Exploring glucosinolates diversity in Brassicaceae: a genomic and chemical assessment for deciphering abiotic stress tolerance

Highlights

• Brassicaceae genes involved in GLS biosynthesis were identified using a multi-database approach.

• UPGMA and PCA separation between genes in GLS core structure and CYP450/MYB gene families.

• Phylogenetics revealed a recent diversification of aliphatic genes and an earliest for indolic.

• Distinct GLS chemo-profiles between Brassica crops and Diplotaxis species, wild relatives.

• GLS-rich species as a new source of taxa with great agronomic potential for abiotic stress tolerance.

Abstract

Brassica is one of the most economically important genus of the Brassicaceae family, encompassing several key crops like Brassica napus (cabbage) and broccoli (Brassica oleraceae var. italica). This family is well known for their high content of characteristic secondary metabolites such as glucosinolates (GLS) compounds, recognize for their beneficial health properties and role in plants defense. In this work, we have looked through gene clusters involved in the biosynthesis of GLS, by combining genomic analysis with biochemical pathways and chemical diversity assessment. A total of 101 Brassicaceae genes involved in GLS biosynthesis were identified, using a multi-database approach. Through a UPGMA and PCA analysis on the 101 GLS genes recorded, revealed a separation between the genes mainly involved in GLS core structure synthesis and genes belonging to the CYP450s and MYBs gene families. After, a detailed phylogenetic analysis was conducted to better understand the disjunction of the aliphatic and indolic genes, by focusing on CYP79F1–F2 and CYP81F1–F4, respectively. Our results point to a recent diversification of the aliphatic CYP79F1 and F2 genes in Brassica crops, while for indolic genes an earliest diversification is observed for CYP81F1–F4 genes. Chemical diversity revealed that Brassica crops have distinct GLS chemo-profiles from other Brassicaceae genera; being highlighted the high contents of GLS found among the Diplotaxis species. Also, we have explored GLS-rich species as a new source of taxa with great agronomic potential, particularly in abiotic stress tolerance, namely Diplotaxis, the closest wild relatives of Brassica crops.

Introduction

The Brassicaceae is one of the world's most economically important plant family (Ishida et al., 2014). It includes important crop species such as Brassica oleracea (e.g., cauliflower, Brussels sprouts, cabbage, broccoli, and Kai Lan), Brassica rapa (e.g., pakchoi, choy sum, and Chinese cabbage), Nasturtium officinale (e.g., watercress), and Raphanus sativus (e.g., daikon radish and red cherry radish). Other species such as Diplotaxis tenuifolia and Eruca vesicaria, commonly referred as ‘rocket salads', have also attracted a considerable interest as culinary vegetables because of their strong flavor and content of putative health-promoting compounds (Verkerk et al., 2010). These species and their crop wild relatives (CWR – taxa closely related to crops) grown primarily in the Euro-Mediterranean region, which contains the highest proportion of agronomically important plants representing an important reservoir of genetic resources for crop improvement (Kell et al., 2008). CWR are likely to contain a great genetic diversity necessary to combat climate change because of the diversity of habitats in which they grow and the wide range of conditions they are adapted to (Ford-Lloyd et al., 2011).

Among the most important chemical compounds produced by Brassicaceae species are the Glucosinolates (GLS), which proved to have health promoting effects and importance in abiotic stress tolerance (Cartea and Velasco, 2007). They are constituted by a common structure comprising a β-D-thioglucose group, a sulfonated oxime moiety and a variable side-chain derived either from methionine, tryptophan, phenylalanine, or from other branched chain amino acids. GLS are found in 16 dicotyledonous plant families where, at least, 130 different structures have been identified so far (Fahey et al., 2001; Collett et al., 2014).

GLS are present at different concentrations throughout the plant organs. They can reach 1% of the dry weight in some tissues of Brassica (Fahey et al., 2001). Within a single species, up to 4 different GLSs dominate the GLS occurrence in the plant (Verkerk et al., 2008). The type, concentration and distribution of the GLS in the plants of Brassicaceae family vary according to a high number of factors, namely species (Bellostas et al., 2004), variety (Choi et al., 2014), plant organ (Brown et al., 2003; Bellostas et al., 2004) or plant age (Fahey et al., 1997; Brown et al., 2003) and developmental cycle. Moreover, environmental conditions such as season (Cartea and Velasco, 2007), biotic (Verkerk et al., 2008) or abiotic stress factors such as salinity or drought, are also known to play a role on the production and content of these compounds (Khan et al., 2011; Martínez-Ballesta et al., 2015).

