High production of recombinant protein using geminivirus-based deconstructed vectors in Nicotiana benthamiana

Abstract

We focused on the geminiviral vector systems to develop an efficient vector system for plant biotechnology. Begomoviruses and curtoviruses, which belong to the Geminiviridae family, contain an intergenic region (IR) and four genes involved in replication, including replication-associated protein (Rep, C1), transcriptional activator (TrAP, C2), and replication enhancer (REn, C3). Geminiviruses can amplify thousands of copies of viral DNA using plant DNA polymerase and viral replication-related enzymes and accumulate viral proteins at high concentrations. In this study, we optimized geminiviral DNA replicon vectors based on tomato yellow leaf curl virus (TYLCV), honeysuckle yellow vein virus (HYVV), and mild curly top virus (BMCTV) for the rapid, high-yield plant-based production of recombinant proteins. Confirmation of the optimal combination by co-delivery of each replication-related gene and each IR harboring the Pontellina plumata-derived turbo green fluorescence protein (tGFP) gene via agroinfiltration in Nicotiana benthamiana leaves resulted in efficient replicon amplification and robust protein production within 3 days. Co-expression with the p19 protein of the tomato bush stunt virus, a gene-silencing suppressor, further enhanced tGFP accumulation by stabilizing mRNA. With this system, tGFP protein was produced at 0.7-1.2 mg/g leaf fresh weight, corresponding to 6.9-12.1% in total soluble protein. These results demonstrate the advantages of rapid and high-level production of recombinant proteins using the geminiviral DNA replicon system for transient expression in plants.

Keywords: Nicotiana benthamiana; geminivirus; transient expression; turbo green fluorescence protein; viral vector.

Figure 1

Diagrams of the T-DNA region of the geminivirus-based deconstructed vectors used in this study. Plasmid pSPtGFP contains tGFP under the CaMV 35S double promoter (d35SP) with 5′-leader sequence of tobacco mosaic virus (Ω) and potato protease inhibitor II terminator (PinIIT). The intergenic region (IR) of TYLCV, HYVV, or BMCTV is located in pSPtGFP inside of T-DNA with a hairpin structure cloned into the Level M vector and named pTIRtGFP, pHIRtGFP, and pBIRtGFP. C1 (C1/ΔC2; start codon of C2 was modified), C12 (C1/C2/ΔC3; start codon of C3 was modified) and C123 (C1/C2/C3) of TYLCV, HYVV, or BMCTV ORFs that encode for replication initiation protein were cloned into Level 2 vector under CaMV 35S short promoter (s35SP) and Nos terminator (NosT). The vectors were named pTC1, pTC12, and pTC123 with TYLCV ORFs; pHC1, pHC12, and pHC123 with HYVV ORFs; pBC1, pBC12, and pBC123 with BMCTV ORFs, respectively. LB and RB indicate the T-DNA left and right border, respectively.

Figure 2

Conformation of IR-carrying tGFP replicons in co-infiltrated N. benthamiana leaves with different vector combinations. (A) Outward-facing primer design, which can be only amplified using replicon DNA as a template. (B) Genomic PCR products using outward-facing primers. Lane M: 1 kb DNA ladder, Lane 1: tGFP only, Lanes 2–4: co-infiltrated leaf with TIRtGFP+TC1, TC12 or TC123, Lanes 5–7: co-infiltrated leaf with HIRtGFP+HC1, HC12 or HC123, Lanes 8–10: co-infiltrated leaf with BIRtGFP+BC1, BC12 or BC123. (C to E) The transgene copy number analyzed via qPCR with primers specific for tGFP in Agroinfiltrated N. benthamiana leaves. (C) The tGFP copy number of co-infiltration with TIRtGFP+TC1, TC12, or TC123. (D) The tGFP copy number of co-infiltration with HIRtGFP+HC1, HC12, or HC123. (E) The tGFP copy number of co-infiltration with BIRtGFP+BC1, BC12, or BC123. tGFP in a single agroinfiltrated leaf was used as a negative control. Relative copy number was normalized using tGFP expression at 3 DPI. Black arrows represent primer binding sites for the polymerase chain reaction. Data means ± SE from three independent infiltrated samples. Significant differences were assessed via Dunnett’s one-way ANOVA. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant.

Figure 3

mRNA expression and fluorescence analysis of tGFP in co-infiltrated N. benthamiana leaves with different vector combinations over times. Total RNA and total soluble proteins were extracted from agroinfiltrated leaf tissues. (A−C) mRNA of tGFP analyzed via qRT-PCR with specific primers for tGFP and (D−J) fluorescence assay analyzed via microplate reader. (A, D) Co-infiltration with TIRtGFP and TC1 or TC12 or TC123 and TIRtGFP; (B, E) Co-infiltration with HIRtGFP and HC1 or HC12 or HC123; (C, F) Co-infiltration with BIRtGFP and BC1 or BC12 or BC123 in N. benthamiana. Relative transcripts and fluorescence of tGFP were normalized to tGFP single agroinfiltrated leaf at 7 DPI. tGFP is a single agoinfiltrated leaf as a negative control. (G) Illustration of the approach to directly compare tGFP expression between different vector combinations. Images are representative tGFP fluorescence intensity and photograph of infiltrated leaves in (H) TYLCV, (I) HYVV, and (J) BMCTV at 3, 5, and 7 DPI under UV and visible light. Data means ± SE from three independent infiltrated samples. Significant differences were assessed via Dunnett’s one-way ANOVA. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant.

Figure 4

Effects of p19 on tGFP transcript level and fluorescence in co-infiltrated N. benthamiana leaves. Expression of tGFP mRNA in co-infiltrated N. benthamiana leaves with TIRtGFP+TC123, HIRtGFP+HC123, and BIRtGFP+BC1 with p19 over time. (A) Relative mRNA expression and (B) fluorescence assay were normalized using tGFP single agroinfiltrated leaf at 7 DPI. The images are representative of tGFP fluorescence intensity on 1, 3, 5, and 7 DPI through co-expression with p19 in the (C) tGFP, (D) TIRtGFP+TC123, (E) HIRtGFP+HC123, and (F) BIRtGFP+BC1 under UV light. Data means ± SE from three independent infiltrated samples. Significant differences were assessed via Dunnett’s one-way ANOVA. **p < 0.01; ***p < 0.001; ns, not significant.

Figure 5

Effects on p19 of tGFP protein expression in co-infiltrated N. benthamiana leaves. (A, C) SDS-PAGE gels, (B, D) Western blot analysis, and (E, F) indirect ELISA showing tGFP protein in co-infiltrated N. benthamiana leaves both without p19 and with p19 at 3 DPI. Lane M: prestained protein marker; Lane PC: purified bacterial tGFP protein; Lane NC: infiltration buffer as a negative control; Lane 1: infiltration with tGFP alone; Lane 2: co-infiltration with TIRtGFP+TC123; Lane 3: co-infiltration with HIRtGFP+HC123; Lane 4: co-infiltration with BIRtGFP+BC1. β-actin used as an internal loading control. Data means ± SE from three independent infiltrated samples. Significant differences were assessed via Dunnett’s one-way ANOVA. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant.