Date Published: April 2, 2019
Publisher: Public Library of Science
Author(s): Priyakshee Borpatragohain, Terry J. Rose, Lei Liu, Carolyn A. Raymond, Bronwyn J. Barkla, Graham J. King, Yong Pyo Lim.
Brassica crops require high amounts of inorganic sulfur (S) for optimum yield, and are characterized by the synthesis of S-rich glucosinolates (GSL). Although it is well established that seed and GSL yield can be increased by S fertilizer, the detailed relationship between S supply as primary source and the harvestable sinks of seed GSL and storage proteins is poorly understood. We tested the hypothesis that Brassica juncea mustard seed acts as a secondary S sink, and so require a higher rate of S to achieve maximum seed GSL compared to rates required to attain maximum seed biomass. Our experimental strategy involved comparing responses to available S for seed biomass, GSL, and protein. This was carried out in a protected environment using sand culture for a high-GSL condiment-type homozygous B. juncea genotype. A low-GSL canola-type was used as a control, in order to establish a base-line of response. Significantly more S was required to achieve maximum seed GSL than was required to achieve maximum seed mass. Total seed protein content was not significantly affected by increased S. The high-GSL line appeared to have an efficient mechanism of S supply to the secondary S sink, given the observed increase in seed S with increased S availability. From a practical point of view, increases in seed GSL with S availability suggests that S fertilizer rates should be optimized for maximum seed GSL yield, rather that optimizing for seed yield, as occurs for most other crops.
Oleiferous brassicas such as canola (Brassica napus), Indian mustard and condiment mustard (B. juncea) and Chinese cabbage, sarson and Indian rapeseed (B. rapa) require high amounts of inorganic sulfur (S) supply for optimum yield. This can be up to 5–8 times the amount required for wheat [1, 2]. Most inorganic S in the mature seed of brassica is sequestered in the storage proteins cruciferin and napin, and in the secondary metabolite glucosinolate (GSL) [3, 4]. GSLs have a wide range of beneficial effects in crop production and plant defense, with some contributing positively to human nutrition, such as the anti-tumorigenic 4 carbon (C4) side-chain aliphatic GSL glucoraphanin found in broccoli (B. oleracea var. italica) [5, 6]. In contrast, anti-nutritional effects of GSLs on livestock  have led to the secondary domestication and widespread cultivation of canola-type rapeseed containing low seed C3 and C4 side-chain aliphatic GSLs. Canola now plays an important role in cereal rotations of global temperate arable systems.
The majority of inorganic S (as sulfate) taken up by the roots (secondary S source) of Brassicaceae is transported to shoots (primary S sink), where it undergoes enzymatic reduction to organic S forms that include glutathione, cysteine, methionine and PAPS (3′-phosphoadenosine 5′-phosphosulphate) . These assimilated S forms produced in the primary S sink at an early stage of plant development may later act as a S source for primary and secondary GSL sinks, such as siliques and seeds . However, a complete picture of how S source and sink distribution changes over crop developmental stages has not been fully resolved . We previously developed a model of the complex network of transporters, signaling molecules and transcription factors regulating S-metabolism in the context of source-sink relationship within brassicas. This suggested that the mature seed embryo acts as the ultimate sink for S-containing metabolites .
Brassica juncea homozygous condiment-type line with high-GSL content and low GSL canola-type line responded differently to increased S availability. The former required significantly higher S to achieve maximum seed GSL than that was S required for maximum seed mass. The high-GSL line appeared to have an efficient mechanism to supply S to the secondary S sink, given the observed increased in seed S with increased S availability. This contrasts with the apparent defect in either early GSL synthesis or in GSL transport in the low-GSL line. From a practical point of view, the increase in seed GSL with higher rates of S availability suggests that S fertilizer application rates in a given environment should be optimized for maximum seed GSL yield, rather that optimizing S rates for seed yield, as occurs for most other crops. These preliminary findings will be explored further in a population segregating for seed GSL content.