Recent studies have revealed that GLS and their derivatives have beneficial effects on humans. They can help in suppressing tumor growth of various types of cancers namely: breast, brain, blood, bone, colon, gastric, liver, lung, oral, pancreatic and prostate (Zhang et al., 2003; Soundararajan and Kim, 2018). Significant reduction in plasma LDL-C levels has also been reported as being directly linked to consumption of GLS-rich broccoli (Armah et al., 2015). Some GLS derived products are reported to have antimicrobial effects and well documented health benefits (Cavaiuolo and Ferrante, 2014; Bischoff, 2016). Exclusive or excessive feeding of vegetables and/or seeds from the Brassica plants have been associated with toxic effects in livestock (VanEtten and Tookey, 1983; Tripathi and Mishra, 2007) and strategies have been explored to reduce GLS content in Brassica vegetables to increase their palatability for animal consumption (Verker et al., 2008).

The GLS biosynthetic pathway has been partially elucidated by studies on Arabidopsis (e.g. reviewed in Grubb and Abel, 2006; Halkier and Gershenzon, 2006). The GLS, synthesized from amino acids, are grouped in three subtypes according to their corresponding precursors: i) aliphatic GLS, derived from alanine, leucine, isoleucine, valine, and methionine; ii) indole GLS, derived from tryptophan; and iii) aromatic GLS, derived from phenylalanine and tyrosine (Fahey et al., 2001; Halkier and Gershenzon, 2006). Different authors have reported on aliphatic GLS accounting for 70–97% of the total GLS content in leaves of Brassica oleracea (Cartea and Velasco, 2007), leaves and stems of Brassica napus (Cleemput and Becker, 2011), leaves and seeds of Brassica juncea (Gupta et al., 2012), and sprouts and mature leaves of Brassica rapa (Wiesner et al., 2013). The formation of the GLS core structure involves the action of enzymes from different families, namely the CYP79 (Hansen et al., 2001; Chen et al., 2003), CYP83 (Bak and Feyereisen, 2001), UGT74 (Grubb et al., 2014), C–S-lyases (Mikkelsen et al., 2004) and of sulfotransferases (SOTs or STs) (Piotrowski et al., 2004). These enzymes are involved in the biosynthesis of basic GLS structures from elongated and non-elongated amino acids. The basic GLS structures are subjected to a range of secondary side chain modification and transformation pathways catalyzed by enzymes such as flavin monooxygenase (FMOOXs) (Hansen et al., 2007), GLS-AOPs (Mithen et al., 1995), GLS-OH (Hansen et al., 2008) and CYP81Fs (Pfalz et al., 2009, 2011) to generate different types of GLS structures, that are the last finalizing gene family involved in the indolic biosynthetic pathway (Clarke, 2010; Fahey et al., 2001).

The most important mechanism for the wide production of secondary metabolites as glucosinolates relies on whole-genome events, which occurred in Brassicaceae evolution history (Kliebenstein et al., 2001a,b; Kroymann, 2011). The availability of the whole-genome sequences gives an opportunity for using comparative genomics, which, in turn, can lead to a better understanding of the genome evolution in this family. Whole-genome sequences are available for more than 100 plant species (Tohge et al., 2014). The massive contribution, resulting from next-generation technologies, cannot be currently matched by metabolomics, especially if high-quality and species-optimized approaches are adopted (Fukushima et al., 2014). With the increasing number of whole-genome sequences and the freely available genomic resources, the opportunities for conducting an analysis based on comparative genomics is foreseen.

In this paper, we investigated gene clusters involved on the biosynthesis of GLS, by combining genome analysis with biochemical pathways and compound structure assessment. Considering the high diversity in GLS content in Brassicaceae species, we aim to: i) contribute to the global GLS gene inventory in Brassicaceae; ii) compare gene diversity within the three GLS sub-pathways; iii) assess a potential genetic basis for GLS divergence using 6 CYP genes (CYP79F1–F2 and CYP81F1–F4), known to be key genes of indolic and aliphatic GLS biosynthetic pathways, respectively; and iv) increase the knowledge on the chemical diversity of GLS compounds in major Brassica crops compared to the CWR of the genus Diplotaxis. By combining chemical data with genomic sequences, we expect to provide information of interest for promoting the use of the neglected Diplotaxis genus as a potential viable CWR of economically important Brassica crops.