Development Defects of Hydroxy-Fatty Acid-Accumulating Seeds Are Reduced by Castor Acyltransferases

Abstract

Researchers have long endeavored to produce modified fatty acids in easily managed crop plants where they are not natively found. An important step toward this goal has been the biosynthesis of these valuable products in model oilseeds. The successful production of such fatty acids has revealed barriers to the broad application of this technology, including low seed oil and low proportion of the introduced fatty acid and reduced seed vigor. Here, we analyze the impact of producing hydroxy-fatty acids on seedling development. We show that germinating seeds of a hydroxy-fatty acid-accumulating Arabidopsis (Arabidopsis thaliana) line produce chlorotic cotyledons and suffer reduced photosynthetic capacity. These seedlings retain hydroxy-fatty acids in polar lipids, including chloroplast lipids, and exhibit decreased fatty acid synthesis. Triacylglycerol mobilization in seedling development also is reduced, especially for lipids that include hydroxy-fatty acid moieties. These developmental defects are ameliorated by increased flux of hydroxy-fatty acids into seed triacylglycerol created through the expression of either castor (Ricinus communis) acyltransferase enzyme ACYL-COA:DIACYLGLYCEROL ACYLTRANSFERASE2 or PHOSPHOLIPID:DIACYLGLYCEROL ACYLTRANSFERASE1A. Such expression increases both the level of total stored triacylglycerol and the rate at which it is mobilized, fueling fatty acid synthesis and restoring photosynthetic capacity. Our results suggest that further improvements in seedling development may require the specific mobilization of triacylglycerol-containing hydroxy-fatty acids. Understanding the defects in early development caused by the accumulation of modified fatty acids and providing mechanisms to circumvent these defects are vital steps in the development of tailored oil crops.

Figure 1.

Early development of fae1, CL37, DG2, and PD1a. A, Representative images of seedlings at 10 DAS. B, Proportion of seedlings with an emerged radicle at 1 DAS. C, Proportion of seedlings with fully opened cotyledons at 4 DAS. D, Proportion of seedlings with two true leaves of 1 mm at 10 DAS. Colors denote fae1 (red), CL37 (blue), DG2 (green), and PD1a (gray). Values indicate means ± _SD; n = 5 independent replicates of 30 seeds. Statistical analysis: one-way ANOVA (P < 0.001) with post-hoc Tukey’s test. Columns with different letters are statistically different.

Figure 2.

Proportion of HFA in major chloroplast polar lipids during early development. Total DGDG (A), total MGDG (B), proportion of HFA in DGDG (C), and proportion of HFA in MGDG (D) are shown at 0 to 4 DAS. Colors denote fae1 (red), CL37 (blue), DG2 (green), and PD1a (gray). Values indicate means ± _SD; n = 3 independent replicates of 90 seeds. Statistical analysis: Student’s t test. Values statistically different from CL37 are marked with asterisks (***, P < 0.001).

Figure 3.

Proportion of HFA in PC during early development. Total PC (A) and proportion of HFA in PC (B) are shown at 0 to 4 DAS. Colors denote fae1 (red), CL37 (blue), DG2 (green), and PD1a (gray). Values indicate means ± _SD; n = 3 independent replicates of 90 seeds. Statistical analysis: Student’s t test. Values statistically different from CL37 are marked with asterisks (***, P < 0.001).

Figure 4.

Rate of FA synthesis during early development. The dpm of ³H₂O normalized per seedling is shown at 1 to 4 DAS. Colors denote fae1 (red), CL37 (blue), DG2 (green), and PD1a (gray). The data for DG2 and PD1a are virtually identical, so the DG2 data are obscured by the PD1a data. Values indicate means ± _SD; n = 3 independent replicates of 90 seeds.

Figure 5.

TAG mobilization during early development. Total TAG (A) and total HFA content in TAG (B) are shown. Colors denote fae1 (red), CL37 (blue), DG2 (green), and PD1a (gray). Values indicate means ± sd; n = 3 independent replicates of 90 seeds.

Figure 6.

TAG species during early development. A to C, The percentage of each HFA TAG species remaining is compared with 0 DAS for CL37 (A), DG2 (B), and PD1a (C). Shapes denote 0-HFA (diamonds), 1-HFA (squares), and 2-HFA (triangles). D, Proportion of each species mobilized at 4 DAS. Colors denote CL37 (blue), DG2 (green), and PD1a (gray). Values indicate means ± sd; n = 3 independent replicates of 90 seeds. Statistical analysis: Student’s t test. Values statistically different from CL37 are marked with asterisks (***, P < 0.001).

Figure 7.

Reduced hypocotyl elongation and soluble sugar levels in seedlings. A, Length of hypocotyls at 4 DAS. B, Soluble sugar levels at 2 DAS. Colors denote fae1 (red), CL37 (blue), DG2 (green), and PD1a (gray). Values indicate means ± sd; n = 30. Statistical analysis: one-way ANOVA (P < 0.001) with post-hoc Tukey’s test. Columns with different letters are statistically different.