Engineering endosperm storage compartments enables simultaneous enrichment of multiple vitamins in rice

Abstract

Engineering rice endosperm with modules for vitamin binding, lipid storage, and antioxidant protection enriched water‐ and fat‐soluble vitamins without compromising key agronomic traits and maintained vitamin higher levels after cooking, demonstrating its potential to improve crop nutrition.

Vitamin biofortification has primarily focused on individual vitamins in crops by engineering metabolic pathways to either enhance biosynthesis or reduce catabolism, eventually increasing accumulation of the target micronutrient (Van Der Straeten et al., 2020). Although this strategy has delivered significant successes, achievement of simultaneous enhancement of multiple vitamins remains challenging, given the complexity of balancing metabolic flux across multivitamin metabolic pathways, which may lead to unintended trade‐offs (Li et al., 2026). Furthermore, vitamin accumulation depends not only on pathway output but also on intracellular storage capacity, metabolite retention, and resistance to oxidative degradation (Asensi‐Fabado and Munné‐Bosch, 2010; Fitzpatrick et al., 2024). These constraints could be particularly important for chemically distinct vitamins, such as water‐soluble B vitamins and lipid‐soluble vitamin E, which require fundamentally different storage environments and may be unstable during grain maturation, desiccation, and storage. These limitations prompted us to explore whether vitamin accumulation could be improved by targeting storage capacity itself, through enhancement of sequestering structures, sink strength, and antioxidant capacity.

Previous studies provided evidence that metabolite‐binding proteins may affect the retention of individual vitamins. In particular, endosperm‐specific expression of a sesame‐derived thiamine‐binding protein (SiTBP) increased vitamin B1 accumulation in rice grain, suggesting that metabolite storage may itself influence nutritional output (Fitzpatrick et al., 2024). Whether this concept can be extended from single‐vitamin retention to the simultaneous accumulation of multiple vitamins remains unresolved. We therefore designed a multigene strategy in rice endosperm that integrated four heterologous modules (Table S1): GgALB, encoding albumin from Gallus gallus, for water‐soluble vitamin binding (Banipal et al., 2016; Bagoji et al., 2017); AtDGAT1, encoding diacylglycerol acyltransferase 1 from Arabidopsis thaliana, was included to promote triacylglycerol biosynthesis and increase lipid droplets abundance, thereby expanding lipid storage capacity for lipid‐soluble vitamin accumulation (Jako et al., 2001); AtOLEO1, which encodes oleosin 1 from Arabidopsis thaliana, was included to stabilize lipid droplets for enhancing lipid soluble vitamin storage (Siloto et al., 2006); and DrSOD, encoding superoxide dismutase from Deinococcus radiodurans, was included to help scavenge reactive oxygen species and reduce oxidative damage to vitamins (Slade and Radman, 2011). We tested whether coordinated deployment of these storage and protection components could simultaneously elevate multiple vitamins in the edible part of rice grains.

Here, we constructed four gene expression cassettes comprising GgALB, AtDGAT1, AtOLEO1, and DrSOD (Figure 1A; Table S1). Because the edible portion of polished rice is mainly the endosperm, all transgenes were placed under the control of endosperm‐specific promoters and introduced to rice cultivar Zhonghua 11. T0 transgenic plants were identified by PCR (Figure S1; Table S1), and the high expression of GgALB, AtOLEO1 and AtDGAT1, and DrSOD in developing T2 seeds was confirmed in the representative transgenic lines, TG3 and TG6 (Figure 1B–E). Compared with the wild‐type (WT) grains, TG3 and TG6 grains contained enlarged oil‐body‐like structures, and quantitative analysis confirmed a significant increase in oil‐body area (Figure 1F, G). No significant differences were observed between WT and the positive lines in polished‐grain appearance and chalkiness, plant height and tiller number, mature‐grain appearance and 1,000‐grain weight, grain dimensions, embryo morphology and area, or early seedling growth (Figures 1H, I, S2–S4). Co‐expression of the four modules did not cause detectable growth defects or yield penalties, indicating that the synthetic storage system is compatible with normal seed development.

Figure 1.

Enhanced storage capacity and antioxidant protection promote multivitamin accumulation in rice grains

(A) Schematic representation of recombinant vector construction and its application in rice genetic transformation. (B–E) RT‐qPCR analysis of GgALB, AtOLEO1, AtDGAT1, and DrSOD transcript accumulation in developing T2 seeds at 14 d after flowering (14 DAF) in WT, TG3, and TG6. ND, not detected. (F) TEM images of oil‐body‐like structures in developing seeds at 14 d after flowering (14 DAF). Red dashed outlines highlight representative regions containing oil‐body‐like structures and black arrows indicate individual oil‐body‐like structures. Scale bar, 5 μm. (G) Quantification of oil‐body area. (H) Polished grains of the WT, TG3, and TG6. Scale bar, 0.5 cm. (I) Chalkiness degree of WT, TG3, and TG6 polished grains. (J–Q) Quantification of thiamine, riboflavin, niacinamide, niacin, pantothenic acid, pyridoxine, α‐tocopherol, and α‐tocotrienol in polished rice of the WT, TG3, and TG6. (R) Comparative analysis of metabolite profiles in polished rice of WT, TG3, and TG6 lines. Data for each row are standardized by Z‐score. Data are presented as means ± SD in (B–E, G, I, J–Q) (n = 3); gray dots represent individual measurements. Different lowercase letters indicate significant differences among genotypes in (G, I, J–Q) and between TG3 and TG6 in (B–E) (one‐way ANOVA, followed by Tukey's multiple‐comparison test, P < 0.05). WT, wild type; TG3 and TG6, independent transgenic rice lines.

