Vertically Transmitted Epichloë Systemic Endophyte Enhances Drought Tolerance of Achnatherum inebrians Host Plants through Promoting Photosynthesis and Biomass Accumulation

Achnatherum inebrians (drunken horse grass, DHG) plants, a dominant grass species in the arid and semi-arid regions of northwest China, symbiotic with an Epichloë fungal endophyte, is well adapted to drought. However, little is known about how the presence of the foliar Epichloë endophyte enhances the tolerance of DHG to drought at the molecular level. This study explored the positive effects of the presence of the Epichloë endophyte on plant growth, biomass, and photosynthetic efficiency and processes of DHG under non-drought and two drought (moderate and severe) treatments, using RNA sequencing to compare transcriptomes. The transcriptome results showed that 32 selected unigenes involved in the photosynthesis processes within Epichloë symbiotic plants were differently expressed (DEGs) versus non-symbiotic plants. The majority of these selected DEGs were upregulated in Epichloë symbiotic plants versus non-symbiotic plants, such as upregulated unigenes (c51525.graph_c1, c47798.graph_c0 & c64087.graph_c0) under drought conditions. In line with the transcriptomes data, the presence of the Epichloë endophyte promoted the photosynthetic rate and biomass accumulation of DHG plants, and the relationship between the photosynthetic rate and biomass is linear and significant. The presence of the endophyte only increased the biomass per tiller of DHG plants under drought. This study provides further insights into the molecular mechanisms that underlie the enhanced plant growth and drought tolerance of Epichloë-symbiotic DHG plants.


Introduction
Drought, a primary environmental factor, limits plant productivity in natural ecosystems [1]. In grasslands in the arid and semi-arid regions of northwest China, grasses, including forage species in the family Poaceae, are typically dominant species. Grassland species adapt and respond to drought through many strategies, such as changes in plant physical responses, biomass accumulation and/or allocation, and the accumulation of some protective metabolites [2,3]. Natural selection and plant breeding can also enhance drought tolerance [4]. Some symbiotic beneficial microbes (e.g., root arbuscular mycorrhizal fungi and foliar Epichloë endophytes) enhanced plant drought tolerance through many strategies, such as absorbing water and nutrients through external hyphae of mycorrhizal and ectomycorrhizal fungi under drought conditions [3,5].
Seeds of E. gansuensis symbiotic DHG plants were generated from one grass population collected from the grassland in Sunan County, Gansu, China (101 • 01 E, 38 • 35 N, attitude 3297 m). Epichloë non-symbiotic DHG plants were generated by treating symbiotic seeds from F0 generation with a systemic fungicide (Thiophanate-methyl, 70% effective component) with 100 times dilution and 2 h treatment [30]. In order to multiply seeds, fungicide-treated and untreated seeds were planted in contiguous plots in an experimental field of the Yuzhong campus of Lanzhou University. All DHG plants were checked via microscopic examination for the presence of seldom-branched hyphae characteristic of Epichloë spp. in leaf sheath pieces stained with aniline blue, in the seeds of individual plants, and plants were individually labeled as EI or EF plants, respectively. Additionally, we did not observe the effects of fungicide treatment on the morphology, phenology, and growth of our experimental plants through pot and field experiments [25,30,31]. Seeds were collected from EI and EF DHG plants grown in the experimental field and stored at −4 • C before planting.
Three EI or EF seeds were planted in one plastic pot (diameter: 24 cm, height: 15 cm) filled with 200 ± 2 g of sterilized vermiculite (120 • C for 5 h), and later thinned to one seedling per pot. Hoagland's solution was used to quantitatively water these experimental pots every other day after the appearance of the second fully expanded leaf of individual plants [26]. Pots were maintained at a constant-temperature (26 ± 2 • C) greenhouse. After one month, pots containing similar sized EI (n = 27) and EF (n = 27) seedlings were cut 15 cm above the vermiculite surface, and the water-holding capacity of each pot was reduced to 15% relative saturation moisture content (RSMC) [26]. Subsequently, severe drought (SD, 15% RSMC), moderate drought (MD, 30% RSMC), and no drought (CK, 60% RSMC) treatments were established and sustained for 50 days. There were 9 replicates per treatment.

