Metabolomics Analysis of Tilletia Controversa Kühn Infected and Non-infected Grains of Wheat


 Dwarf bunt caused by the pathogen Tilletia controversa Kühn is one of the most serious quarantine disease of winter wheat. Metabolomics studies provide detailed information about biochemical changes at the cell and tissue level of the plants. In the present study, liquid chromatography/mass spectrometry (LC/MS) metabolomic approach was used to investigate the changes in the grains metabolomics of T. controversa infected and non-infected samples. PCA analysis suggested that T. controversa infected and non-infected samples scattered separately during the interaction. LC/MS analysis showed that 62 different metabolites were recorded in the grains, among them total of 34 metabolites were up-regulated and 28 metabolite were down-regulated. The prostaglandins (PGs) and 9-hydroxyoctadecaenoic acids (9-HODEs) are fungal toxin related substances and their expression significantly increased in T. controversa infected grains. Additionally, the concentration of cucurbic acid and octadecatrienoic acid were changed significantly after pathogen infection, which has great role in plant defense. The eight different metabolic pathways activated during the T. controversa and wheat plants interactions including phenylalanine metabolism, isoquinoline alkaloid biosynthesis, starch and sucrose metabolism, tyrosine metabolism, sphingolipid metabolism, arginine and proline metabolism, alanine, aspartate, glutamate metabolism, and tryptophan metabolism.


Introduction
Wheat is one of the most important staple crops and plays a fundamental role in food security worldwide 1 . Wheat production was often negatively affected by the infection of a wide variety of pathogens. Wheat dwarf bunt (caused by a fungal pathogen Tilletia controversa Kühn) is a destructive disease which causes signi cant quality and quantity losses in wheat growing regimes 2 . Metabolomics is an omics technology for comprehensive evaluation of small, endogenous molecules (such as nucleotides, organic acids, sugar and amino acids) which helps to analyse the interrelationships between genetic structure, gene expression, protein function, and environmental impact 3 . These compounds are the substrates and by-products of cell processes (such as enzymatic reactions) and, as such have a direct effect on phenotypes. The sub-discipline of plant metabolomics is a growing research in the plant-microbe interaction area 4,5 . Plant defense mechanisms can be regulated by metabolomics in fungus-infected plants. When plant infected by the pathogen, the plant activates a multi-component biochemical and physical response involving radical changes in the expression patterns of proteins, genes and metabolites 6,7 . Metabolites are the end products of translation and transcription; therefore, the changes in metabolites abundance may be regarded as a major feature of plant interactions with pathogen and environment 3,8 Previous studies showed that wheat metabolites were different in F. graminearum infected and normal plants 9 . Magnaporthe oryzae altered the alanine contents in rice compared to normal rice leaves 10 . Doehleman 11 investigated the metabolites in maize tumor after U. maydis infection. Similarly, metabolic pro ling strategies were used to determine the mechanisms of plant defense against R. solani in soybean, rice, lettuce, and potato [12][13][14][15] . Metabolomics analysis have been performed in B. cinerea infected tomato, strawberry, Arabidopsis and grapes 16,17 . GC-MS based metabolomics approach was done on the S. sclerotiorum infected susceptible and resistance cultivars of soybean. Results of this study showed that antifungal activity increased in the resistant cultivar by reprograming the phenylpropanoid pathway 18 . Similarly, untreated LC-MS metabolomics strategy was performed to investigate metabolome alterations in the anthracnose causing C. sublineolum 19,20 . Additionally, GC-ESI-MS/MS technique was used in wheat crop infected by S. nodorum and found that the concentration of secondary metabolites was 200 higher in the mutants strain compared to wide 21 . Metabolomics studies are very useful in plant sciences because it offers the ability to recognise biochemical alterations relatively quickly, usually before any change in the phenotypes of crops 10 .
Plant-pathogen relationships are extremely interesting in terms of both biological importance and metabolite richness and thus an ideal area for exploration using metabolomics techniques 10 . Liquid chromatography-mass spectrometry (LC-MS) is a sensitive tool for metabolic pro ling that perfectly compensates for this defect and has become an important research method in the eld of metabolomics 22,23 . LC-MS-based analysis aims to compare multiple biological groups to detect metabolites that have changed signi cantly after pathogen infection. In the present work LC-MS technique was used to study the response of metabolites in T. controversa infected grains (galls) and normal grains. Clear differences were obtained in metabolic pro les of wheat grains after T. controversa infection. According to our knowledge, it is the rst study to evaluate the metabolites in wheat grains after T. controversa.

