Immunogenicity of Fusion Protein of Cholera Toxin B Subunit-
Porphyromonas gingivalis 53-kDa Minor Fimbrial Protein Produced in Nicotiana benthamiana
Tae-Geum Kim, Tran Thuy Lan, and Jin-Yong Lee
Received: 11 May 2019 / Revised: 29 July 2019 / Accepted: 31 July 2019
© The Korean Society for Biotechnology and Bioengineering and Springer 2019
Abstract Porphyromonas gingivalis induces destruction of periodontal tissues that surround and support the teeth, contributing to the development of periodontitis, which eventually results in tooth loss in adults. A 53-kDa protein of P. gingivalis is a major subunit variant protein of minor fimbriae (Mfa1), and is reported to be highly immunogenic and considered as a potential vaccine candidate. The gene encoding 53-kDa protein, was divided into three segments, and each DNA segment was fused to the gene coding for cholera toxin B subunit (CTB) to improve gut mucosal immune responses. The ctb-mfa1 fusion genes were expressed in the leaf tissues of Nicotiana benthamiana using agroinfiltration. Immunoblot analysis revealed that CTB- Mfa1 fusion proteins were produced in the agroinfiltrated leaves. The interaction of the plant-produced CTB-Mfa1 fusion proteins with GM1-ganglioside, which acts as the binding site for native CTB, was confirmed by GM1- ELISA. Mice immunized orally with the agroinfiltrated leaf powder containing the CTB-Mfa1 fusion proteins elicited serum IgG and fecal IgA antibodies to CTB and Mfa1.
These results suggest that CTB-Mfa1 fusion proteins produced in plants can be used as an oral vaccine to control P. gingivalis infection-associated periodontitis.
Keywords: Porphyromonas gingivalis, 53-kDa minor fimbrial protein, periodontitis, agroinfiltration, plant-based vaccine
1. Introduction
Periodontitis, a common but severe form of periodontal disease, is an inflammatory condition induced by subgingival plaque bacteria, affecting the tissues surrounding the teeth and eventually leading to tooth loss [1]. Chronic periodontitis, which is more prevalent in adults, is a constant potential source of infection and inflammation, and thus is considered as a risk factor for systemic diseases such as atherosclerotic cardiovascular and cerebrovascular diseases, rheumatoid arthritis, obesity-related diabetes, preterm-low birth weight, respiratory diseases, osteoporosis, and cancers [2]. Therefore, an effective vaccine is urgently needed to prevent perio- dontitis and subsequently systemic diseases.
Porphyromonas gingivalis, a black-pigmented gram- negative anaerobic rod, is a major oral pathogen associated with periodontitis [3]. P. gingivalis possesses several virulence factors such as fimbriae, cysteine proteases, hemagglutinins, and lipopolysaccharide. These factors strongly interact with host cells and support the potency of P. gingivalis as a pathogen [3-6]. Bacterial fimbriae are filamentous proteinaceous appendages projecting from the cell surface.
P. gingivalis fimbriae are an important virulence factor involved in the bacterial colonization of the oral cavity by
Tae-Geum Kim
Center for Jeongup Industry-Academy-Institute Cooperation, Chonbuk National University, Jeongup 56212, Korea
Tran Thuy Lan
Department of Molecular Biology, Chonbuk National University, Jeonju 54896, Korea
Tran Thuy Lan
Institute of Biotechnology, Hue University, Hue, Vietnam Jin-Yong Lee*
Department of Oral Microbiology, School of Dentistry, Kyung Hee University, Seoul 02447, Korea
Tel: +82-2-960-2838; Fax: +82-2-960-2838 E-mail: [email protected]
RESEARCH PAPER
[8,9]. Another study also observed that sera from more than 60% of patients with these periodontal diseases recognized 53-kDa protein, and furthermore, some of the sera reacted primarily with the protein [10]. These results indicate that 53-kDa protein is highly immunogenic and may be an important virulence factor involved in periodontal patho- genesis. The protein was purified and characterized to be a novel short fimbriae [11]. Recently, the finding was supported by Nagano et al. who observed that 53-kDa protein is a major subunit variant protein of Mfa1, of which 67-kDa protein (size varies up to 75 kDa) had been regarded as the only major subunit protein [12]. Overall, these findings suggest that 53-kDa protein can be a potential target candidate for eliciting protective immunity to P. gingivalis- induced periodontitis.