Differentially Expressed Genes
At the end of the soil moisture treatments, three fresh leaves of three EF or EF DHG plants for each soil moisture treatment were collected and immediately frozen in liquid nitrogen, and then stored at −80  [32]. Clean data were mapped back onto the assembled transcriptome; read count for each gene was obtained from the mapping results. For each treatment, with three biological replicates, differential expression analysis of the two groups (the EF plants were the control group) was performed using the DESeq2 package in R software (version 1.10.1), which provides statistical routines for determining differential expression of genes using a model based on the negative binomial distribution. The resulting p values were adjusted using the Benjamini and Hochberg's approach for controlling the false discovery rate (FDR) at 0.05. These log 2 [fold changes (FC)] of unigene FPKM were used to identify whether these unigenes were differentially expressed genes (DEGs) [33]. We used the KOBAS2.0 software to test the statistical enrichment of DEGs in KEGG pathways [34,35]. DEGs involved in photosynthesis (ko00195) and photosynthesis antenna proteins (ko00196) from the present transcriptomic data were selected for further analysis. Here, a total of 32 unigenes were selected for the subsequent analysis.
Amino acid sequences of these selected unigenes were blasted (blastx) against the genome of a related species to get the referenced sequences, and the resulting and reference sequences were used to construct Neighbor-Joining (NJ) phylogenetic trees using Molecular Evolutionary Genetics Analysis (MEGA, version 10.0.5) software [36]. Total leaf RNA of each sample used for RNA-Seq analysis was also used for quantitative real-time PCR (qRT-PCR) analysis. Single-stranded cDNAs were synthesized from 2.5 µg of total RNA with MMLV reverse transcriptase (TaKaRa, Dalian, China). The qRT-PCR was performed using SYBR Premix Ex Taq II Kit (TaKaRa, Dalian, China) on a 7500 Fast Real-time PCR system (Applied Biosystems, Waltham, MA, USA). The specific primers sequences of the selected unigenes used in the present study are listed in Table S1. Three technical replicates were carried out for each reaction, and the relative expression levels were normalized to the expression of the unigene (ID: c56016.graph_c0 detected in the present study) and calculated using the 2 − CT method.

Indexes of Plant Growth and Photosynthesis
In order to test the tolerance of Epichloë symbiotic DHG plants to severe and moderate drought, we assessed the indices of plant growth (plant height, tiller number and biomass), chlorophyll content, and photosynthesis (photosynthetic rate, intercellular carbon dioxide (CO 2 ) concentration, stomatal conductance, and transpiration rate). The heights and tiller numbers of EI and EF DHG plants were measured upon completion of the soil moisture treatments. The effects of the presence of the Epichloë endophyte on the photosynthetic rate of host plants were obtained through a standard meta-analysis (see the detailed information in the Supplementary Materials).
The chlorophyll content of three leaves of one individual EI or EF DHG plant was measured using a chlorophyll meter (SPAD-502Plus, Konica Minolta Sensing, Inc., Osaka, Japan). Then, the mean of three measurements represented the actual value of this individual plant. The photosynthetic indexes were performed using a portable photosynthesis system (LI-6400, LI-COR, Lincoln, NE, USA) between 9:00 and 11:00 on the morning of the final day of soil water treatments. The concentration of air CO 2 was 410 ± 10 µmol CO 2 mol −1 , the chamber was equipped with a red/blue LED light source (LI6400-02B), with the photo flux density set at 1200 µmol m −2 s −1 , and the detection conditions were at 28 ± 1 • C. Finally, the shoots and roots of sampled plants were collected from the nine pots to measure the fresh weight of shoots and roots, and then the dry weight of shoots and roots were recorded when a constant weight had been reached in an 80 • C oven for 48 h.