Plant material and pathogen inoculation
Wheat (Triticum aestivum L) cv Dongxuan 3 seeds were collected from Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing-China. Wheat seeds were sterilized with 30% sodium hypochlorite for 5 min, rinsed 5 times in water, and germinated for 30 days in incubator (AUCMA, Qing Dao, China) to vernalize. After vernalization, seedlings were grown in a 1:2 mixture of organic matter (peat moss, Beijing, China) and soil (Beijing, China) in pots. The pots were kept in incubator (AUCMA, Qing Dao, China) under a 14 h light: 10 h dark (8-10 ˚C, 70% humidity) regime and temperature was increased to 20 ˚C during boot stage. The pathogenic fungus T. controversa Kühn was gifted by Blair Goates (the United States Department of Agriculture (USDA), Agricultural Research Service (ARS), Aberdeen, Idaho, USA). T. controversa was propagated in soil agar medium (20 g agar powder, 75 g soil in one liter distilled water) and autoclaved for pouring the media into sterilized plates. After pouring, the plates were incubated at 5 °C in a 24 h light in incubator (MLR 352H, Panasonic, USA) after covering with para lm for 60 days incubation. Mycelium production were observed under an automated invert uorescence microscope (IX83, Olympus, Japan). Hyphae were collected with distilled water and used to inoculate wheat plants. During the boot stage, hyphae was injected into spike with the syringe. The hyphae was inoculated twice a day (5 mL per spike) and consecutively for 5 days, while only ddH 2 O was used for control plants. The pathogen infected grains (galls) and normal grains was harvested and quickly dipped in liquid nitrogen and stored at -80 ˚ C for further use. Nine replications of each treatment was used for reproducibility.

Sample Processing
Metabolites were extracted from a 50 mg of crushed grains using ultrapure (Watsons, China) water. The choice of extraction method is an important factor in any metabolomics study. The grains were crushed by following the developed method of our laboratory. Brie y, two-three grains were dipped into 2 ml Eppendorf tube containing sterilized steel ball (1.5 mm) with appropriate amount of steel balls and grinded in grinder machine (FastPrep 24 5G, MP Biomedicals, USA) for I minute at 70 Hz for crushing.
Every sample grinded 3 times for better results. Powdered tissue sample (50 mg) were rst mixed with 500 µl of a ultrapure water (1:10 w/V) and sonicated for 10 min. After this time, 50 µl was taken from homogenate mixture and added into 450 µl of precipitant containing internal standard (methanol: acetonitrile = 1:1). The samples were then vortex for 1 min, centrifuging at 13000 rpm for 10 min. The nally resultant product (100 µL) was then transferred to separate new sterilized Eppendorf tubes for LC-MS analysis.

LC-MS analysis
LC-MS analysis of the precipitant was done in 450 µl of a methanol-acetonitrile mix. Analysis was done using a liquid chromatograph mass spectrometry (Dionex ultimate 3000, Thermo Fisher) with a C18 column (2.1 × 100 mm, 3 µm particle size, Waters). After a 5 µl of sample injection chromatographic analysis was achieved with a liquid phase of 2 mmol / L ammonium acetate and 0.1% formic acid in water (A) and acetonitrile (D). The detailed gradient elution program was shown as followed, 95% (A) and 5% (D) to 0-2 min; 5% (A) and 95% (D) to 42-47 min; 95% (A) and 5% (B) to 47.10-50 min.

Data processing
All LC-MS dates were further ltered by using the R platform loaded with the xcms tool kit including; peak matching, retention time correction, variable integration and data standardization. The preprocessed data was then analysed by multivariate analysis, including principal components analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) by EZinfo software, and hierarchical cluster analysis (HCA) was performed by MetaboAnalyst 4.0 software. Metabolic pathways were further analyzed using KEGG (http://www.genome.jp/kegg/). The correlation analysis of differential metabolites was calculated by using R software and Cytoscape software was used for network construction.

Metabolite identi cation-LC-MS
The chromatogram of the positive ions and negative ions of wheat grains extract respectively. The total ion concentration of positive ions (Fig. 1a) and negative ions (Fig. 1b) demonstrated with different time intervals using high-resolution mass spectrometry in different time intervals. The peak of positive and negative ions was obtained within 25-30 minutes (time interval) during analysis. The qualitative relative abundance of aspartic and its molecular obtained by using LC-MS (Fig. 1 cd). Additionally, all metabolite was found and their statistical signi cance were demonstrated in Table 1.

PCA analysis of T. controversa infected and control samples
Principle component analysis score plot exhibited good clustering between T. controversa infected and non-infected samples. Results showed that the distribution of T. controversa infected and non-infected metabolites in these two groups was signi cantly different (Fig. 2a). OPLS-DA plot revealed that the samples of T. controversa was mainly distributed on the right side and non-infected samples distributed left side (Fig. 2b). Receiver operating characteristic (ROC) curve was generated by using Monte Carlo cross validation (MCCV) under balanced sub sampling. In each MCCV, two-thirds of the samples was used to assess the importance of features. Results showed that metabolites successfully detected under ROC curve (Fig. S1). The list of metabolites and their signi cant difference on the base of VIP value demonstrated in Table 1. According to results presented in Table 1 and in Fig. 2b, 62 differential metabolites were screened. Out of the 62 metabolites, 34 were up-regulated and 28 were down-regulated.

