Riboswitch-mediated inducible expression of an astaxanthin biosynthetic operon in plastids

Abstract

The high-value carotenoid astaxanthin (3,3'-dihydroxy-β,β-carotene-4,4'-dione) is one of the most potent antioxidants in nature. In addition to its large-scale use in fish farming, the pigment has applications as a food supplement and an active ingredient in cosmetics and in pharmaceuticals for the treatment of diseases linked to reactive oxygen species. The biochemical pathway for astaxanthin synthesis has been introduced into seed plants, which do not naturally synthesize this pigment, by nuclear and plastid engineering. The highest accumulation rates have been achieved in transplastomic plants, but massive production of astaxanthin has resulted in severe growth retardation. What limits astaxanthin accumulation levels and what causes the mutant phenotype is unknown. Here, we addressed these questions by making astaxanthin synthesis in tobacco (Nicotiana tabacum) plastids inducible by a synthetic riboswitch. We show that, already in the uninduced state, astaxanthin accumulates to similarly high levels as in transplastomic plants expressing the pathway constitutively. Importantly, the inducible plants displayed wild-type-like growth properties and riboswitch induction resulted in a further increase in astaxanthin accumulation. Our data suggest that the mutant phenotype associated with constitutive astaxanthin synthesis is due to massive metabolite turnover, and indicate that astaxanthin accumulation is limited by the sequestration capacity of the plastid.

Figure 1

Generation of homoplasmic transplastomic tobacco plants for riboswitch-inducible expression of a synthetic astaxanthin operon. A, Physical maps of the targeting region in the tobacco plastid genome (ptDNA; upper panel) and the modified region in transplastomic Nt-iAXT lines harboring the synthetic astaxanthin operon and the elements of the RAmpER system (Verhounig et al., 2010; Emadpour et al., 2015). The recognition sites of restriction endonucleases used for RFLP analysis and the resulting fragment sizes are indicated. The binding sites of the hybridization probe for RFLP analysis are represented as black horizontal bars. Cr PpsbA: chloroplast psbA promoter from Chlamydomonas reinhardtii; Nt Prrn: plastid rRNA operon promoter from N. tabacum; T7 P: T7 RNA polymerase promoter from bacteriophage T7; Cr TrbcL: 3′ UTR of rbcL from C. reinhardtii; Nt TpsbA: 3′ UTR of psbA from N. tabacum; Cr TatpA: 3′ UTR of atpA from C. reinhardtii; Nt TrbcL: 3′ UTR of rbcL from N. tabacum; Nt Trps16: 3′ UTR of rps16 from N. tabacum; T7 Lg10: 5′ UTR of gene10 from phage T7; IEE (Zhou et al., 2007; Legen et al., 2018). B, RFLP analysis of transplastomic tobacco plants generated with the iAXT construct for RAmpER-dependent inducible expression of the synthetic astaxanthin operon. Total DNA was digested with the restriction enzyme BglII, and fragments were detected by hybridization with a radiolabeled psaB-specific probe (cf. panel A). Nt-WT: wild-type tobacco; M: molecular weight marker. C, Seed assays to confirm homoplasmy of transplastomic plants. Wild-type (Nt-Wt) seeds and T1 seeds from an Nt-iAXT plant were germinated on synthetic medium in the presence or absence of spectinomycin. Absence of antibiotic-sensitive progeny and absence of green seedlings indicate the homoplasmic state of the transplastomic line with respect to the presence of both the aadA gene and the astaxanthin operon. + Spectinomycin: 500 mg L⁻¹ spectinomycin in the culture medium; − Spectinomycin: control with no antibiotic in the culture medium.

Figure 2

Phenotype of transplastomic Nt-iAXT plants in comparison to wild-type tobacco plants. A–H, Plants photographed at different stages of growth; transplastomic Nt-iAXT plants are shown on the right (orange-brown phenotype), and wild-type tobacco plants are shown on the left (green). A, 10 d, (B) 17 d, (C) 24 d, (D) 31 d, (E) 38 d, (F) 48 d, (G) 52 d, and (H) 70 d after sowing. In panel (H), a transplastomic plant constitutively expressing the astaxanthin pathway (Nt-AXT; Lu et al., 2017) is included for comparison. I–K, Flower phenotype of Nt-iAXT plants. I, Side view of transplastomic (right) and wild-type (left) flowers shortly before opening of the corolla. J, Flowers after opening of the corolla. K, Top view showing the red stigma and style of Nt-iAXT flowers.

