The pBY3R geminiviral expression vector was provided by Professor Hugh Mason (Arizona State University, Tempe, Arizona, USA). Taq DNA polymerase used for polymerase chain reaction (PCR) was a product of Vivantis Technologies (Selangor, Malaysia). Restriction endonucleases (BsaI, NheI, AflII, XbaI, and SacI) and T4 DNA ligase were purchased from New England Biolabs (Hitchin, UK). Standard native human IgG1 was purchased from Abcam (Cambridge, United Kingdom). Amintra^® Protein A resin for antibody purification was a product of Expedeon (Cambridge, UK). All blue pre-stained protein standard and nitrocellulose membrane (0.45 micron) were bought from BioRad (US). Coomassie brilliant blue was purchased from Applichem (Darmstadt, Germany). Anti-human kappa antibody was purchased from Southern Biotech (Alabama, US) and anti-human gamma antibody was purchased from The Binding Site (Birmingham, UK). Amersham ECL prime western blotting detection reagent was a product of GE Healthcare (Illinois, US). Humanized Rat Basophilic Leukaemia cell line (RS-ATL8) used for IgE cross-linking-induced luciferase expression was kindly provided by Prof. Tanapat Palaga, Department of Microbiology, Faculty of Science, Chulalongkorn University. Other chemical reagents were of analytical grade.

Wild-type N. benthamiana plants were cultivated in a controlled environment in the greenhouse. The plants were grown in pots containing a soil mixture of peat moss, vermiculite, and perlite in the ratio of 4:2:1 (Ji et al., 2018). The greenhouse was maintained at a temperature of 25 ± 2°C with 16 ± 8 hours of day night cycle, light intensity 80–100 μmol/m²s^{− 1} and around 60% humidity. The plants were grown for 4-6 weeks before agroinfiltration.

The gene fragments encoding the variable coding regions of the heavy chain (HC) and light chain (LC) of the Omalizumab (Drugbank accession number: DB00043) and human IgG₁ constant region (Huang et al., 2010) sequences were retrieved for the construction of expression vector. The variable region of heavy chain and light chain sequences were codon optimized in silico for expression in N. benthamiana plants using GeneArt GeneOptimizer software (Invitrogen, Thermo Fisher Scientific, MA, United States). The codon-optimized sequences were synthesized at Biomatik, Canada. The variable regions in the heavy and light chain genes of anti-IgE antibody synthesized were digested using BsaI/NheI and BsaI/AflII restriction enzymes, respectively and the digested DNA products on agarose gel electrophoresis were further purified by using Accuprep PCR/gel purification kit (Bioneer, Korea). For the heavy chain, the variable (V_H) domain was ligated with constant heavy chain domain (C_H1-C_H2-C_H3), in the pBY3R geminiviral expression vector using T4 DNA ligase (New England Biolabs, Hitchin, UK). For the light chain, the variable (V_L) domain was ligated with constant (C_L) regions in pBY3R expression vector. The HC and LC encoding DNA sequences were cloned with the endoplasmic reticulum (ER) retention signal peptide SEKDEL (Ser-Glu-Lys-Asp-Glu-Leu) at the carboxyl terminus (C-terminus) to generate full length HC and LC gene sequences.

In this study, the geminiviral replicon system derived from the bean yellow dwarf virus (Chen et al., 2011) was used for rapid production of anti-human IgE mAb. The positive recombinant plasmid was confirmed by restriction enzyme analysis using XbaI and SacI. Then the A. tumefaciens strain GV3101 was separately transformed with the Omalizumab-heavy chain expression vector pBY3R (pBY3R-Oma-HC) and Omalizumab-light chain (pBY3R-Oma-LC) genes by electroporation. The transformed A. tumefaciens cells were cultured in lysogeny broth media (supplemented with rifampicin, kanamycin and gentamicin at a final concentration of 50 µg/L) and confirmed using PCR with gene-specific forward and reverse primers (PCR cycling condition: an initial denaturation at 96°C for 5 min, followed by 35 cycles of 96°C for 30 s, 52°C for 30 s, and 72°C for 1.5 min, and a final extension at 72°C for 10 min). The positive clones of transformed A. tumefaciens were chosen for further experiments.