Data Analysis of Plant Parameters
Differences in plant growth performance, biomass, and photosynthetic index under the Epichloë endophytic status and different soil moisture levels were tested using a two-way analysis of variance (ANOVA) using the datarium package of R software. A statistically significant two-way interaction was followed up by simple main effect analyses; that is, evaluating the effect of endophytic status during each soil moisture treatment. All values are means ± SE of the mean.

Differentially Expressed Genes in Photosynthesis
A total of 462,911,295 clean reads were obtained from all samples, and 64.88-70.95% reads of each sample were mapped and used for further analysis (  (Table S4). The results indicated that expression of all unigenes differed between EI and EF plants regardless of non-drought and drought treatments ( Figures 1A and S1).  KEGG pathways (top 20) results indicated that these DEGs involved in the photosynthesis processes responded to the presence of the Epichloë endophyte in the CK treatment, such as for photosynthesis (ko00195), antenna proteins (ko00196), chlorophyll metabolism, carbon fixation in photosynthetic organisms, and other metabolites processes (Figure 1C). Similar unigenes of these DEGs are also reported in some crop and model plants in the sub-family Pooideae. The identity of these DEGs with reference genes was over 80%, from 80.18% to 100% (Table 1). There were 16 DEGs associated with the process of photosynthesis, such as photosystem Ⅱ (8 DEGs: psbB, psbC, psbE, two psbS, psbQ, and psb27), photosystem Ⅰ (3 DEGs: psaO, psaE, and psaG), photosynthetic electron transport (5 DEGs: petE, petF, two petH, and petJ), and F-type ATPase (1 DEG: atpH) ( Table 1 and KEGG pathways (top 20) results indicated that these DEGs involved in the photosynthesis processes responded to the presence of the Epichloë endophyte in the CK treatment, such as for photosynthesis (ko00195), antenna proteins (ko00196), chlorophyll metabolism, carbon fixation in photosynthetic organisms, and other metabolites processes ( Figure 1C). Similar unigenes of these DEGs are also reported in some crop and model plants in the sub-family Pooideae. The identity of these DEGs with reference genes was over 80%, from 80.18% to 100% (Table 1). There were 16 DEGs associated with the process of photosynthesis, such as photosystem II (8 DEGs: psbB, psbC, psbE, two psbS, psbQ, and psb27), photosystem I (3 DEGs: psaO, psaE, and psaG), photosynthetic electron transport (5 DEGs: petE, petF, two petH, and petJ), and F-type ATPase (1 DEG: atpH) ( Table 1 and Figure S3). There were 16 DEGs identified that were associated with the process of photosynthesis antenna proteins, including lhcB1 (10), lhcB2 (2), lhcB3 (1) and lhcB5 (2), and lhcB6 (1) unigenes (Table 1 and Figure S3).

Photosynthesis
As the consequence of the upregulation of the majority of photosynthesis DEGs, we assessed whether photosynthetic rates were higher in EI versus EF DHG plants. Twoway ANOVA results indicated that plant chlorophyll content and photosynthetic indices responded differently to the drought treatments and the Epichloë presence (Tables 2 and  S5). The effects of the presence of the Epichloë endophyte on leaf chlorophyll content depended on the soil moisture and symbiosis x soil moisture treatments: F (2,48) = 15.73, p = 0.000 ( Table 2). The Epichloë presence significantly increased the leaf chlorophyll content only in the MD and SD treatments in 19.7% (p = 0.000) and 7.1% (p = 0.040), respectively ( Figure 3A). The chlorophyll content in EI DHG plants was only significantly higher than in EF DHG plants under MD and SD treatments ( Figure 3A). The MD and SD treatments reduced the photosynthetic efficiency of DHG plants compared to the CK level of soil moisture: F (2,48) = 47.80, p < 0.001 ( Figure 3B). Epichloë presence increased the photosynthetic efficiency of DHG plants regardless of soil moisture treatments, symbiosis status: F (1, 48) = 14.08, p < 0.001, with increases of 17.2% (p = 0.000), 10.7% (p = 0.022) and 10.9% (p = 0.030) under CK, MD and SD treatments, respectively ( Figure 3B). The intercellular carbon dioxide concentration was significantly higher in EI versus EF DHG plants under two drought conditions, with an increase of 19% and 22% under MD and SD, respectively ( Figure S4A). The transpiration rate was only significantly lower in EI versus EF plants under non-drought conditions ( Figure S4C). drought and drought treatments ( Figure 2). Meanwhile, the majority of DEGs were upregulated in CK and SD treatments (Figure 2A,C,D).