Figure 3

Comparison of leaf biomass of wild-type plants (Nt-WT) and two independently generated transplastomic (Nt-iAXT) tobacco lines. Samples were collected from plants at the 12-leaf stage, and the total fresh weight of all leaves of a given plant was determined. Error bars indicate the standard deviation (n = 6).

Figure 4

Northern blot analyses to examine mRNA accumulation in Nt-iAXT transplastomic plants and assess the induction of operon expression in response to theophylline application. mRNA accumulation was analyzed in leaf number 5 of 6-week-old plants after watering with 5 or 10 mM theophylline over a period of 3 d. Plants were watered daily and the numbers on top of the lanes indicate the number of waterings with theophylline solution the plants received. Asterisks indicate the expected transcript sizes for *monocistronic, **dicistronic, and ***tricistronic mRNAs derived from the astaxanthin operon. Additional minor hybridizing transcript species were not characterized. Larger transcript species are likely the result of read-through transcription due to inefficient transcription termination in plastids (Zhou et al., 2007; Lu et al., 2013). The ethidium bromide-stained gel prior to blotting is shown as a control for equal loading below each blot. M, RNA size marker; -theo, control plants without theophylline application; Nt-WT, wild type; Nt-AXT, transplastomic plant constitutively expressing the astaxanthin pathway (Lu et al., 2017).

Figure 5

Time course analysis of pigment accumulation upon induction with 5 mM or 10 mM theophylline for 3 d. A, Astaxanthin content. B, Carotenoid contents. C, Chlorophyll contents. Plants were induced at the six-leaf stage by watering with theophylline solution (once per day for three consecutive days). DW, leaf dry weight; Day, number of times the plants were watered with theophylline solution (once per day); Conc., concentration of the theophylline (Theo) solution used for induction; WT, wild type. Error bars represent the sd (n = 6). Significant changes in astaxanthin accumulation on induction compared to the uninduced state are marked by asterisks (Student’s t test; *P < 0.05; **P < 0.01). Note that the increase in astaxanthin contents in line Nt-iAXT-2 is slightly above the significance criteria (P = 0.078 at 10 mM theophylline after 2 d).

Figure 6

Microscopic images of leaf mesophyll cells from wild-type (Nt-WT) and transplastomic tobacco plants expressing the astaxanthin biosynthetic operon either inducibly (Nt-iAXT) or constitutively (Nt-AXT). A–C, Light microscopic images. Dark red particles within the chloroplasts of transplastomic leaves represent sites of aggregation and/or crystallization of astaxanthin (Lu et al., 2017). D–F, TEM images. Note that, while the transplastomic plants engineered to produce astaxanthin constitutively (Lu et al., 2017) show an underdeveloped thylakoid network (T) and accumulation of large plastoglobules (P), the inducible transplastomic plants display normally developed thylakoid stacks and are similar to wild-type chloroplasts. CW, cell wall.

Figure 7

Comparison of astaxanthin and chlorophyll accumulation in transplastomic plants engineered for RAmpER-dependent expression of the synthetic astaxanthin operon (Nt-iAXT) and those engineered to express the operon constitutively (Nt-AXT; Lu et al., 2017). The diagram shows a developmental series of eight consecutive leaves (leaves number 4–11 from plants at the 12-leaf stage; leaves numbered from the bottom). Note that, although the astaxanthin content is similar in the two transplastomic lines, the chlorophyll content in Nt-iAXT is nearly twice as high as in Nt-AXT. Error bars represent the sd (n = 3). DW, dry weight.

Figure 8

Quantification of selected metabolites by LC/GC-MS analysis. A, Long-chain oxidation products of astaxanthin. B, α-citral, a volatile isoprenoid, specifically accumulating in astaxanthin-synthesizing plants. C, Short-chain oxidation products of astaxanthin. D, Glutathione and ascorbate, two polar metabolites that act as antioxidants. Error bars represent the sd (Nt-iAXT: n = 6, Nt-AXT: n = 3, Nt-WT: n = 3). See also Supplemental Dataset S1.