Plant-derived anti-human IgE mAb was transiently expressed in N. benthamiana using the geminiviral vector. Recombinant A. tumefaciens harboring the plant expression vector containing either HC or LC were inoculated in lysogeny broth supplemented with rifampicin, gentamicin, and kanamycin at a final concentration of 50 mg/L. The cultures were grown in an incubator shaker at 28°C with 200 rpm overnight. The overnight bacterial culture medium was collected and harvested by centrifuging at 1000 g for 10 min. The cell pellet was gently washed and resuspended in infiltration buffer [10 mM 2-N-morpholino-ethanesulfonic acid (MES), pH 5.5 and 10 mM MgSO₄] to obtain a final optical density (OD₆₀₀) of 0.4 at 600 nm. Agrobacterial suspension containing HC and LC were mixed together in the ratio of 1:1 and co-infiltrated into 30-35 day-old wild-type N. benthamiana leaves by syringe infiltration whereas the large-scale infiltration was performed by vacuum infiltration. After infiltration, the plants were returned to the greenhouse with the controlled condition (temperature at 25 ± 2°C with 16-h light/8-h dark cycle) before being harvested.

For time-course experiment of plant-produced anti-human IgE mAb expression levels in plants, infiltration buffer containing the recombinant pBY3R-Oma-HC and pBY3R-Oma-LC (ratio 1:1) were infiltrated into plant leaves. The infiltrated leaves were collected on different days post-infiltration (dpi) that include 2, 4, 6, 8, and 10 dpi to estimate the optimal incubation time that can express high yields of plant-produced anti-human IgE. The infiltrated leaves were ground in liquid nitrogen using a mortar and pestle. The powdered leaf was added with 1:2 w/v cold extraction buffer (Phosphate buffered saline (1X PBS); 137 mM NaCl, 2.68 mM KCl, 10.1 mM Na₂HPO₄, and 1.76 mM KH₂PO₄, pH 7.4) and vortexed for 5 mins The clarified crude extracts were spun down by centrifugation at 26000 g for 30 min at 4°C. The supernatants were used to measure the total soluble protein (TSP) using Bradford assay (Bio-Rad, USA) following manufacturer’s instructions. The clarified crude extract was further analyzed by SDS-PAGE and Western blot analysis.

The harvested leaves after agroinfiltration were collected and extracted in extraction buffer (1X PBS, pH 7.4) using an electric drill. Supernatants were collected by centrifugation at 26000 g for 30 min and 0.45µm filtration (Millipore Sigma, United States) was performed to clarify plant crude extract containing anti-IgE mAb. For purification, the anti-human IgE mAb in the clarified crude extract was specifically captured by Protein A affinity chromatography. Briefly, 1.0-1.5 mL of Amintra Protein A resin was packed into a 10 mL polypropylene column to a bed height of 1 cm and equilibrated with 5 column volumes (CV) of the extraction buffer (1X PBS, pH 7.4) at a flow rate of 2 mL/min. After column equilibration, the clarified crude extracts were applied at a flow rate of 1 mL/min and the wash step was performed with 10 CV of wash buffer (1X PBS, pH 7.4). Elution was done at a flow rate of 1 mL/min using 5 mL of elution buffer (0.1 M glycine, pH 2.7) with an incubation time of 5 mins prior to the collection of each 1 mL. The elutes were neutralized with 1.5 M Tris-HCl (pH 8.8). The neutralized elutes were desalted and concentrated with Amicon ultra-centrifugal filter (Merck, Massachusetts, United States) in 1X PBS buffer (pH 7.4). The protein’s structure integrity in the purified mAb was confirmed by Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting analysis under reducing and non-reducing conditions.

The protein structure integrity and product-related variants were characterized using SDS-PAGE and Western blot analysis. Commercial anti-human IgE mAb (Omalizumab; Xolair, Novartis, United States) was used as positive control. The samples were analyzed under reducing and non-reducing conditions. For non-reducing conditions, the protein samples were mixed with SDS-loading buffer [125 mM Tris-HCl, pH 6.8, 12% w/v SDS, 10% v/v glycerol, and 0.001% (w/v) bromophenol blue]. The protein sample denatured at 95°C for 5 min was separated on 12% freshly prepared polyacrylamide gels. For reducing conditions, the sample was added to SDS-loading buffer containing 22% β-mercaptoethanol, denatured at 95°C for 5 min, and then separated on 12% polyacrylamide gels. The gels were stained with Coomassie brilliant blue (AppliChem, Germany), and the bands were visualized.

