Integrated soybean transcriptomics, metabolomics, and chemical genomics reveal the importance of the phenylpropanoid pathway and antifungal activity in resistance to the broad host range pathogen Sclerotinia sclerotiorum

Sclerotinia sclerotiorum, a predominately necrotrophic fungal pathogen with a broad host range, causes a significant yield limiting disease of soybean called Sclerotinia stem rot (SSR). Resistance mechanisms against SSR are poorly understood, thus hindering the commercial deployment of SSR resistant varieties. We used a multiomic approach utilizing RNA-sequencing, Gas chromatography-mass spectrometry-based metabolomics and chemical genomics in yeast to decipher the molecular mechanisms governing resistance to S. sclerotiorum in soybean. Transcripts and metabolites of two soybean recombinant inbred lines, one resistant, and one susceptible to S. sclerotiorum were analyzed in a time course experiment. The combined results show that resistance to S. sclerotiorum in soybean is associated in part with an early accumulation of JA-Ile ((+)-7-iso-Jasmonoyl-L-isoleucine), a bioactive jasmonate, increased ability to scavenge reactive oxygen species (ROS), and importantly, a reprogramming of the phenylpropanoid pathway leading to increased antifungal activities. Indeed, we noted that phenylpropanoid pathway intermediates such as, 4-hydroxybenzoate, ferulic acid and caffeic acid were highly accumulated in the resistant line. In vitro assays show that these metabolites and total stem extracts from the resistant line clearly affect S. sclerotiorum growth and development. Using chemical genomics in yeast, we further show that this antifungal activity targets ergosterol biosynthesis in the fungus, by disrupting enzymes involved in lipid and sterol biosynthesis. Overall, our results are consistent with a model where resistance to S. sclerotiorum in soybean coincides with an early recognition of the pathogen, leading to the modulation of the redox capacity of the host and the production of antifungal metabolites. Author Summary Resistance to plant fungal pathogens with predominately necrotrophic lifestyles is poorly understood. In this study, we use Sclerotinia sclerotiorum and soybean as a model system to identify key resistance components in this crop plant. We employed a variety of omics approaches in combination with functional studies to identify plant processes associated with resistance to S. sclerotiorum. Our results suggest that resistance to this pathogen is associated in part with an earlier induction of jasmonate signaling, increased ability to scavenge reactive oxygen species, and importantly, a reprogramming of the phenylpropanoid pathway resulting in increased antifungal activities. These findings provide specific plant targets that can exploited to confer resistance to S. sclerotiorum and potentially other pathogens with similar lifestyle.

INTRODUCTION million raw reads were generated, with 95.7 -96.6% of reads mapping to the host reference 148 genomes of soybean and S. sclereotiorum at all timepoints. On average, 96% of the total reads 149 mapped uniquely to the soybean reference genome in the uninfected plants of both lines. In the S 150 line, 91.6, 91.8, and 68.9% of the reads mapped to the soybean genome at 24, 48 and 96 hpi, 151 respectively. In the R line, 92.8, 91.9, and 88.2% of the reads mapped to the soybean genome at 152 24, 48 and 96 hpi, respectively (Table 1) (Table S4). Each metabolite was 210 characterized by its distinct retention time and mass to charge ratio (m/z). All the 164 identified 211 metabolites found in infected soybean were also detected in non-inoculated stems and therefore 212 are likely of plant origin. Despite this, we note that as disease progresses, contributions from S. 213 sclerotiorum cannot be ruled out. 214 MetaboAnalyst 3.0 (35) was used for the analysis of the 164 identified metabolites. One-metabolites, the accumulation of which was significantly (FDR <0.05) affected by S. sclerotiorum 217 inoculation at 48 and 72 hpi in R-line compared to S-line (Table S5). These metabolites included 218 polar compounds, such as nucleotides, amino acids, alcohols, organic acids, and carbohydrates, 219 along with nonpolar compounds including fatty acids, and long-chain alcohols (Table S5). The 220 multivariate analysis of identified metabolites was performed using Partial Least Squares -  Supplementary Table 5. Significantly regulated metabolites were assigned 226 to distinct functional categories according to the chemical groups to which they belong (Table S5) 227 and to specific plant pathways in which they may function (Table S6). Our data revealed several 228 differentially regulated metabolic processes between the R and S lines (Fig. 4). However, those 229 involved in phenylalanine metabolism are particularly interesting given the differential expression transcriptomics and metabolomics data with respect to the differential modulation of the 236 phenylpropanoid pathway indicates its potential key participation in resistance to S. sclerotiorum. 237 Similar to transcriptomic data, the comparative metabolite profiles also implicated 238 phytohormones in this interaction. Namely, the fatty acids linolenic acid (a precursor of jasmonic acid) and cyanoalanine (an indicator of ethylene biosynthesis) (36,37) are both significantly 240 induced at 48 and 72 hpi, in the R line (Fig. S3). Interestingly, the most highly upregulated 241 metabolite in our R line is mucic acid with an ~86-fold higher accumulation (Table S5). Mucic 242 acid, also referred to as galactaric acid, can be produced by the oxidation of d-galacturonic acid, 243 the main component of pectin. 244 Galactose metabolism and the TCA cycle were the pathways most affected by this analysis, 245 and the metabolites assigned to them were primarily carbohydrates and organic acids, respectively. 246 Interestingly, of the metabolites downregulated in R plants in comparison to S plants, 81.8% (9/11) 247 belonged to one of these two groups (Table S5). Although these metabolites relate to multiple 248 pathways, their downregulation in R plants may be a strategy to reduce S. sclerotiorum access to 249 preferable carbon sources. Of the top six potentially affected pathways, three of them (Glyoxylate 250 and dicarboxylate metabolism; Alanine, aspartate and glutamate metabolism; the TCA cycle) 251 contain both fumaric and succinic acid. These organic acids were downregulated at 48 and 72 hpi, 252 respectively, and demonstrate the potentially broad impacts that changes in individual metabolites 253 can have on plant biosynthesis and metabolism.

