Soybean seeds: a practical host for the production of functional subunit vaccines

Abstract

Soybean seeds possess several inherent qualities that make them an ideal host for the production of biopharmaceuticals when compared with other plant-based and non-plant-based recombinant expression systems (e.g., low cost of production, high protein to biomass ratio, long-term stability of seed proteins under ambient conditions, etc.). To demonstrate the practicality and feasibility of this platform for the production of subunit vaccines, we chose to express and characterize a nontoxic form of S. aureus enterotoxin B (mSEB) as a model vaccine candidate. We show that soy-mSEB was produced at a high vaccine to biomass ratio and represented ~76 theoretical doses of human vaccine per single soybean seed. We localized the model vaccine candidate both intracellularly and extracellularly and found no difference in mSEB protein stability or accumulation relative to subcellular environment. We also show that the model vaccine was biochemically and immunologically similar to native and recombinant forms of the protein produced in a bacterial expression system. Immunization of mice with seed extracts containing mSEB mounted a significant immune response within 14 days of the first injection. Taken together, our results highlight the practicality of soybean seeds as a potential platform for the production of functional subunit vaccines.

Figure 1

Gene construct design. (a) The pPTN ST108 binary vector used for Agrobacterium-mediated transformation comprising the following regulatory elements: 7S soybean β-conglycinin promoter (P-7S), tobacco etch virus translational enhancer element (TEV), native SEB bacterial signal peptide (SP-N), mutant SEB gene (mSEB), and 35S cauliflower mosaic virus terminator element (T-35s) followed by the selectable marker cassette (nopaline synthase promoter (P-nos), phosphinothricin acetyltransferase gene (bar), and nopaline synthase terminator element (T-nos)). (b) The pPTN 764 binary vector contained soybean 11S glycinin promoter (P-Gly), soybean glycinin signal peptide (SP-Gly), mSEB, and T-35S, followed by the selectable marker cassette. Arrows show orientation of cassettes relative to the right border (RB) and left border (LB) sequences.

Figure 2

Molecular characterization of soy-mSEB events. (a) and (b) Duplex PCR of 8 T1 progeny from the indicated transformation events. WT: nontransgenic (negative control); +: plasmid DNA (positive control). Arrow shows position of amplified DNA fragments derived from mSEB and vegetative storage protein (VSP). Sizes of molecular weight markers are shown in base pairs. (c) and (d) Western blot of protein derived from the T1 progeny shown in (a) and (b). Arrow indicates soy-mSEB immunoreactive protein. Sizes of molecular weight standards are shown as kDa. (e) and (f) Western blots of T2 progeny from the indicated events. (g) and (h) Western blots of T3 progeny from the indicated events.

Figure 3

Quantification of soy-mSEB. (a) Known amounts of total seed protein (ST108, T3 generation) and various known amounts of purified E. coli-derived mSEB protein (standards) were separated under nonreducing SDS-PAGE and subjected to Western analysis. (b) A standard curve generated from the five known standards following densitometric analysis of the film shown in (a). (c) Chart showing theoretical number of vaccine doses present within a single transgenic soybean seed and in a 1 liter volume of crushed soybean powder. Calculations assume 200 soybeans per plant, 160 mg average seed weight, 40% seed protein content, 1.2% mSEB expression, and a 10 μg human vaccine dose, which is similar to the dose recommended for recombinant hepatitis B surface antigen immunizations [18]. The calculations above do not account for any losses during the purification procedures.

Figure 4

Characterization of soybean-derived mSEB. (a) and (b) Western blot analysis of soy-mSEB, E. coli-derived mSEB, and native SEB under nonreducing and reducing conditions. The ST108 soy-mSEB fragments detected under reducing conditions are labeled I and II, while those derived from 764 soy-mSEB are labeled III and IV. (c) and (d) N-terminal sequencing of soy-mSEB fragments detected under reducing conditions. Amino acids identified from N-terminal protein sequencing are shown in shaded boxes and aligned with the relevant portion of the mSEB protein sequence. The bacterial and soybean signal peptide sequences are underlined with bold typeface. Solid arrows indicate the predicted location for signal peptide cleavage and open arrows indicate observed N-termini. (e) and (f) SignalP 4.1 analysis of the ST108 and 764 soy-mSEB amino acid sequences.

Figure 5

Immunohistochemical detection of soy-mSEB in T2 seeds. (a) ST108 seed section. (b) 764 seed section. (c) Nontransgenic (WT) seed section (control). Red fluorescence represents soy-mSEB protein that is either secreted into apoplastic spaces (ST108) or localized throughout the cell (764). DAPI staining of nuclear material is shown in blue. Samples were viewed at 20x magnification using confocal microscopy, and identical microscope parameters were used for photography of all samples shown.

Figure 6

Western blot analysis of promoter specificity. Nonreducing SDS-PAGE conditions were used to separate 10 μg total protein extracted from leaf, stem, and root tissues of the indicated T2 progeny. Equal amounts of T1 seed protein (parent) and nontransgenic (WT) seed protein were also included as controls. Top panels show X-ray film of the resulting Western blots while bottom panels show the blots used in this experiment following staining with Coomassie blue. Sizes of molecular mass standards are shown as kDa. (a) Data for events derived from ST108; (b) data for events derived from 764.

Figure 7

Immunogenicity of SEB proteins. ELISAs were used to determine relative immunogenicities of purified native SEB (nSEB), E. coli-derived recombinant mutant SEB (rmSEB), and soy-derived mutant SEB (smSEB) proteins. Cholera toxin (CT) was included as a negative control. 100 ng purified protein was coated in each well. All assays were performed in quadruplicate. (a) ELISA results using an in-house rabbit anti-mSEB polyclonal detection antibody. (b) ELISA results using a commercial sheep anti-SEB polyclonal detection antibody (Abcam number ab15925). (c) ELISA results using a commercial mouse anti-SEB monoclonal detection antibody (Abcam number ab6064). Values shown represent average absorbance values (405 nm). Error bars represent standard deviation.

Figure 8

Anti-mSEB titers in mice following immunization. Groups of female BALB/c mice (n = 4) were immunized intraperitoneally on day 0 and boosted on days 14 and 28 days with 1 mg transgenic seed extract plus adjuvant. Bleeds were collected just prior to immunization on days 0, 14, and 28, and again on day 42. ELISAs were performed to determine serum IgG anti-mSEB reactivity. Absorbance values (405 nm) represent serum tested at a 1 : 27000 dilution and are presented as mean anti-mSEB titers and error bars represent standard deviation.