For Western blot analysis, the separated protein samples were transferred on to nitrocellulose membranes (Bio-Rad, USA). The nitrocellulose membranes were blocked with 5% skim milk in 1X PBS (pH 7.4) for 1 h at room temperature of 25 ± 2°C. The membranes were washed thrice with PBS before incubation with either a 1:5,000 dilution (0.2 mg/L) of HRP-conjugated anti-human gamma antibody (The binding site, UK) or a 1:5,000 dilution (0.1 mg/L) of HRP-conjugated anti-human kappa antibody (Southern Biotech, USA) in 3% skim milk prepared in 1X PBS (pH 7.4). After incubation, the membrane was washed thrice with 1X PBST, and the bound antibody was detected using Amersham ECL prime western blotting detection reagent (GE Healthcare, UK) and the signal was captured on medical X-ray green (MXG) film (Carestream, USA).

The binding efficiency of plant-produced anti-human IgE mAb with IgE was determined using enzyme-linked immunosorbent assay (ELISA). 5 µg/mL human IgE was coated onto a 96-well ELISA plate (Griner Bio One GmbH, Austria) and incubated at 37°C for 4 h. Then, the plate was blocked with 200 µL of 5% skim milk (BD, Franklink Lakes, NJ, United States) in 1X PBS (pH 7.4) at 37°C for 2 h and washed three times with 1X PBST. Various concentrations of plant-produced anti-human IgE, commercial Omalizumab (positive control) were diluted in PBS and added to triplicate wells followed by incubation at 37°C for 2 h. After washing three times with 1X PBST (1X PBS containing 0.05% Tween-20), the plate was incubated at 37°C for 1 h with HRP-conjugated anti-human gamma IgG1 antibody (The Binding site, United Kingdom) at a dilution of 1:1,000 in 1X PBS. The plate was developed by adding SureBlue TMB 1-Component Microwell Peroxidase Substrate (Promega, United States), and the reaction was stopped by 1 M H₂SO₄. The absorbance was measured using a microplate reader at an optical density of 450 nm (OD₄₅₀).

The RS-ATL8 cells, humanized rat basophilic leukemic cells, were cultured in minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), 1% Penicillin/Streptomycin, 0.5 µg/mL geneticin, 0.2 µg/mL hygromycin B, and GlutaMAX-I. All the reagents for RS-ATL8 cell culture were procured from Gibco; ThermoFisher Scientific, USA. Measurement of luciferase activity was performed following the procedures described previously (Nakamura et al., 2012; Gericke et al., 2015; Jarupalee et al., 2018) with some modifications. Briefly, the RS-ATL8 cells were seeded at 5×10⁴ cells per well in 96 well-plate (50 μL per well) and incubated for 3 h. Cells were then sensitized overnight with the pre-mixture of 5 μg/mL of human-IgE (Abcam, USA) and various concentrations (0.5-512 μg/mL) of plant-produced anti-human IgE (stock 3.4 mg/mL) or Omalizumab (stock 75 mg/mL) diluted in PBS, for 1 h at 37°C. After sensitization, cells were washed once gently with sterile PBS to remove the unbound antibodies. The pre-mixtures that contain only human IgE and only 512 μg/mL of plant-produced anti-human IgE or Omalizumab were used as positive and negative control, respectively. The sensitized cells were then crosslinked with 5 μg/mL of goat anti-human IgE (Millipore, USA). After 3 h of cross-linking, luciferase activity was measured by the ONE-Glo Luciferase Assay System (Promega Corporation, Madison, WI, USA) using a Varioskan microplate reader (ThermoFisher Scientific, Vantaa, Finland) with luminometric mode. The relative luciferase unit (RLU) was calculated by subtracting the signal from the negative control wells. The study was performed in duplicate wells for three independent runs and the average was used for analysis and the values were expressed as mean ± SD.

