Engineering vitamin E content: from Arabidopsis mutant to soy oil

Abstract

We report the identification and biotechnological utility of a plant gene encoding the tocopherol (vitamin E) biosynthetic enzyme 2-methyl-6-phytylbenzoquinol methyltransferase. This gene was identified by map-based cloning of the Arabidopsis mutation vitamin E pathway gene3-1 (vte3-1), which causes increased accumulation of delta-tocopherol and decreased gamma-tocopherol in the seed. Enzyme assays of recombinant protein supported the hypothesis that At-VTE3 encodes a 2-methyl-6-phytylbenzoquinol methyltransferase. Seed-specific expression of At-VTE3 in transgenic soybean reduced seed delta-tocopherol from 20 to 2%. These results confirm that At-VTE3 protein catalyzes the methylation of 2-methyl-6-phytylbenzoquinol in planta and show the utility of this gene in altering soybean tocopherol composition. When At-VTE3 was coexpressed with At-VTE4 (gamma-tocopherol methyltransferase) in soybean, the seed accumulated to >95% alpha-tocopherol, a dramatic change from the normal 10%, resulting in a greater than eightfold increase of alpha-tocopherol and an up to fivefold increase in seed vitamin E activity. These findings demonstrate the utility of a gene identified in Arabidopsis to alter the tocopherol composition of commercial seed oils, a result with both nutritional and food quality implications.

Figure 1.

Tocopherol and Tocotrienol Structures and Biosynthesis. (A) Detailed chemical structures of tocopherol and tocotrienol. The canonical structures are shown, with R-group modifications illustrated in the gridded box at top right. (B) Final steps of the tocopherol biosynthetic pathway (Bramley et al., 2000). MEP, methylerythritol phosphate; VTE1, tocopherol cyclase; VTE2, homogentisic acid prenyltransferase; VTE3, 2-methyl-6-phytylbenzoquinol (MPBQ) methyltransferase; VTE4, γ-tocopherol methyltransferase.

Figure 2.

Tocopherol Content and Composition of Seeds from Wild-Type and High Delta (hd) Arabidopsis Mutants. (A) HPLC scan of wild-type seed, with tocol standard. (B) HPLC scan of hd2 mutant seed, with tocol standard. (C) Percentage of γ-tocopherol and δ-tocopherol in Arabidopsis wild type and high δ-tocopherol mutants defective in VTE3. Col, Columbia-0; Ler, Landsberg erecta.

Figure 3.

Map-Based Cloning of a Gene Associated with the hd2 Mutant Phenotype. Lowercase letters a to h refer to genetic markers on BACs with the nomenclature as follows: BAC number_nucleotide position on BAC. a, T12J13_32820; b, MCB17_75129; c, K11J14_6841; d, F28P10_21630; e, T5P19_12659; f, F15B8_57589; g, F17J16_24643; h, T12C14_1563. cM, centimorgan. (A) First-pass mapping of hd2 showed linkage to genetic markers near the end of chromosome III. (B) Narrowing with additional markers to five BACs at the end of chromosome III. (C) BACs in the region of the hd2 mutation. (D) Scheme of the predicted exon-intron structure of MAA21_40 (At-VTE3) with sites of vte3-1 through vte3-5. Numbers above the schematic of the gene model refer to nucleotides from the first codon, while numbers below represent amino acids.

Figure 4.

Percentage of γ-Tocopherol and δ-Tocopherol in Arabidopsis T2 Seeds from Wild-Type and hd2 Mutant Plants Expressing At-VTE3 or CTP-Anab-VTE3 Proteins under the Control of Napin. (A) Scheme of the constructs that were used to transform Arabidopsis. (B) Transgenic (n = 16) and control (n = 4) Arabidopsis T2 seed data. Error bars represent standard errors. Ler, Landsberg erecta; wt, wild type.

Figure 5.

2-Methyl-6-Phytylbenzoquinol Methyltransferase Activity of Recombinant Proteins. (A) Scheme of the constructs that were transformed into E. coli. (B) Activity data for E. coli extracts expressing Anab-VTE3, At-VTE3, or At-vte3-1 protein, empty vector negative control, or positive control extract from pea chloroplasts. MPBQ transferase activity is given in units. One unit of MPBQ activity is defined as 1 μmol of 2,3-dimethyl-5-phytylbenzoquinol formation per minute per milligram of protein.

Figure 6.

Tocopherol Composition Analysis of Segregating R1 Single Soybean Seed. (A) Data for wild-type control and eight seeds of transgenic line 27930, containing P_7Sα′-At-VTE3. (B) Data for wild-type control and eight seeds of transgenic line 28906, containing both P_7Sα′-At-VTE3 and P_7Sα′-At-VTE4.

Figure 7.

Protein Sequence Alignment of VTE3 from Arabidopsis and Two Cyanobacteria. The protein sequence of Arabidopsis VTE3 was aligned with the Anabaena VTE3 and Synechocystis VTE3 sequences. Motifs I to III are as described (Kagan and Clarke, 1994). Labels under the sequence correspond to the location of the amino acid change in each of the five hd mutants as described in Figure 3, and the affected amino acid residues are indicated by shading. A putative C-terminal membrane anchor proposed by Motohashi et al. (2003) is underlined.