Source: TEXAS A&M UNIVERSITY submitted to NRP
MANAGING PLANT MICROBE INTERACTIONS IN SOIL TO PROMOTE SUSTAINABLE AGRICULTURE
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1019749
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
W-4147
Project Start Date
May 29, 2019
Project End Date
Sep 30, 2023
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
TEXAS A&M UNIVERSITY
750 AGRONOMY RD STE 2701
COLLEGE STATION,TX 77843-0001
Performing Department
Plant Pathology & Microbiology
Non Technical Summary
Members of the fungal genus Trichoderma are well known for providing multiple beneficial effects to their hosts including enhanced plant growth and development, induced systemic defense (IRS), and defense against pathogens through mycoparasitism. Ingress and colonization of plant roots is often the pivotal event that initiates these beneficial effects. For instance, the elicitor SM1 is secreted by T. virens within the rhizosphere and root to prime the plant for an induced defense response (ISR). Colonization of the root by T. virens provides the fungus with a source of sucrose and enables the endophyte to attack invading pathogens by producing novel secondary metabolites and cell wall degrading enzymes. To enhance the ability of T. virens to promote these many beneficial effects requires a deeper understanding of the genes and their products that are essential for colonization. We have developed databases of the interaction of T. virens with maize roots over time as well as with maize mutants that react differently than the wild type to T. virens. These databases can be analyzed to biologically illustrate which genes are critically involved in root colonization and ISR. In addition, we will use mathematical approaches to develop gene connections to reveal networks that influence beneficial effects.
Animal Health Component
15%
Research Effort Categories
Basic
70%
Applied
15%
Developmental
15%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
2154020104070%
2151510116030%
Goals / Objectives
To determine how microbial populations function to suppress disease and how plants and the environment relate to this function.
Project Methods
Trichoderma virens is a an endophyte that efficiently colonizes roots of many plant species resulting in plant growth promotion, direct protection of plant roots from pathogens through mycoparasitism, and indirect protection of plants by the priming induced systemic resistance (ISR). Understanding the genes and their products that enable T. virens to penetrate and colonize live roots will result in the development of enhanced biocontrol strains, identification of new signaling molecules for direct application on plants, and provide a foundation of the genes the interact with other microbes in the rhizosphere. To achieve this understanding, the following sub-objectives are proposed.Sub-objective A. Select gene targets from Trichoderma virens that are involved in the early colonization of maize roots via RNA-seq time course analysis. We have a unique set of transcriptomic data that includes the interaction of the wild-type strain of T. virens and two mutants, Δsm1 (impaired in priming ISR) and Δsir1 (enhances ISR), with the inbred maize hybrid, B73, and two mutants in maize oxylipin production. The interaction events occurred in a hydroponic system with sampling at 6 and 30 hours. These data were developed under a USD-NIFA grant (2016-67013-24730), and the first set of analyses have been published (Malinich et. al. 2019, BMC Genomics 20:280). We are in the process of analyzing a second set of transcriptomic data collected from wild-type T. virens with maize hybrid Silver Queen in a similar hydroponic system, but sampled at 6, 12, 15, 24 and 36 hours. In both data sets, the experimental approach has been to define differentially expressed genes followed by gene characterization by programs including Kyto Orthology of Genes, Gene Ontology, Carbohydrate-Active enZymes Database, and protein domains as annotated in Joint Genome Institute expression analysis version 2.0 T. virens Gv29-8 catalogue. Gene expression identified by RNA-seq was validated with quantitative reverse-transcriptase polymerase chain reaction of selected genes representing a log2fold change. In addition, we have developed maps of signature genes involved in secondary metabolism, which allowed us to query their specific expression. We are in the process of completing the analysis of the second database, which will be used as a comparison with the first set to determine complementary differentially expressed genes (DEGs) for a "rack and stack" target list.Sub-objective B. Identify genes with a role in root colonization and IRS. As we have temporal sets of DEGS, we have the opportunity to define those expressed during recognition of the maize roots (~6 hours), early events of ingress (12 to 24 hours), and expansion of colonization (>24 hours). Two mutants of maize, lox3 and lox10, are phenotypically characterized as constitutively demonstrating ISR and induced systemic susceptibility (ISS) in the presence of wild type T. virens, respectively. Using this two maize mutants will enable us to examine for the presence of T. virens genes involved in these ISR and ISS. While these efforts represent a best biological "rack and stack" approach of defined processes, we intend to employ an additional mathematical approach that can define gene connections. We computationally seek to identify key functional genes (hub genes) as well as functional subnetwork modules. This approach will be conducted in collaboration with Dr. Mansuck Kim, who has developed algorithms for this approach and has published this approach with Fusarium verticillioides (Kim et. al., 2015, IEE/ACM Trnasction on Bomputational Biology and Bioinformatics PP 99 and Kim et. al., 2015 BMC Bioinformatics 16:S12). For the first dataset, we will construct T. virens co-expression networks with RNA-seq data two time points (6 and 30 hours after inoculation of hydroponic jars containing maize seedlings with strains of T. virens). Using highly upregulated genes (seed genes), subnetwork modules can be generated. These modules must meet certain criteria: maintain large number of edges to the seed gene, possess significant GO term annotation, and restrict the maximum distance from the seed gene to all other member genes to less than five. Initial efforts have developed five subnetwork modules.A second approach will be developed to use the second set of temporal data with wild-type T. virens and maize hybrid Silver Queen. The normalized read counts obtained from our RNA-seq analysis pipeline will be used to generate clusters from each of six time points by DBSCAN algorithm. Using the generated co-expression network, centrality analysis will be performed to identify the most highly connected genes in each cluster. The top three best connected genes from each cluster at each time point will be selected as potential hub genes for further analysis. We will make comparisons between the two approaches in an effort to reduce the number of hub genes to analyze.Sub-objective 2C. Generate gene knockouts of the targeted hub genes in T. virens. We will use the OSCAR system (Paz et. al., 2011 Fungal Genetics Biology 48:677-684) to construct knockout mutants to assess the role of the selected hub genes or DEGs for their role in fungal-maize interactions, especially colonization, plant growth and ISR. In addition, mutants can be screened for mycoparasitism and secondary metabolite production. We have found the OSCAR system to be very efficient for developing vectors for Agrobacterium transformation of conidia leading to a high number of positive mutants (in some cases >75 % of putative mutants screened). This will allow us to screen more genes than our previous protoplast transformation system.Sub-objective 2D. Determine changes in the maize metabolome during colonization by T. virens. As there are signals from T. virens that elicit an alteration or reprograming of the maize transcriptome, metabolites in the maize sap should reflect these changes. We are proposing to examine the composition of maize seedling sap over a time course of colonization by T. virens to better understand the putative long distance signals that influence plant growth and ISR. Xylem sap will be collected from stems excised from seedlings inoculated with selected strains of T. virens. Sap that is excreted at the wound will be prepared for untargeted and targeted metabolite analysis using LCMS/MS analysis.

