Transcriptional response of a target plant to benzoxazinoid and diterpene allelochemicals highlights commonalities in detoxification

Abstract

Background: Plants growing in proximity to other plants are exposed to a variety of metabolites that these neighbors release into the environment. Some species produce allelochemicals to inhibit growth of neighboring plants, which in turn have evolved ways to detoxify these compounds.

Results: In order to understand how the allelochemical-receiving target plants respond to chemically diverse compounds, we performed whole-genome transcriptome analysis of Arabidopsis thaliana exposed to either the benzoxazinoid derivative 2-amino- 3H-phenoxazin-3-one (APO) or momilactone B. These two allelochemicals belong to two very different compound classes, benzoxazinoids and diterpenes, respectively, produced by different Poaceae crop species.

Conclusions: Despite their distinct chemical nature, we observed similar molecular responses of A. thaliana to these allelochemicals. In particular, many of the same or closely related genes belonging to the three-phase detoxification pathway were upregulated in both treatments. Further, we observed an overlap between genes upregulated by allelochemicals and those involved in herbicide detoxification. Our findings highlight the overlap in the transcriptional response of a target plant to natural and synthetic phytotoxic compounds and illustrate how herbicide resistance could arise via pathways involved in plant-plant interaction.

Keywords: 2-amino-3H-phenoxazin-3-one; Allelochemical; Benzoxazinoid; Detoxification; Diterpene; Momilactone B.

Fig. 1

Allelochemicals elicit overlapping transcriptional changes in A. thaliana. A Principal component analysis (PCA) of APO- and momilactone B-treated seedlings and controls. In the APO dataset, PC1 separates timepoints and PC2 separates treatments, while it is the other way around in the momilactone B dataset. B Clustering of differentially expressed genes (DEGs). Heatmap of DEGs compared to the 0 h timepoint. Genes were clustered using WGCNA [29], the resulting clusters are indicated on the right. Clusters containing genes that were more strongly up-regulated in the treated than in the control samples are highlighted in green (A3, A6 and A7 for APO and M1 for Momilactone B)

Fig. 2

Gene ontology (GO)-term enrichment analysis of genes up-regulated by exposure to allelochemicals. GO-term analysis was performed on genes contained in clusters A3, A6 and A7 or M1 (see Fig. 1B). Orange bars indicate the relative fraction of genes associated with the respective GO-term among all genes in the respective cluster; blue bars indicate the relative fraction of these genes among all genes in the genome. Redundant terms were removed. Fill color of the dots indicates -log₁₀(p) of the hypergeometric test, adjusted using the method of Benjamini-Hochberg [30]. Only GO-terms that were significantly (p < 0.05) enriched are shown

Fig. 3

Up-regulation of A. thaliana cytochrome P450 oxidases (CYPs). Phylogenetic tree of all CYPs in the A. thaliana genome, based on protein sequence. Bootstrap values > 70 are shown. CYP74 was used to root the tree. Coloured nodes indicate CYPs up-regulated by APO (blue), momilactone B (red) or both (yellow), colored edges indicate CYPs up-regulated by BOA (blue; Baerson et al., 2005 [32]), fenclorim (red; Brazier-Hicks et al. 2018 [33]), or both (yellow)

Fig. 4

Up-regulation of A.thaliana UPD-dependent glycosyltransferases (UGTs) and glutathione-S-transferases (GSTs). Phylogenetic trees of all UGTs (A) and GSTs (B) in the A. thaliana genome, based on protein sequences. Coloured nodes indicate UGTs (A) and GSTs (B) up-regulated by APO (blue), momilactone B (red) or both (yellow), coloured edges indicate UGTs (A) and GSTs (B) up-regulated by BOA (blue; Baerson et al., 2005 [32]), fenclorim (red; Brazier-Hicks et al., 2018 [33]), or both (yellow)