For the anti-IgE antibody production in plant host N. benthamiana, the variable regions of heavy and light chain genes of Omalizumab were fused with human IgG₁ along with signal peptide and SEKDEL sequences at N-C terminals respectively and were cloned into pBY3R geminiviral expression vector ( Figure 1 ) and subsequently electroporated into A. tumefaciens strain GV3101. Agrobacterium harboring the Omalizumab genes were co-infiltrated into the N. benthamiana plant leaves to further evaluate the expression of anti-IgE antibody. The infiltrated leaves were harvested at specific timepoints, extracted, and purified for the anti-human IgE antibody by Protein-A affinity chromatography. Further, the purified anti-human IgE antibody was analyzed using SDS-PAGE (under both reducing and non-reducing conditions) stained with Coomassie brilliant blue stain to visualize the antibody fragments and assembling of the purified plant-produced anti-IgE mAb. Western blot analysis was performed to confirm the purified plant-produced mAb by probing with anti-human gamma-HRP and anti-human kappa-HRP. As shown in Figure 2 , the SDS-PAGE profile under reducing condition ( Figure 2A ) showed protein bands at molecular sizes of approximately 50 and 25 kDa respectively corresponding to the heavy and light chains of the antibody, whereas the non-reducing condition showed bands at 150 kDa ( Figure 2D ) in similar to the commercial Omalizumab.

The protein bands were further confirmed by western blot using specific goat anti-human kappa and goat anti-human gamma IgG antibodies antibody. As presented in Figures 2B, C, E , the western blot results showed a detectable signal, confirming the expression of target protein. The results from western blot analysis were correlated with SDS-PAGE, confirming that the protein bands (both HC and LC) observed are at an approximate molecular weight of 50 and 25 kDa respectively ( Figures 2B, C ). Light chains interact with heavy chains and heavy chains can also interact with each other. The recombinant mAbs can exist as combination of heavy and light chains and mostly the F(ab’)₂ fragments exhibit multiple unfolded states in non-reducing SDS (Kiessig et al., 1992). This could be the possible reason for multiple bands in both plant-produced and commercial Xolair ( Figure 2E ).

Preliminary studies with anti-IgE gene constructs having with and without SEKDEL at C-terminus were performed to determine the expression levels of anti-IgE mAb in N. benthamiana. The anti-IgE construct with SEKDEL demonstrated higher anti-IgE accumulation ( Supplementary figure S1 ). The anti-IgE antibody expression with the SEKDEL gene construct was screened on different days after post-infiltration on 2, 4, 6, 8, and 10. The infiltrated leaves were harvested on different days and the expression level was detected by western blot analysis. As shown in Figure 3 , the highest expression of plant-produced anti-human IgE from infiltrated leaves was apparently observed with intense band on 4 dpi with the expression yields up to 41.2 µg/g of leaf fresh mass after subsequent steps of purification. The onset of necrosis was observed in the infiltrated plant leaves on day 4 post infiltration and increased in the following days of 6, 8 and 10. Hence, the 4 dpi was chosen for further analysis. The significant difference in the expression levels on different days post infiltration was not established by statistical methods, owing to the small sample size used in the study.

The binding efficiency of plant produced anti-IgE was determined by ELISA as depicted in Figure 4 . The interaction of the anti-human IgE mAb with its target binding site on IgE was assessed in this assay, which is the prime mode-of-action of Omalizumab in allergic conditions (Guntern and Eggel, 2020). Both plant produced and commercial Omalizumab showed similar binding to human IgE target receptors in all concentrations (0-50 µg of each antibody) analyzed. This implicates the functional ability of the plant expressed anti-IgE mAb. The dissociation constant (K_d) was found to be 14.8 µg/mL and 0.9043 µg/mL for plant produced anti-IgE and commercial Omalizumab respectively, using GraphPad Prism 8.1.2 (GraphPad Software, Inc. La Jolla, CA). Human IgG₁ that was used as the negative control could not bind to the human IgE receptor protein.

Determination of plant produced anti-human IgE to inhibit IgE cross-linking-induced luciferase expression using RS-ATL8 cell line was performed. The results revealed that plant-produced anti-human-IgE could inhibit IgE crosslink in a dose-dependent manner ( Figure 5 ). The reduction of relative luminescence units (RLU) generated by plant produced anti-human-IgE was found to be comparable with commercial Omalizumab in every concentration analyzed. The minute difference in RLU generated by plant produced anti-human-IgE and Xolair was analyzed by Mann-Whitney test (p > 0.05) in every concentration. The 50% inhibitory concentration (IC50) for plant produced anti-human-IgE and commercial Omalizumab were 1.208 and 2.887 µg/mL, respectively. This could be confirmed that the plant-produced anti-human IgE exhibited the bioactivity and function, at least, similar to those of commercial Omalizumab (Xolair).