Photosynthesis
As the consequence of the upregulation of the majority of photosynthesis DEGs, we assessed whether photosynthetic rates were higher in EI versus EF DHG plants. Two-way ANOVA results indicated that plant chlorophyll content and photosynthetic indices responded differently to the drought treatments and the Epichloë presence (Tables 2 and S5). The effects of the presence of the Epichloë endophyte on leaf chlorophyll content depended on the soil moisture and symbiosis x soil moisture treatments: F (2,48) = 15.73, p = 0.000 (Table 2). The Epichloë presence significantly increased the leaf chlorophyll content only in the MD and SD treatments in 19.7% (p = 0.000) and 7.1% (p = 0.040), respectively ( Figure  3A). The chlorophyll content in EI DHG plants was only significantly higher than in EF DHG plants under MD and SD treatments ( Figure 3A). The MD and SD treatments reduced the photosynthetic efficiency of DHG plants compared to the CK level of soil moisture: F(2,48) = 47.80, p < 0.001 ( Figure 3B). Epichloë presence increased the photosynthetic efficiency of DHG plants regardless of soil moisture treatments, symbiosis status: F(1, 48) =  The transpiration rate was only significantly lower in EI versus EF plants under nondrought conditions ( Figure S4C).

Plant Growth and Biomass
The plant performance and shoot/root/total biomass significantly responded to the presence of the Epichloë endophyte and drought treatments ( Table 1 and Table S5). The effect of Epichloë on plant height depended on the soil moisture, symbiosis x soil moisture treatments: F (2,48) = 17.27, p = 0.000 (Table S5). The Epichloë endophyte significantly increased the plant height only in the MD and SD treatments by 13.1% (p = 0.000) and 9.6% (p = 0.000), respectively ( Figure S1A). The effects of the Epichloë endophyte on total biomass depended on the soil moisture, symbiosis x soil moisture treatments: F (2,48) = 7.67, p = 0.001 (Table S5). The Epichloë endophyte significantly increased the total biomass in the CK, MD, and SD treatments in 12.7% (F (1,16) = 288.0, p = 0.000), 11.3% (F (1,16) = 58.9, p = 0.000) & 21.4% (F (1,16) = 88.1, p = 0.000), respectively ( Figure 3C). The drought treatments significantly decreased the tiller number of DHG plants compared with CK treatments regardless of symbiosis status, F (2,48) = 17.27, p = 0.000 ( Figure 3D). Here, the present results indicated that the Epichloë endophyte only had significant positive effects on the average per-tiller biomass under the MD and SD treatments ( Figure 3E). Additionally, the total biomass of EI and EF DHG plants was significantly (p < 0.001) and positively associated with the photosynthetic efficiency regardless of the presence or absence of the Epichloë endophyte ( Figure 3F).

Discussion
With plants exposed to a water deficit, production was linked to the net photosynthetic efficiency. Adaption mechanisms of plants responding to abiotic and biotic stresses include the forming of symbiotic associations with beneficial microbes. In this study, which we conducted to examine the effects of different levels of drought stress on A. ine-