We next quantified vitamin accumulation in mature polished grains (Figure 1J–Q; Table S2). All eight vitamins were significantly elevated in TG3 and TG6, compared to WT. Thiamine and riboflavin increased by 2.22‐ to 2.36‐fold and 2.72‐ to 3.10‐fold, respectively, whereas niacinamide, niacin, pantothenic acid, and pyridoxine increased by 1.67‐ to 3.01‐fold (Figure 1J–O; Table S2). The greatest increase was observed for α‐tocopherol, which increased from 0.53 mg kg⁻¹ in WT to 3.12 and 4.95 mg kg⁻¹ in TG3 and TG6, respectively, corresponding to 5.89‐ and 9.34‐fold increases; α‐tocotrienol increased by 1.88‐ and 2.29‐fold, respectively (Figure 1P, Q; Table S2). This increase is consistent with the proposed role of the AtOLEO1–AtDGAT1 module in providing additional lipid storage capacity for lipid‐soluble vitamin deposition (Figure S5). To evaluate the practical value of the biofortified grains, we further assessed the vitamin profile after processing and storage. Following cooking, eight target vitamins remained significantly higher in TG3 and TG6 than in the WT, ranging from 1.67‐ to 8.05‐fold the corresponding WT levels (Figure S6; Table S3). A similar enrichment pattern was observed after storage at 46.5°C and 90% relative humidity for 6 d, with vitamins remaining significantly elevated in both TG3 and TG6 lines relative to WT (Figure S7). Consistent with the proposed role of the DrSOD module in antioxidant protection, total SOD activity was significantly higher in mature seeds of TG3 and TG6 than in those of WT (Figure S8). These findings indicate that the enhanced vitamin was maintained after cooking and under the tested storage conditions. The simultaneous enhancement of both water‐soluble and lipid‐soluble vitamins supports the coordinated function of the integrated storage system.

As B vitamins serve as cofactors or cofactor precursors in central metabolism, we next examined whether their simultaneous accumulation was accompanied by broad metabolic reprogramming in polished grains. To this end, we profiled 194 metabolites associated with amino acid, lipid, and saccharide metabolism, pathways in which vitamin‐derived cofactors are widely involved (Figure 1R; Table S4). Although vitamin levels were strongly increased, most other detected metabolites showed only limited changes between WT and engineered grains, suggesting that a substantial fraction of the accumulated vitamins may be retained by engineered storage‐related components rather than being released solely as freely available cofactors. This could explain why the engineered grains accumulated more vitamins without obvious changes in overall metabolism, plant growth, or grain quality.

Unlike conventional biofortification strategies centered on precursor supply or pathway enzymes, this approach targets the intracellular environment in which metabolites are stored and stabilized. In this context, the present work extends beyond earlier examples of single‐vitamin retention and establishes a broader synthetic system integrating hydrophilic reservoirs, artificially increased lipid sinks, and oxidative protection. These results support the idea that intracellular sink engineering can be treated as a generalizable design principle for crop biofortification. Because the framework is modular, it may be adaptable to additional labile or poorly retained micronutrients and could also be combined with conventional breeding or metabolic engineering strategies for further improvement. Taken together, these results establish synthetic storage‐compartment engineering as a distinct route for crop biofortification and provide a framework for simultaneous enhancement of multiple vitamins in staple cereals.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

J.Luo, Y.F.L., J.Li., and A.R.F. designed the research. X.W.C., H.T.H., Y.Y.W., and W.Y.T. performed the experiments. M.B., H.F.Z., X.Q.L., Z.Y., and R.X. analyzed the data. X.W.C., H.T.H., and Y.F.L. wrote the manuscript. A.R.F., J.Luo, and J.Li revised and improved the manuscript. All authors have read and approved the contents of this paper.

Supporting information

Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.70363/suppinfo

Figure S1. PCR‐based indentification of rice transgenic lines

Figure S2. Agronomic performance and grain morphology of the wild type, TG3 and TG6 Lines

Figure S3. Grain dimensions of WT, TG3, and TG6

Figure S4. Embryo morphology and early seeding traits of WT, TG3 and TG6 rice

Figure S5. Isolation of oil‐body ‐enriched fractions and analysis of vitamin E components in polished rice grains

Figure S6. Vitamin contents in cooked polished rice from WT, TG3, and TG6

Figure S7. Vitamin contents in polished rice following accelerated storage

Figure S8. Total SOD activity in mature seeds of WT, TG3, and TG6

Table S1. Promoters, gene, and primers information used in this study

Table S2. Quantitative analysis of vitamins in polished rice(mg/kg)

Table S3. Vitamin contents in cooked polished rice (mg/kg)

Table S4. Quantification of metabolites in metabolic pathways where vitamins serve as confactors in polished rice

Cai X, Hou H, Wang Y, et al. (2026) Engineering endosperm storage compartments enables simultaneous enrichment of multiple vitamins in rice. J. Integr. Plant Biol. 68: 3892–3894.