 1-fold and significance levels of P ≤ 1e−4. Gene expression was compared between the pathogen-infected and control samples, a comparison of the 2 and 0 DPI libraries identified 1930 DEGs, of which 1341 were upregulated and 589 were downregulated, a comparison of 3 and 0 DPI generated 2200 DEGs with 1407 upregulated and 793 downregulated. Of all the comparisons, the most DEGs, 4381, were found when the 4 and 0 DPI libraries were compared, in that analysis, 2039 were upregulated and 2342 were downregulated, suggesting that this time is particularly important for the poplar in the response to A. alternata infection (Fig. 2A). Then, the results above were illustrated as a Venn diagram, both unique and shared DEGs occur between and among pairs (Fig. 2B). For example, 421 DEGs (21.81% of the total) in 2 DPI vs. 0 DPI, 709 DEGs (32.22% of the total) in 3 vs. 0 DPI and 2,803 DEGs (63.98% of the total) in 4 vs. 0 DPI were shared with the other compared libraries. In addition, 584 DEGs were shared across all comparisons. These results suggest that, more genes become involved in the defense response during the pathogen infection progresses, and genes that respond early to pathogen infection may also subsequently function.
Validation of RNA-Seq Data using RT–qPCR
To verify the accuracy and reproducibility of the RNA-Seq data, the transcript of 15 randomly selected DEGs, including 10 up-regulated genes and 5 down-regulated genes, were analyzed by RT-qPCR. All 10 upregulated genes were significantly induced after pathogen infection, while the five downregulated genes were correspondingly repressed, which were in accordance with the RNA-Seq results (Fig. S1). Therefore, the consistency between the RT–qPCR results and the RNA-Seq analyses confirmed the reliability of our RNA-Seq data.
GO enrichment and KEGG pathway analysis of DEGs
The functional categories of the DEGs induced by pathogenic infection were obtained by GO enrichment and KEGG analysis. The GO enrichment analysis was performed via eggNOG-mapper (Fisher’s exact test, P value ≤0.05). The most significant enriched biological processes (BPs) were “secondary metabolic process” and “defense response to fungus”. In contrast, the most significant cellular component (CC) terms were related to “plant-type cell wall” and “integral component of plasma membrane”, while molecular function (MF) enrichments were related to “beta-glucosidase activity” and “glucosidase activity” (Fig. 2C). The pathways that displayed significant changes (P value ≤0.05) were identified via the KEGG database. A total of 17 KEGG pathways were significantly enriched, among which “plant hormone signal transduction” (ko04075), “phenylpropanoid biosynthesis” (ko00940), “MAPK signaling pathway-plant” (ko04016) and “flavonoid biosynthesis” (ko00941) were the most highly represented. Both “plant hormone signal transduction” (ko04075) and “phenylpropanoid biosynthesis” (ko00940) exhibited many DEGs, suggesting that in poplar, plant hormones and phenylpropanoid compounds play significant roles in resistance to pathogen infection. The “MAPK signaling pathway” (ko04016) exhibited the third largest number of DEGs, indicating that during pathogen infection, the expression of internal genes in plants is regulated by various signaling substances for pathogen defense (Fig. 2D). Moreover, the disease resistance genes involved in the “plant–pathogen interaction pathway” (ko04626) or “Ras signaling pathway” (ko04014), such as RPM, RGA and DRL, were mainly upregulated after pathogen infection (Fig. S2). Taken together, these results suggest that poplar has evolved a range of different molecular defense strategies depending on the infection stage of the pathogen. Additionally, among these DEGs, 380 TFs that could be divided into 34 TF families were identified, such as the bHLH, MYB, NAC, bZIP (basic leucine zipper), WRKY and HSF families (Table S2). Some of these TFs have previously been reported to be closely related to the plant resistance response to biotic stress. For example, the largest group of pathogen-induced TFs belongs to the WRKY family, which is well known for plant defense, and several MYB, NAC, and bHLH family members were also induced during infection (Fig. 2E). This result suggests that TFs also play roles in the resistance of P. davidiana × P. bollena to A. alternata infection.
DEGs Involved in the Phenylpropanoid Pathway
Secondary metabolites are known to be involved in plant defense against pathogens by forming physical or chemical barriers. In this study, DEGs related to secondary metabolism showed significant changes during pathogen infection, especially changes in “phenylpropanoid biosynthesis” (ko00940) and “flavonoid biosynthesis” (ko00941). In total, over one hundred DEGs with significant expression involved in 32 different biosynthetic pathways were enriched in these pathways. A gene for phenylalanine ammonia-lyase (PAL), which is involved in the first step of phenylpropanoid biosynthesis and catalyzes lignin accumulation, was upregulated at 3 DPI. Other genes involved in lignin biosynthesis, such as the 4-coumarate-CoA ligase (4CL) and shikimate O-hydroxycinnamoyltransferase (HCT) genes, were upregulated, while the trans-cinnamate 4-monooxygenase (C4H) gene was downregulated. Key genes leading to lignin formation, including p-coumarate 3-hydroxylase (C3H), cinnamyl alcohol dehydrogenase (CAD) and cinnamoyl-CoA reductase (CCR), were up- or downregulated at different time points in this study (Fig. 3A). Moreover, genes for chalcone synthases (CHS), a branch point enzyme in flavonoid biosynthesis, were upregulated during pathogen infection, while flavanone 3-hydroxylase (F3H), which leads to the formation of quercetin derivatives, was downregulated or upregulated. Additionally, genes encoding leucoanthocyanidin dioxygenase (LDOX) and leucoanthocyanidin reductase (LAR), which are involved in the synthesis and extension of proanthocyanidins, respectively, were both upregulated at 2 DPI and 3 DPI (Fig. 3B).
