Oilseed-based metabolic engineering of astaxanthin and related ketocarotenoids using a plant-derived pathway: Lab-to-field-to-application

Abstract

Ketocarotenoids, including astaxanthin, are red lipophilic pigments derived from the oxygenation of β-carotene ionone rings. These carotenoids have exceptional antioxidant capacity and high commercial value as natural pigments, especially for aquaculture feedstocks to confer red flesh colour to salmon and shrimp. Ketocarotenoid biosynthetic pathways occur only in selected bacterial, algal, fungal and plant species, which provide genetic resources for biotechnological ketocarotenoid production. Toward pathway optimization, we developed a transient platform for ketocarotenoid production using Agrobacterium infiltration of Nicotiana benthamiana leaves with plant (Adonis aestivalis) genes, carotenoid β-ring 4-dehydrogenase 2 (CBFD2) and carotenoid 4-hydroxy-β-ring 4-dehydrogenase (HBFD1), or bacterial (Brevundimonas) genes, β-carotene ketolase (crtW) and β-carotene hydroxylase (crtZ). In this test system, heterologous expression of the plant-derived astaxanthin pathway conferred higher astaxanthin production with fewer ketocarotenoid intermediates than the bacterial pathway. We evaluated the plant-derived pathway for ketocarotenoid production using the oilseed camelina (Camelina sativa) as a production platform. Genes for CBFD2 and HBFD1 and maize phytoene synthase were introduced under the control of seed-specific promoters. In contrast to prior research with bacterial pathways, our strategy resulted in nearly complete conversion of β-carotene to ketocarotenoids, including primarily astaxanthin. Tentative identities of other ketocarotenoids were established by chemical evaluation. Seeds from multi-season US and UK field sites maximally accumulated ~135 μg/g seed weight of ketocarotenoids, including astaxanthin (~47 μg/g seed weight). Although plants had no observable growth reduction, seed size and oil content were reduced in astaxanthin-producing lines. Oil extracted from ketocarotenoid-accumulating seeds showed significantly enhanced oxidative stability and was useful for food oleogel applications.

Keywords: antioxidant; aquaculture; camelina; β‐carotene hydroxylase; β‐carotene ketolase.

Figure 1

Astaxanthin biosynthesis pathway identified in Adonis aestivalis (a) and schematic diagram of the T‐DNA region of the binary vectors, ASTA, crtWZ, COMBI and pASTA, used in the study (b). 35S‐P, Cauliflower Mosaic Virus 35S promoter; 35S‐T, Cauliflower Mosaic Virus 35S terminator; CBFD2, carotenoid β‐ring 4‐dehydrogenase (Genbank: AY644758.1); crtW, β‐carotene ketolase (Genbank: QVQ68840.1); crtZ, β‐carotene hydroxylase (Genbank: WP_183216731.1); GGPP, geranylgeranyl pyrophosphate; Gly‐P, glycinin promoter; Gly‐T, glycinin terminator; HBFD1, carotenoid 4‐hydroxy‐β‐ring 4‐dehydrogenase (Genbank: DQ902555.1); PSY, phytoene synthase; TP, transit peptide for small subunit of Rubisco complex; ZmPSY, PSY gene from Zea mays (Genbank: NM_001114652.2). The Basta gene encodes phosphinothricin N‐acetyltransferase, which provides resistance to herbicide applications. Adonis‐specific enzymatic reactions are shown in red.

Figure 2

TLC analysis of Nicotiana benthamiana leaves transiently expressing Adonis genes or bacterial genes involved in astaxanthin biosynthesis. (a) Images of N. benthamiana leaves transiently expressing astaxanthin biosynthetic genes. (b) The carotenoids were extracted using a mixture of methanol and dichloroform (75:25, v/v) then separated on a TLC plate with mobile phase toluene:acetone (80:20, v/v). 1, Ketocarotenoid (unknown). 2, Lutein. 3, Adonixanthin. 4, Astaxanthin. 5, Adonirubin/phoenicoxanthin. 6, Canthaxanthin. 7, Echinenone. 8, β‐carotene.

Figure 3

Characterization of seed development, yield and germination from planting to harvest in WT and pASTA. (a) Images of representative cotyledons excised from the seed coat from 15 to 38 days after flowering (DAF). Scale = 1 mm. (b, c) Dry seed and extracted seed oil by a press. Scale = 2 mm. (d) Histogram of seed area. Seed area was measured using digital images of 400 seeds each using ImageJ (https://imagej.net/ij/). M_pASTA, median value of pASTA. M_WT, median value of WT. Student's t‐test was used to generate the P values. n.s., non‐significant. (e) Seed weight per 100 seeds. Seed weight was measured in 28 replicates. *P < 0.001, student's t‐test. (f) Seed yield per pot from greenhouse. Three plants were grwon in each of 16 pots, and seeds were harvested separately from each pot in greenhouse. *, p<0.001, student's t‐test. (g) Seed germination test. Hundred seeds were tested on the wet paper and the radical appearance was counted. Values are the mean of three replicates ± SD. *P < 0.001, student's t‐test. DAS, days after sowing. (h) Seven‐day‐old and 3‐week‐old camelina plants growing in the greenhouse. (i) Mature camelina plants observed 5 days before seed harvest, grown in the greenhouse. (j) Camelina growing at the field at ENREEC, NE.

Figure 4

TLC analysis (a and b) and HPLC analysis (c and d) of total carotenoids extracted from pASTA seeds grown in a greenhouse. (a) Separation of carotenoids on TLC plate (left) and iodine‐stained TLC image (right). Total extracts from seeds of wild type or pASTA were analysed using TLC. TLC was run using two different mobile phases. The first mobile phase was heptane:ethyl ether:acetic acid (70:30:1, v/v), second mobile phase was toluene:acetone (80:20, v/v). (b) Separation of carotenoids on TLC plate. The neutral lipid (wax esters and triacylglycerol) were separated on the TLC plate with the mobile phase heptane:ethyl ether:acetic acid (70:30:1, v/v). Then the carotenoids were scraped and purified. The extracts with acetone were separated on the TLC plate with the mobile phase toluene:acetone (80:20, v/v). (c) The HPLC chromatogram of ketocarotenoids extracted from pASTA seeds. Peak 1, ketocarotenoid 1; peak 2, astaxanthin; peak 3, ketocarotenoid 2. (d) Absorbance spectrum of ketocarotenoid 1, astaxanthin, and ketocarotenoid 2, which are shown in (c).

Figure 5

The carotenoids content in seeds of wild type and pASTA grown in the field at ENREEC, NE. Values are means (μg/g DW) ± SD of analysis of three independent samples. −S, without saponification. +S, with saponification.

Figure 6

Tocopherol (a), ABA (b) and oil (c) concentrations of seeds from wild type and pASTA grown in the field at Eastern Nebraska Research, Extension and Education Center (ENREEC). Values are means ± SD of analysis of three independent samples. Student's t‐test was compared to wild type. *P < 0.01. n.s., not significant.

Figure 7

Evaluation of the oxidative stability of oils and oleogels from pressed seed oil of wild‐type and pASTA plants grown in the field at ENREEC, NE. (a) Formation of oleogels by mixing sorghum grain waxes with camelina oil pressed from either non‐engineered seeds or pASTA seeds. (b) Peroxide values (PV) for primary oxidation products. (c) p‐anisidine values for secondary products.