Discussion
With plants exposed to a water deficit, production was linked to the net photosynthetic efficiency. Adaption mechanisms of plants responding to abiotic and biotic stresses include the forming of symbiotic associations with beneficial microbes. In this study, which we conducted to examine the effects of different levels of drought stress on A. inebrians plants with and without an Epichloë systemic endophyte, our results provide a comprehensive overview of unigene changes associated with photosynthesis processes ( Figure 5). We found that the majority of DEGs in photosynthesis were upregulated in Epichloë symbiotic plants and thus had a higher NPE than non-symbiotic plants ( Figure 5). Many studies have confirmed that the presence of an Epichloë endophyte promoted plant growth, biotic resistance, and abiotic tolerance of their host grasses [3,8,[36][37][38]. Studies have provided an understanding of how symbiotic microbes improved drought tolerance through different strategies [5,11,39,40]. The secondary metabolites produced or induced by the presence of an Epichloë endophyte contribute to a plant's abiotic/biotic tolerance [38,39]. Plant metabolite processes begin with the products of the photosynthesis process. The presence of an Epichloë endophyte in grasses stimulates the accumulation of plant metabolites in the aboveground tissues and in roots, such as SA, flavonoids, and total phenolic compounds [39,[41][42][43]. As we expected, the unigenes in flavonoids and fatty acids biosynthesis were differently expressed in response to the Epichloë endophyte.
The presence of Epichloë spp. in aboveground tissues altered the transcription levels of their host cool-season grasses [29,[44][45][46][47]. The expression of dehydrin and heat shock protein genes in F. arundinacea was enhanced by the symbiotic Epichloë endophyte in water-unstressed conditions [47]. Similarly, the majority of unigenes that differently responded to the Epichloë presence were found in the control, and in abiotic and biotic conditions such as those involved with SA biosynthesis [14,29]. As expected, this study also Many studies have confirmed that the presence of an Epichloë endophyte promoted plant growth, biotic resistance, and abiotic tolerance of their host grasses [3,8,[36][37][38]. Studies have provided an understanding of how symbiotic microbes improved drought tolerance through different strategies [5,11,39,40]. The secondary metabolites produced or induced by the presence of an Epichloë endophyte contribute to a plant's abiotic/biotic tolerance [38,39]. Plant metabolite processes begin with the products of the photosynthesis process. The presence of an Epichloë endophyte in grasses stimulates the accumulation of plant metabolites in the aboveground tissues and in roots, such as SA, flavonoids, and total phenolic compounds [39,[41][42][43]. As we expected, the unigenes in flavonoids and fatty acids biosynthesis were differently expressed in response to the Epichloë endophyte.
The presence of Epichloë spp. in aboveground tissues altered the transcription levels of their host cool-season grasses [29,[44][45][46][47]. The expression of dehydrin and heat shock protein genes in F. arundinacea was enhanced by the symbiotic Epichloë endophyte in waterunstressed conditions [47]. Similarly, the majority of unigenes that differently responded to the Epichloë presence were found in the control, and in abiotic and biotic conditions such as those involved with SA biosynthesis [14,29]. As expected, this study also detected a large number of DEGs in EI plants versus EF DHG plants under the three soil moisture treatments, and the DEGs detected in the present transcriptome data included some involved in photosynthesis (PSI, PSII, and PET), which are in line with our hypothesis. Photosynthesis begins with harvesting light within leaves, and the present transcriptome data indicated that genes in antenna proteins and the chlorophyll metabolism process were upregulated in response to the Epichloë endophyte, and a higher chlorophyll content was found in EI versus EF DHG plants. The present results are also supported by two studies [19,48]. The abundance of lhcI and lhcII proteins in EI D. glomerata plants was higher than that of EF plants [19]. Meanwhile, the genes of lhcI type II were upregulated in the E. festucae-infected F. rubra compared to EF plants [48].
Ambrose and Belanger (2012) also noted that some genes involved in the photosynthesis process of red fescue (F. rubra) are upregulated and downregulated in response to the presence of an Epichloë endophyte [48]. The Epichloë endophyte increased the rate of carbon assimilation, PSII photochemistry, and grass biomass associated with D. glomerate plants [19]. A study showed that the 1000 D7 gene (CP47) was downregulated in perennial ryegrass symbiotic with E. festucae var. lolii [44]. Similar to the study of Epichloë symbiotic perennial ryegrass, our results indicated that a unigene (c61885.graph_c1, encoding PSII CP47 reaction center protein) was downregulated in the EI DHG plants regardless of nondrought and drought treatments. Another study found that the electron transport rate was enhanced by 31% in EI plants and reduced by 13% in EF plants under water stressed versus CK treatments [49]. As we expected, one (c51525.graph_c1, petH) and two (c47702.graph_c0, petF; c46095.graph_c0, petJ) upregulated unigenes were only in EI DHG plants with under drought and non-drought treatments, respectively. This is in line with a study that indicated that the NADPH activity in EI D. glomerata plants was significantly greater (c. 4.28) than in EF plants [19]. Our results and present meta-analysis indicated that the Epichloë endophyte promoted the photosynthetic rate of host plants ( Figure 5). Meanwhile, the presence of the Epichloë endophyte on the net photosynthesis rate of L. perenne is independent of endophyte concentration in planta [18].
Greater photosynthetic rates are commonly associated with higher production, and our results showed the liner relationship between photosynthetic rates and the total biomass of EI and/or EF DHG plants. In line with our hypothesis and certain studies, the presence of an Epichloë endophyte increased the biomass and photosynthetic rate of host plants under drought [12,16,25,26]. Another study found that EI F. arizonica plants produced more shoot biomass and had a greater plant growth rate versus EF plants under low water availability [10]. Our results have shown that the presence of the Epichloë endophyte had no effect on the per-tiller biomass under non-drought treatment, while it did, however, promote the per-tiller biomass accumulation under (MD and SD) drought conditions. This study provides an enhanced understanding of the enhancement of drought tolerance in Epichloë symbiotic plants ( Figure 5).