254
Reprogramming of the phenylpropanoid pathway in resistance to S. sclerotiorum. 255 Many secondary metabolites derived from multiple branches of the phenylpropanoid 256 pathway, including lignin, isoflavonoid-phytoalexins, and other phenolic compounds such as 257 benzoic acid, have been proposed as important components of defense responses (38,39). In this 258 study, we found differential expression of transcripts and metabolites related to the 259 phenylpropanoid pathway between the R and S soybean lines. At the transcript level, we observed  Table 2 and Fig. 5A). Transcript levels of select genes within these pathways were validated 270 using RT-qPCR, thus confirming the RNA-Seq results (Fig. 6).

271
In accordance with the transcriptomics data, we observed a marked accumulation of the 272 metabolites phenylalanine, a phenylpropanoid pathway precursor, and the lignin intermediates  (Table 2 and Fig. 5A). We reasoned that the contribution of these ferulates to resistance 278 may be due to their antifungal activity, and their ability to inhibit S. sclerotiorum growth was tested 279 in vitro. A significant reduction of fungal growth was observed when S. sclerotiorum was grown 280 on PDA with increasing concentrations of ferulic acid (Fig. S2). While caffeic acid did not 281 significantly affect colony size, it clearly affected S. sclerotiorum growth patterns on PDA with 282 the appearance of abnormal concentric ring growth patterns and premature sclerotia formation 283 (Fig. S2).
We mined our metabolomics data for differentially accumulated metabolites that may serve 308 as ROS scavengers or antioxidants. Dehydroascorbic acid (DHA), the oxidized form of ascorbate, 309 an important antioxidant, was specifically accumulated later in the infection time course (72 hpi), 310 but not at the early stages in the R line (Fig. 7). Similarly, the proline derivative, trans-4-hydroxy-311 L-proline, a known osmoprotectant and antioxidant (43,44) is significantly accumulated in the R 312 line at the later stages of the infection process (Fig. 7). Proline plays a major role as an antioxidant, 313 owing to its ROS scavenging capacity (45,46). Overall, these results and the earlier observation of  intriguing. This resistant response was also associated with the development of a prominent red 349 coloration at the site of infection (Fig. 1). Considering the metabolomics data, we reasoned that 350 total stem extract from the resistant line should also exhibit antifungal activity against S. 351 sclerotiorum. To test this hypothesis, we prepared an ethanol extract of stem sections harvested 352 from the R line,10 days post inoculation, that we termed the red stem extract. Equal amount of green stem extracts harvested from non-inoculated plants served as control. The final extracts were 354 diluted in DMSO following ethanol evaporation. S. sclerotiorum growth was assayed on potato 355 dextrose broth (PDB) amended with the red stem extract, green stem extract, or DMSO control for 356 48 hours. Fungal biomass as determined by mycelial fresh weight was markedly reduced (12-14 357 fold) in PDB cultures containing the red stem extract compared to PDB amended with the green 358 stem extract or DMSO (Fig. S4 B, C). These results confirm that the resistant response associated 359 with our R line clearly involves the accumulation of antifungal compounds that inhibit S. 360 sclerotiorum growth. 361 We next examined the mechanism by which the red stem extract inhibits fungal growth by in the ergosterol biosynthetic pathway, had the greatest sensitivity to the extract, and this was a 369 highly significant response (p<1e-7). ERG2, which is also involved in ergosterol biosynthesis, was 370 also significantly sensitive (p<0.01). Mutants of ARO7, which encode a gene involved in amino 371 acid biosynthesis was significantly sensitive. While ARO7 is not known to be directly involved in 372 lipid/sterol biosynthesis, it has many genetic interactions with lipid related genes (54). CHO2 and 373 OPI3 mutants were also significantly sensitive. Cho2p and Opi3p are both involved in 374 phosphatidylcholine biosynthesis. A deletion mutant of PAH1 was the most significantly resistant 375 mutant. Pah1p is a phosphatase that regulates phospholipid synthesis. Deletion mutants of PAH1 have increased phospholipid, fatty acid and ergosterol ester content (55). Further, in two out of 377 three replicates, the chemical genomic profile of the red stem extract had significant correlation 378 (p<0.05) with the profile of fenpropimorph (56), an ergosterol biosynthesis inhibitor that targets 379 ERG2 and ERG24 in yeast. Taken together, these data suggest that the red stem extract may exert 380 toxicity by either disrupting enzymes involved in lipid/sterol biosynthesis, or alternatively 381 physically binding membrane lipids and causing cell leakage.