Progress 10/01/19 to 09/30/20

Outputs
Target Audience:Research community interested in the role of Trichodema in the rhizosphere of host plants. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?A graduate student (PhD) and postdoctoral researcher were involved in aspects of this research. The graduate student completed his degree during the time covered by this report, and is employed in the biotechnology industry developing a vaccine for COVID-19. Both were active members of our weekly laboratory meetings (my laboratory and a colleague's group) where they presented their current research findings or a recent published article relevant to our research. Both have collaborated with other laboratories on campus as well as the ARS group in College Station. The collaborations involved identification of secondary metabolites or proteins through HPLC, MS/MS, and NMR. Both have received training through Texas A&M System for working in a BSL-2 laboratory and required training in discrimination and research and institutional situations. How have the results been disseminated to communities of interest?We have published two papers. Both the graduate student and postdoctoral researcher presented departmental seminars on their research efforts. In addition, we have held zoom meetings with our colleague in Israel and his research group. What do you plan to do during the next reporting period to accomplish the goals?We anticipate submitting two additional manuscripts this next reporting period. One will address a unique hydrophobin from T. virens that we have characterized for its role in root colonization and induced systemic resistance. The second manuscript will report our findings on identifying HUB genes by computational approaches.

Impacts
What was accomplished under these goals? A previous transcriptomic analysis of the interaction of maize roots and T. virens identified several differentially-expressed genes (DEGs) of T. virens that were significantly upregulated in the presence of the roots. One of the genes is Tv_51662, which codes for a hypothetical protein predicted to be a salicylic acid monooxygenase (SAM), which degrades salicylic acid (SA) to catechol. As expression of Tv_51662 was only detectable in T. virens when grown with maize, we hypothesized that this specific SAM gene may play a major role in regulating interactions between T. virens and plant host roots. More specifically, we predicted that this SAM may play a major role in degrading SA to dampen host defenses to facilitate root colonization. Compared to several other SAM genes (Tv_223757, Tv_29989, Tv_147231, Tv_228034, and Tv_76629), Tv_51662 was the only SAM gene that was significantly overexpressed in the presence of maize. We subsequently generated several Δ51662 knockout mutants of T. virens. Compared to WT, these mutants displayed no changes in radial growth on minimal media (VMS), but demonstrated reduced growth when grown on VMS supplemented with SA (1mM). We also grew WT and Δ51662 mutants under hydroponic conditions with maize seedlings and extracted root hormones and metabolites for LC-MS/MS analysis. The preliminary results of this experiment showed that SA levels were highest in untreated maize roots and lowest with roots treated with WT T. virens. Roots treated with Δ51662 had SA levels that were between untreated and WT suggesting that Tv_51662 does play a significantly important role in controlling host SA levels. We subsequently examined root colonization by Δ51662. Compared to WT, the Δ51662 mutants demonstrated significantly reduced levels of root colonization. Most surprisingly, the Δ51662 mutants were no longer able to trigger ISR in treated maize, as foliar lesions caused by Colletotrichum graminicola were much larger in size compared to those treated with WT T. virens. Currently, there is work to clone Tv_51662 to express in E. coli in order to confirm its function in degrading SA in vitro. Using the same transcriptomic data, we employed a different streamlined computational network analysis pipeline (Kim, et al. 2015, Kim, et al. 2018) to investigate the systemic changes across DEGs in the T. virens-maize root interaction. Rather than focusing on read counts from a list of DEGs, this computational approach provides a mechanism to make meaningful predictions from the generated datasets. Based on this analysis of the transcriptomic data (WT with and without roots), several candidate hub genes were identified that may play significant roles in regulating interactions between T. virens and maize. Of the eight candidate genes, we arbitrarily chose three genes for further characterization to confirm the validity of our approach. The first gene, TRIVIDRAFT_18067, encodes for a hypothetical protein that may be an extracellular matrix protein. The knockout mutants exhibited no significant morphological changes, with radial growth, germination, conidiation, direct antagonism of the fungal pathogens Rhizoctonia solani or Pythium ultimum, and maize root colonization being indistinguishable from that of WT. Surprisingly, the mutants were no longer able to trigger ISR in associated maize plants. The second gene, TRIVIDRAFT_72259, encodes for a hypothetical protein and predicted to be a gluconolactonase. Knockout mutants of TRIVIDRAFT_72259 displayed no different phenotypic changes, antagonism towards R. solani and P. ultimum, or ISR compared to WT. However, they were significantly reduced in their ability to colonize maize roots compared to WT. The third gene, TRIVIDRAFT_75844, encodes for a small GTPase Rho3. Unlike the other two mutants, knockout mutants of TRIVIDRAFT_75844 displayed drastically reduced radial growth, germination, conidiation, direct antagonism of fungal pathogens R. solani and P. ultimum, and root colonization compared to WT. The only characteristic assayed that was no altered was the mutants ability to trigger ISR in host maize plants.

Publications

  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Wang, K.D., Gorman, Z., Huang, P.C., Kenerley, C.M., and Kolomiets, M.V. Trichoderma virens colonization of maize roots triggers rapid accumulation of 12-oxophytodienoate and two alpha-ketols in leaves as priming agents of induced systemic resistance. Plant Signaling and Behavior. 15. DOI: 10.1080/15592324.2020.1792187
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Taylor, J.T., Mukherjee, P.K., Puckhaber, L.S., Dixit, K., Igumenova, T.I., Suh, C., Horwitz, B.A., and Kenerley, C.M. Deletion of the Trichoderma virens NRPS, Tex7, induces accumulation of the anti-cancer compound heptelidic acid. Biochemical and Biophysical Research Communications. 529: 672-677.


Progress 05/29/19 to 09/30/19

Outputs
Target Audience:Research community interested in the role of Trichoderma in the rhizosphere of host plants. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?A graduate student has been participated in this project. The student has been an active member of our weekly laboratory meetings where he has the opportunity to present research updates and contribute to paper discussions. As some of his research has involved NMR and MS/MS, he has had the opportunity to interact and collaborate with these advanced facilities on the Texas A&M campus. An undergraduate has also been involved in aspects of the project by assisting in the experiments and presenting her findings at laboratory meetings. How have the results been disseminated to communities of interest?We have published two papers and anticipate the completion of the graduate student's dissertation this year. He will present a Departmental seminar this year, and has monthly interactions with our collaborator in Israel via Zoom. What do you plan to do during the next reporting period to accomplish the goals?We have several pieces of research to conclude, which are directed towards the analysis of selected HUB genes that were identified by the analysis of transcriptomic data of T. virens colonizing maize roots. The research will involve the creation of several mutants that will be assessed for their colonization potential, ability to induce host resistance, and developmental aspects of fungal mycelia. A second set of experiments will test the role of over-expression a novel hydrophobin in mutants for effects on host growth and resistance induction.