Screening and statistical analysis of differential metabolites
The correlation between T. controversa infected and non-infected samples demonstrated in Fig. 3a. Red color indicates the positive correlation while blue color indicates the negative correlation. Results showed that metabolites in the same branch along horizontal axis had strongest positive correlation. Additionally, metabolites among the control and infected grain groups were seen in the hierarchical clustering analysis (Fig. 3b).

Metabolic pathways analysis
In uence coe cient of the metabolic pathways, the pathway with the impact value higher than 0.1 was found to be a potential target pathway. As a result, phenylalanine metabolism, isoquinoline alkaloid biosynthesis, starch and sucrose metabolism, tyrosine metabolism, sphingolipid metabolism, arginine and proline metabolism, alanine, aspartate and glutamate metabolism and tryptophan metabolism were closely related pathways in this study (Table 2 and Fig. 4).

Discussion
Wheat is one of the major staple food crops worldwide. Wheat crops have evolved e cient mechanisms to inhibit the negative effects of pathogen attack 24 . The critical differences between plants that become diseased or remain healthy after pathogen infection is the recognition of elicitors molecules released by the pathogens 25,26 . The earliest response in plants after pathogen infection is due to the oxidative burst that can triggered hypersensitive cell death. This is celled the hypersensitive response (HR) and is one of the basic cellular response following successful pathogen recognition in plants and a major element of plant disease resistance 27 . Therefore, any change in the primary and secondary metabolites of plants in response to pathogen infection may thus be the key difference in the successful defence against pathogen.
Proline is a multi-functional amino acid which confer resistance against plant pathogens 28,29 and abiotic factors 30,31 . Resistance plants increase the proline accumulation against pathogen infections and vice versa in the susceptible plants 28 . Similarly, cucurbic acid, a compound similar to JA molecule actively involved in the defense mechanisms and tuberization in different crops 32 . Our results showed the concentration of proline, hydroxyproline, hydroxymethylproline and cucurbic acid decreased after T. controversa infection ( Table 1). The decreasing of the proline contents is the common reaction and energy demand increase for cross-signaling in plant pathogen interaction. It is also possible that elicitor molecules here not recognized by susceptible plants thus cause decrease in proline and cucurbic level, which can facilitate the pathogen infection. Prostaglandin molecule act in multi-dimensional manner in the fungi, but its biological function in the metabolism of fungi still unclear 33,34 . However, the impact of prostaglandin on fungal pathogenesis is still not well clear, but the concentration of prostaglandin increased during the pathogen infection, which has resemblance with our results (Table 1). Prostaglandin used as signal molecule to induce pathogen infection 35 . Previous study reveals that the silenced of tripleppo (gene have role in prostaglandin production) mutant of Aspergillus fumigatus was more virulent comparing to A. fumigatus wild-type in the infected murine test model. The A. fumigatus mutant showed higher resistance compared to A. fumigatus wild against reactive-oxygen species (ROS) produced by the plants for strong defense system after A. fumigatus infection 34 . Additionally, prostaglandin have role to increase other fungal metabolite including tyrosine, phenylalanine and other secondary metabolites 36,37 . The 9-hydroxyoctadecaenoic acids (9-HODEs) molecules are oxidation products of linoleic acid molecules that have role in the defense mechanism and increased plant growth as well as promote a atoxin B1 in Aspergillus parasiticus and sterigmatocystin (ST) from Aspergillus nidulans [38][39][40] . The 9-HODEs also act as a quorum sensing signal molecule to regulate the development and growth of Aspergillus ochraceus 41 . Our results showed that during T. controversa and wheat interactions the concentration of 9-HODE molecule signi cantly increased suggest that T. controversa effects the wheat grains.
An important and preliminary character of metabolomics data is that it can measure a huge number of metabolites in a quantitatively trustworthy way and therefore allows for discovery of correlations between pair of metabolites, even when biological connection between them are not clear. In our results, the metabolites from the grains, little strong correlation was noted. However, for some metabolites there was a positive or negative correlation in the heatmap analysis (Fig. 3ab). Additionally, the relative abundance of pathogen infected and control groups scattered separately, which may suggest that grains metabolites infected by the T. controversa. For example, the metabolites that are decreased or increase after pathogen infection are involved in the energy metabolism. These changes in the plant metabolites may be the result of T. controversa infection.

Declarations
Author contributions LG conceived the experiment work, ZR performed the experiment analysis, GM drafted the experiment, JL,TL and WC revised critically the manuscript.

Declaration of competing interest
Authors declare that there is no competing interest between them  The hierarchical clustering heatmap visualizing the changes in the contents of potential metabolites in T. controversa infected and non-infected samples. a correlation analysis of different metabolites. Deeper color had strong correlation and lighter color had weak correlation. b hierarchical cluster analysis of different metabolites. Red color showed that metabolites increased and blue color showed that metabolites decreased. Figure 4