399
Resistance to fungal pathogens with a predominately necrotrophic lifestyle, such as 400 Sclerotinia sclerotiorum, is not well understood due to the likely complex network of responses to 401 these pathogens or their determinants. Uncovering key components of these defense responses is 402 essential for the deployment of disease resistant crops. Omics approaches offer a unique 403 opportunity to identify global cellular networks in plants in response to these pathogens. S. 404 sclerotiorum is a broad host pathogen that infects over 400 species, mostly dicotyledonous plants. 405 Its pathogenic success most assuredly relies on a broadly effective toolkit that allows it to infect 406 multiple hosts, however, host-specific dialogue between a given host and the pathogen may also 407 be of importance. In this study, we specifically examine soybean resistance mechanisms against Corp., Carlsbad, CA, USA). Briefly, collected tissue from each sample was finely ground in liquid 557 nitrogen. For each 100 mg of tissue, 1 ml of chilled Trizol was added. Samples were centrifuged at 12k rpm for 5 min at 4℃. Supernatant was discarded and 200 µl of chilled chloroform was 559 added, and vortexed at high speed for 15 sec. Samples were centrifuged again at 12k rpm for 15 560 min at 4℃. The aqueous phase was mixed with 0.8x isopropanol and left at room temperature for 561 10 minutes, followed by centrifugation at 12k rpm at 4℃. Supernatant was discarded, and the pellet 562 was washed with 75% ethanol. Pellet was air dried for 10 minutes at room temperature and 563 resuspended in 20µl of nuclease free water followed by incubation at 55℃ for 10 minutes. Samples   ramped to 250°C by 12°C s-1. A helium mobile phase was applied at a flow rate of 1 ml min -1 .

615
Column temperature was 50°C for 1 min, ramped to 330°C by 20°C min-1 and held constant for

Compound extraction from soybean stem
Five hundred mg of infected red stem or unaffected green stem was mixed 1:1 w/v in 100% ethanol 673 at 80°C for 1 h. Samples were resuspended with 100 µl of DMSO and used for S. sclerotiorum 674 inhibition assay, chemical genomics, and cell permeability assays.

676
Chemical genomic analysis 677 Chemical genomic analysis of the red stem extract was performed using the non-essential yeast 678 deletion mutant collection as described previously (53). Briefly, triplicate 200 µL cultures of the 679 pooled deletion collection were exposed to a 1:10 dilution of the red stem extract and allowed to            each pathway which were found to be significantly regulated in this study. Upregulated = Green.