Inhibition of human MCF-7 breast cancer cells and HT-29 colon cancer cells by rice-produced recombinant human insulin-like growth binding protein-3 (rhIGFBP-3)

Abstract

Background: Insulin-like growth factor binding protein-3 (IGFBP-3) is a multifunctional molecule which is closely related to cell growth, apoptosis, angiogenesis, metabolism and senescence. It combines with insulin-like growth factor-I (IGF-I) to form a complex (IGF-I/IGFBP-3) that can treat growth hormone insensitivity syndrome (GHIS) and reduce insulin requirement in patients with diabetes. IGFBP-3 alone has been shown to have anti-proliferation effect on numerous cancer cells.

Methodology/principal findings: We reported here an expression method to produce functional recombinant human IGFBP-3 (rhIGFBP-3) in transgenic rice grains. Protein sorting sequences, signal peptide and endoplasmic reticulum retention tetrapeptide (KDEL) were included in constructs for enhancing rhIGFBP-3 expression. Western blot analysis showed that only the constructs with signal peptide were successfully expressed in transgenic rice grains. Both rhIGFBP-3 proteins, with or without KDEL sorting sequence inhibited the growth of MCF-7 human breast cancer cells (65.76 ± 1.72% vs 45.00 ± 0.86%, p < 0.05; 50.84 ± 1.97% vs 45.00 ± 0.86%, p < 0.01 respectively) and HT-29 colon cancer cells (65.14 ± 3.84% vs 18.01 ± 13.81%, p < 0.05 and 54.7 ± 9.44% vs 18.01 ± 13.81%, p < 0.05 respectively) when compared with wild type rice.

Conclusion/significance: These findings demonstrated the feasibility of producing biological active rhIGFBP-3 in rice using a transgenic approach, which will definitely encourage more research on the therapeutic use of hIGFBP-3 in future.

Figure 1. The rhIGFBP-3 expression constructs used for rice transformation.

All constructs carrying the plant-codon optimized hIGFBP-3 were driven by glutelin 1 (Gt1) promoter. The construct, pSB130/Gt1/IGFBP-3 (B), contained modified hIGFBP-3 cDNA alone (A); pSB130/Gt1/SP/IGFBP-3 (SB) contained glutelin signal peptide only (B); pSB130/Gt1/SP/IGFBP-3/KEDL (SBK) carried both glutelin signal peptide and KDEL (C). All constructs were ligated into the twin-DNA binary vector pSB130 for Agrobacterium transformation. The twin-DNA binary vector pSB130, contains two T-DNAs, one flanking the gene of interest driven by Gt1 promoter while the other flanking the selectable marker, hygromycin phosphotransferase (HYG). (Abbreviations: RB – right border; LB – left border).

Figure 2. Transgenic integration of modified hIGFBP-3 in rice.

Genomic DNA extracted from rice leaves of independent transformants was digested with BamHI and separated on 0.8% agarose gel, blotted on nylon membrane and hybridized with DIG-labeled hIGFBP-3 probe. Lanes 1-3: three independent pSB130/Gt1/hIGFBP-3 transformants (B1 to B3); lane 4: wild type (WT) rice plant; lanes 5-6: two independent pSB130/Gt1/SP/hIGFBP-3 transformants (SB1 to SB2); lanes 7-9: three independent pSB130/Gt1/SP/hIGFBP-3::KDEL transformants (SBK1 to SBK3).

Figure 3. Western blot analysis of transgenic rice grains.

Total protein was extracted from mature rice grains of different transformants. Recombinant hIGFBP-3 protein was used as positive control. (A) Lanes 1-5: five independent pSB130/Gt1/hIGFBP-3 transformants (B1-5); lane 6: WT; lane 7: positive control (+ve) - commercial rhIGFBP-3 protein. (B) Lanes 1-5: five independent pSB130/Gt1/SP/hIGFBP-3 transformants (SB1-5); lane 6: WT; lane 7: +ve - commercial rhIGFBP-3 protein. (C) Lanes 1-5: five independent pSB130/Gt1/SP/hIGFBP-3::KDEL transformants (SBK1-5); lane 6: WT; lane 7: +ve - commercial rhIGFBP-3 protein.

Figure 4. Inhibitory effect of commercial rhIGFBP-3 on human MCF-7 breast cancer cells and HT-29 colon cancer cells.

Different concentrations of commercial rhIGFBP-3 ranged from 9.375 ng/ml to 300 ng/ml were used to treat MCF-7 and HT-29 cells. Data are shown as means ± SD.

Figure 5. Inhibitory effect of rice-produced rhIGFBP-3 on MCF-7 human breast cancer cells.

(A) Equal amount (3.125 mg seed) of WT, SB and SBK total seed protein extracts were added to the MCF-7 cells separately. Data are shown as means ± SD. *p < 0.05 and ** p < 0.01 denote statistically significant and very significant differences, respectively, between transgenic lines and WT.

Figure 6. Inhibitory effect of different concentration of rice-produced rhIGFBP-3 on HT-29 human colon cancer cells.

Different concentration of seed proteins (ranged from 1.042 to 15.625 mg) from WT and transgenic SBK-66 and SB-57 lines were used to treat HT-29 cells. Data are shown as means ± SD. *p < 0.05 and ** p < 0.01 denote statistically significant and very significant differences, respectively, between transgenic lines and WT.

Figure 7. Glycoprotein staining of transgenic lines and WT line in rice seeds.

Total protein was extracted from mature dehulled seeds of SB-57 and SBK-66 transgenic lines and WT. RNase B was used as the glycoprotein control. To evaluate endoglycosidase digestion efficiency, RNase B was digested by PNGase F and Endo H. Lane 1: marker; lane 2: WT; lane 3: SBK-66; lane 4: SK-57; lane 5: RNase B control without digestion; lane 6: RNase B digested by PNGase F; and lane 7: RNase B digested by Endo H. Band 1: PNGase F; band 2: RNase B; band 3: RNase B with Endo H digestion, and band 4: RNase B with PNGase F digestion.

Figure 8. Western blot analysis of rhIGFBP-3 digested by Endo H in transgenic rice seeds.

Total protein was extracted from mature dehulled seeds of SB-57 and SBK-66 transgenic lines and digested by Endo H. Lane 1: SB-57; lane 2: SB-57 digested by Endo H; lane 3: marker; lane 4: SBK-66; and lane 5: SBK-66 digested by Endo H.