Bio-detoxification of ricin in castor bean (Ricinus communis L.) seeds

Abstract

Ricin is a highly toxic ribosome-inactivating lectin occurring in the seeds of castor bean (Ricinus communis L.). Castor bean grows throughout tropical and sub-tropical regions and is a very important crop due to its high seed content of ricinoleic acid, an unusual fatty acid, which has several industrial applications. However, due to the presence of the toxin, castor bean can cause death after the exposure of animals to low doses of ricin through skin contact, injection, inhalation or oral routes. Aiming to generate a detoxified genotype, we explored the RNAi concept in order to silence the ricin coding genes in the endosperm of castor bean seeds. Results indicated that ricin genes were effectively silenced in genetically modified (GM) plants, and ricin proteins were not detected by ELISA. Hemagglutination activity was not observed with proteins isolated from GM seeds. In addition, we demonstrated that seed proteins from GM plants were not toxic to rat intestine epithelial cells or to Swiss Webster mice. After oil extraction, bio-detoxified castor bean cake, which is very rich in valuable proteins, can be used for animal feeding. Gene silencing would make castor bean cultivation safer for farmers, industrial workers and society.

Figure 1

Engineering bio-detoxified castor bean (R. communis L.) seeds by silencing of the ricin genes. (a) A 460 bp fragment from the ricin A-chain gene was cloned in sense and antisense orientations under the control of the 35 S CaMV promoter for the construction of the intron-hairpin RNAi cassette (Δricin). The pRicRNAi vector also contained the reporter gus gene and the mutated Arabidopsis thaliana ahas gene, which confers tolerance to imidazolinones. (b) PCR analyses confirmed the presence of the gus, ahas and Δricin transgenes. (c) Expression of the gus gene in transgenic embryos (TB14S-5D). Non-transgenic (NT) embryos did not show GUS expression. (d) Southern blot analysis revealed the presence of two copies of the Δricin integrated into the genome of two transgenic plants (TB14S-5D). No signal was observed in non-transgenic plants (NT). Genomic DNA was hybridized with probe a. Molecular size markers are indicated on the left. (e) Northern blot analysis was carried out with the fragment of ricin probe (solid bar) and shows the presence of the ricin siRNA and absence of ricin gene transcript in transgenic seed [TB14S-5D (+)]. In contrast, ricin siRNA transcripts were absent and ricin RNA transcripts were present in non-transgenic (NT) or negative segregating seeds [TB14S-5D (−)]. Full-length Southern and Northern blots and gel images (b,d,e) are presented in Supplementary Figure 2.

Figure 2

Detection of ricin in bio-detoxified event TB14S-5D. ELISA was used to detect and quantify ricin in the endosperm of castor bean seeds. Ricin was detected in non-transgenic seeds (control, wild type plants) and in the negative segregating seeds of the T₁ generation [marked with (−)].However, ricin could not be detected in positive transgenic seeds [marked with (+)]. Asterisks represent significant differences compared to control (P < 0.01, n = 9).

Figure 3

Proteins from transgenic event TB14S-5D do not agglutinate red blood cells. Proteins from transgenic (TB14S-5D) and non-transgenic (NT) seeds were tested for their capacity to hemagglutinate red blood cells (RBC, 2% suspension). Protein concentration was serially diluted by a ratio of 0.5 from wells 1 to 12, starting with 2.5 µg/µL. RCA₁₂₀ (starting with 0.1 µg/µL) was used as a positive control and PBS was a negative control. Agglutinated RBC formed a diffuse mat, whereas non-agglutinated RBC sediment formed a dot at the bottom of the well.

Figure 4

Toxicity performance of transgenic event TB14S-5D. (a) Rat small intestine epithelial cells (IEC-6) were incubated with proteins isolated from transgenic (TB14S-5D) and non-transgenic seeds (NT). In proteins from the non-transgenic seeds, 0 to 50.0 µg total protein/mL contained 0 to 1000 ng ricin/mL (numbers in boxes). There was no statistical difference between the values observed in the TB14S-5D viability curve. Values are expressed as number of viable cells as a percentage of control cells (cultivated only in the DMEM medium). n = 9. Asterisks represent significant statistical differences compared to control (P < 0.01). (b) Inhibition of protein synthesis was quantified in IEC-6 cells incubated with total proteins isolated from non-transgenic (NT) and transgenic TB14S-5D seeds. In proteins from the non-transgenic seeds, 0 to 500 ng total protein/mL, contained 0 to 10 ng ricin/mL (numbers in boxes). Data were expressed as the percentage of incorporated L-[¹⁴C(U)]leucine into proteins of the IEC-6 cells relative to the control (cells incubated with DMEM medium). n = 9. Asterisks represent significant statistical differences compared to control (P < 0.01).

Figure 5

Ricin toxicosis (lethal challenge assay) evaluation in Swiss Webster mice. (a) Swiss albino mice were injected intraperitoneally with 100 µL of a solution of total protein extracted from non-transgenic (NT) seeds (20 µg protein/g body weight; 552 µg ricin/kg body weight) and the equivalent amount of protein (20 µg protein/g body weight) isolated from transgenic TB14S-5D seeds. Comparison of survival curves with log rank test yielded statistical significance of P = 0.0145, n = 7. (b) Effect of intraperitoneal administration of proteins from transgenic (TB14S-5D) and non-transgenic (NT) seeds on blood glucose concentration was evaluated for a period of 48 h. n = 7.