DEGs related to ROS accumulation and scavenging
As the H2O2 content in leaves, as well as the enzyme activities involved in ROS scavenging, changed after infection, the DEGs related to ROS accumulation and scavenging were identified. Three genes encoding respiratory burst oxidase homolog (RBOH) proteins, which respond to pathogens by producing ROS, were upregulated during infection. Moreover, genes encoding enzymes involved in ROS scavenging also showed significant changes. The expression levels of genes such as SOD, PPO, CAT, POD and glutathione peroxidase (GPX) were mainly upregulated, especially for PPO and POD, which showed high expression during infection (Fig. 4D). In addition, the expression of glutathione S-transferases (GSTs) was mainly upregulated at 2 DPI, while only GSTFB (Pda_00008118-RA) was continuously activated during infection.
Defense-related Proteins
Pathogenesis-related (PR) proteins, which are indispensable components of plant innate immunity, play important roles in plant defense against pathogens. In this study, many poplar PR genes were induced in response to A. alternata infection, the expression of PR-1 was upregulated at both 2 DPI and 3 DPI, but most of the thaumatin-like protein (PR-5) genes were downregulated, except one that had high transcript levels during pathogen infection (Fig. 4C). Our results also demonstrate that the PR-9s (peroxidases), PR-10s (ribonucleases), PR-14s (lipid-transfer proteins) and PR-16s (germin-like proteins) were up or down regulated throughout the infection period (Fig. 4C,D). Furthermore, most of the chitinases showed high transcript levels at both 2 DPI and 3 DPI but were downregulated at 4 DPI (Fig. 4A), while some of the glucusidas were up regulated at 2 DPI but then down regulated (Fig. 4 B) .
DEGs Related to Plant hormone and signal transduction
Phytohormones, especially salicylic acid (SA), jasmonic acid (JA), and ethylene (ET), are critical regulators of plant–pathogen interactions, among them, the JA/ET signaling pathway is involved in the response to necrotrophic pathogen infection. In this study, DEGs related to JA/ET synthesis and signal transduction were identified, and most were upregulated during infection. Genes involved in JA synthesis, including linoleate lipoxygenase (LOX), allene oxide synthase (AOS), allene oxide cyclase (AOC), 12-oxophytodienoate reductase (OPR) and acyl-coenzyme oxidase (ACX), were induced during infection. The expression levels of LOX, AOC, AOS3 and ACX were upregulated continuously, while the expression of OPR3 was upregulated at 2 DPI but downregulated at 4 DPI. Moreover, the expression levels of MYCs involved in JA signal transduction were upregulated at both 2 DPI and 3 DPI. However, the expressions of JAZ, which contains a conserved TIFY domain and is a negative regulator of JA signal transduction, were mainly downregulated during pathogen infection (Fig. 5A). Similarly, genes for enzymes related to ET synthesis, including S-adenosylmethionine synthase (SAMS), 1-aminocyclopropane-1-carboxylate synthase (ACS) and 1-aminocyclopropane-1-carboxylate oxidase (ACO), were all upregulated during infection. Genes involved in the ET signal transduction pathway were also significantly induced. Five ethylene-responsive transcription factors (ERFs) were upregulated, while one ERF was downregulated. Among these ERFs, two ERFC3s (Pda_00014570-RA and Pda_00008900-RA) were highly induced at all time points (Fig. 5B). The SA response pathway has frequently been shown to be associated with plant resistance to biotrophic and hemibiotrophic pathogens. DEGs related to SA accumulation and signal transduction were also identified. The genes for enhanced disease susceptibility 1 (EDS1) and its coregulator phytoalexin deficient 4 (PAD4), which are related to SA accumulation, were both downregulated during infection. However, a few genes involved in the SA signal transduction pathway were induced. The TGA1 gene was transcribed 1.03-fold at 2 DPI, and two NPR genes were upregulated at 4 DPI (Fig. S3).
The linoleate 13 -lipoxygenase gene (PdbLOX) of P. davidiana × P. bollena is involved in resistance to A. alternata
Lipoxygenase (LOX), a key enzyme in the JA biosynthetic pathway, was shown to be involved in plant defense responses against diverse pathogens. Based on the KEGG analyses and gene expression data, we found that several lipoxygenase genes were strongly induced during the infection of poplar leaves by A. alternata, especially PdbLOX (Pda_00004421-RA), therefore, we speculated that this specific LOX gene plays an underlying role in regulating P. davidiana × P. bollena defense against A. alternata infection. The transgenic poplars with overexpressing PdbLOX (PdbLOX-OE) and RNAi-silenced PdbLOX (PdbLOX-IE) were obtained via Agrobacterium infection, and then the expressions of PdbLOX were analyzed via RT–qPCR. PdbLOX was expressed at significantly higher levels in PdbLOX-OE lines than in the wild type (WT), but lower in PdbLOX-IE lines (Fig. 6B). Next, leaves from transgenic poplars, as well as those from WT, were inoculated with A. alternata. After inoculation, the disease spots on WT leaves were dark, extensive, and irregular, while the disease spots on leaves of PdbLOX-OE plants were light-colored circles, were particularly smaller than WT. In contrast, the disease spots on leaves of PdbLOX-IE plants were darker and larger than that on WT (Fig. 6A,C). All this results above were further confirmed by the percentage disease index (Fig. 6D).