Impacts
What was accomplished under these goals? As most species of Trichoderma are capable of colonizing roots of numerous plants resulting in the activation of host defenses against pathogens, we have sought to define genes involved in the colonization process. One aspect of our focus has been on genes and gene clusters encoding for peptide synthetases in T. virens. Screening a series of core genes in putative peptide synthetases for their expression levels during root colonization of maize resulted in identifying several gene candidates for construction of mutants to further assess their role in root colonization. The knockout mutant in a gene named Tex7 was still capable of root colonization, but inhibited the growth of maize young plants (4 week-old). Testing for general production of secondary metabolites by TLC revealed the appearance of an anomalous compound in large quantities that was not in the wild-type strain. Strains of T. virens has been separated into two different groups based primarily on the production of secondary metabolites. The mutant of Tex7 was constructed in a strain that would have been expected to produce gliotoxin, viridin, viridiol but not heptelidic acid or gliovirin. Following a combination of HPLC, LCMS-MS, and NMR, this compound was surprisingly, identified as heptelidic acid. This discovery alters the strain classification structure and selection process of putative biocontrol strains of T. virens. As hydrophobins produced by fungi have been shown to affect root colonization, we undertook an examination of two class I hydrophobins from T. virens. Deletion mutants of the first hydrophobin (HFB9a) were generated and displayed no differences in growth, mycoparasitism, nor biocontrol capability as compared to wild type. However, the surface hydrophobicity of the mycelium was significantly disrupted in the deletion mutants, and the mutants showed significantly less cellulase and chitinase activity. Plants treated with strains lacking the HFB9a gene were significantly less colonized compared to the WT treated plants. Thus, one hypothesis is that the coordination between the hydrophobin and plant cell wall-degrading enzymes has been altered in the mutant. Additionally, plants treated with the mutant strains were unable to induce systemic resistance (ISR) against the foliar pathogen Colletotrichum graminicola. Recombinant protein was used to complement the enzyme activity in the deletion strains and was found to enhance the activity of commercially sourced chitinase and cellulase. The production of the elicitor SM1 secreted by T. virens in the mutants was compared to the WT by western blot and found to be similar, suggesting that there is another mechanism behind the ISR phenotype. The second class I hydrophobin (HFB3a) produced by T. virens shares high homology with the T. asperellum hydrophobin TasHyd1, which is responsible for attachment of the fungus to plant roots. We generated deletion mutants in this locus and found no changes in growth, mycoparasitism, or biocontrol ability compared to wild type. However, the mutants significantly over-colonized treated plants and demonstrated a lack of ISR against Cochliobolus heterostrophus. We are in the process of developing overexpression strains to determine if colonization and ISR will be enhanced by producing more transcript of this gene. Another study was initiated to access gene expression by wild type T. virens introduced into a hydroponic system containing roots of maize seedlings over a time course of five different time points. RNA was extracted from collected fungal tissues and sequenced by RNA-seq. The reads were analyzed using HISAT2, Stringtie, and EdgeR for read alignment, abundance determination, and differential expression analysis, respectively. This analysis returned a large data set of differentially expressed genes. A computational approach was adopted to enrich the likelihood that a selected target gene would have a high impact on the root colonization process. A gene co-expression network that contains all the genes identified in the RNA-seq analysis and connects the genes together based on the likelihood that the expression between the two would be similar at a given point of time was produced. Then two methods of cluster analysis were used to generate groups of genes based on time and level of expression. The first group was derived from scatter plots with the fold change of two different time points on the x and y axes (eg. 6 hrs vs 12 hrs, 12 hrs vs 15 hrs). These scatter plots were then subjected to clustering using the DBScan algorithm, which creates clusters based on the density of the data points. For the second method, the normalized read counts for all the samples were used to create a principal component analysis. I then plotted the first three components in a 3-dimensional plot and used the same DBscan algorithm to generate a list of clusters. The generated clusters are then related back to the co-expression network with each gene in a cluster analyzed for how central it is in the co-expression network. By selecting genes that are very highly connected as targets for knockout, the probability of detecting a phenotypic change in colonization. Three genes have been selected from a total of 50 identified using this approach for mutant construction. The first mutant has been constructed, and assays are underway to determine the effect on colonization.

Publications

  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Crutcher, F.K., Moran-Diez, M.E., Krieger, I. V., and Kenerley C.M. 2019. Effects on hyphal morphology and development by the putative copper radical oxidase glx in Trichoderma virens suggest a novel role as a cell wall associated enzyme. Fungal Genetics and Biology. 131. Article Number:103245.
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Crutcher, F.K., and Kenerley, C.M. 2019. Analysis of a putative glycosylation site in the Trichoderma virens elicitor SM1 reveals no role in protein dimerization. Biochemical and Biophysical Research Communications. 509:817-821.