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/jof8050512/s1, Figure S1: Principal component analysis (PCA) of all unigenes from Epichloë symbiotic (EI) and non-symbiotic (EF) Achnatherum inebrians plants under normal (CK), moderate (MD) and severe (SD) drought treatments. Figure S2: The number of up-and down-regulated differently expressed unigenes (DEGs, FC > 1.5) of Achnatherum inebrians plants under the several (SD) and moderate drought (MD) treatments compared to the normal (CK) treatment (DEGs in endophyte-infected plants versus endophyte-free plants). Figure S3: The Neighbor-Joining (NJ) tree showing the amino acid sequences of differently expressed genes (DEGs) associated with photosynthesis (A) and photosynthesis-antenna proteins (B,C) of Achnatherum inebrians plants. All bootstrap values > 70% are shown (1000 replicates). Numbers above branches indicate the bootstrap values of the maximum likelihood analysis. Figure S4: The intercellular carbon dioxide concentration (A), transpiration rate (B) and stomatal conductance (C) of Epichloë symbiotic (EI) and non-symbiotic (EF) Achnatherum inebrians plants under normal (CK), moderate (MD) and severe (SD) drought treatments. The asterisk (*) means significant difference at p < 0.05 (independent t-test) between and EI and EF plants at corresponding water content at 0.05 level. The A and B mean significant differences among corresponding water content at 0.05 level. Figure S5: The fresh weight of shoot (a) and root (b), and dry weight of shoot (c) and root (d) of Epichloë symbiotic (EI) and non-symbiotic (EF) Achnatherum inebrians plants under normal (CK), moderate (MD) and severe (SD) drought treatments. The asterisk (*) means significant difference at p < 0.05 (independent t-test) between and EI and EF plants at corresponding water content at 0.05 level. The A and B mean significant differences among corresponding water content at 0.05 levels. Table S1: Selected unigenes associated with processes of photosynthesis and photosynthesis-antenna proteins identified in the RNA-seq analysis in the present study. Table S2: The comparative statistics between RNA sequencing clean data and transcriptome assembly of Epichloë symbiotic (EI) and non-symbiotic (EF) Achnatherum inebrians plants under normal (CK), moderate (MD) and severe (SD) drought treatments. Table S3: Length distribution of transcripts and unigenes of Achnatherum inebrians plants.