Many vaccines, therapeutic or diagnostic monoclonal antibodies, and pharmaceutical proteins have been produced in plants, some of which are now in clinical development or on the market [13,14]. Plant expression systems have several advantages compared to conventional protein expre- ssion systems, including no risk of pathogen contamination;
easy production of vast amounts of recombinant proteins, such as antigen proteins, at low cost; ability to perform eukaryotic post-translational modifications, such as glyco- sylation and disulfide bonding; and ability to assemble multimeric proteins, such as antibodies [15]. Nevertheless, in many cases, especially in the field of plant-based edible vaccine, the low expression level of vaccine antigen proteins in transgenic plants often becomes the cause of oral tolerance and weak immune responses, and it is a big obstacle for the successful commercial use of transgenic plants as oral vaccines. Some recombinant proteins have increased their stable expression level to overcome this problem in transgenic plants with the help of new techniques such as chloroplast transformation [16] and transient expression system utilizing plant viral vectors delivered by Agrobacterium [17]. Use of receptor-targeting ligands is another alternative to overcome this problem, and they have an ability to deliver the ligand-antigen fusion proteins to the mucosal immune system, increasing uptake of the antigen by immune cells and improving the mucosal immune responses. Cholera toxin B subunit (CTB) and enterotoxigenic Escherichia coli heat-labile enterotoxin B subunit are the best known and widely used as carrier ligand
In the present study, fusion genes consisting of mfa1 gene fragments encoding three different segments of 53-kDa major subunit variant protein of P. gingivalis Mfa1 and the gene encoding CTB were constructed. Proteins were expressed in N. benthamiana leaf tissues using agro- infiltration. The plant-produced CTB-Mfa1 fusion proteins elicited serum IgG and fecal IgA antibodies to recombinant CTB and Mfa1 protein segments when mice were orally administered the fusion proteins.
2. Materials and Methods
2.1. Construction of plant expression vectors
The gene encoding 53-kDa major subunit variant protein of Mfa1 from P. gingivalis A7A1-28 was divided into three fragments to be fused to ctb gene. These three gene fragments of mfa1 -1 (1-150 amino acids of mature 53-kDa Mfa1 protein), mfa1-2 (151-300 amino acids), and mfa1-3 (301-448 amino acids) were amplified by PCR using gene- specific primer sets (Fig. 1A) and pRSET/B plasmid (Invitrogen, San Diego, USA) containing mature 53-kDa protein-coding-mfa1gene [19] as a template. The obtained PCR products were cloned into pGEM®-T Easy vector (Promega, Madison, USA), and the sequences of the cloned PCR products were confirmed by DNA sequencing analysis (Genotech, Seoul, Korea). The cloned vectors were subjected to digestion with restriction enzymes (BamHI and KpnI for mfa1-1 and mfa1-2, BglII and KpnI for mfa1-3) and ligated into BamHI and KpnI sites of the plant expression vector containing ctb gene, pMYV498 [20]. The expression of ctb-mfa1 fusion genes was driven by the duplicated 35S Cauliflower mosaic virus promoter.
An endoplasmic reticulum (ER)-retention signal was added to the C-termini of the ctb-mfa1 fusion genes for their accumulation in the ER to increase the protein expression level. The plant expression vectors containing ctb-mfa1-1, ctb-mfa1-2, and ctb-mfa1-3 were designated pTKP29, pTKP30, and pTKP31 (Fig. 1B).
2.2. Plant materials
Surface-sterilization was performed by exposure of seeds of N. benthamiana for 15 min to 2% sodium hydrochlorite with a few drops of Tween 20, and then the seeds were
washed three times with autoclaved distilled water. The surface-sterilized seeds were placed on the surface of MS basal medium supplemented with 3% sucrose and 0.8% plant agar, and incubated at 25oC in a plant growth chamber.
After germination, the grown N. benthamiana plants were transferred to pots in a greenhouse. The plants were grown under conditions with a constant 16-h photoperiod in the greenhouse.
2.3. Agroinfiltration
Agrobacterium tumefaciens LBA4404 was transformed with the cloned plant expression vectors by triparental mating method [21]. The transformed A. tumefaciens harboring each plant expression vector was grown at 28oC in LB broth containing 10 mM 2-(N-morpholino)ethanesulfonic acid (MES, pH 5.5), 10 mM MgCl2, 20 μM acetosyringone, 50 μg/mL kanamycin, and 100 μg/mL rifampicin. The bacterial cells of A. tumefaciens were harvested by centrifuging at 5,000 × g for 5 min at 4oC, resuspended in MS basal medium containing 10 mM MES, 10 mM MgCl2, and 200 μM acetosyringone, and then adjusted to a final O.D.600 of 1.0. The bacterial cells were allowed to stand at room temperature for 2 h for pre-incubation and transferred to a 1-mL disposable plastic syringe. N.
benthamiana leaves were infiltrated with the A. tumefaciens bacterial suspension by pressing the syringe without a needle into the abaxial side of each leaf. The agroinfiltrated leaves were incubated in a green house.
2.4. Immunoblot analysis
The agroinfiltrated leaves were harvested 7 days after the agroinfiltration and ground with liquid nitrogen in a mortar and pestle. Proteins in the powdered leaves were solubilized and extracted with extraction buffer (200 mM Tris-Cl, pH
8.0, 100 mM NaCl, 400 mM sucrose, 10 mM EDTA, 14 mM 2-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, and 0.05% Tween 20) by centrifuging at 13,000 × g for 10 min at 4oC. The extracted total soluble proteins (TSPs) were either treated in sodium dodecyl sulfate (SDS) sample buffer without β-mercaptoethanol (β-ME) for 5 min at room temperature or boiled for 10 min in SDS sample buffer with β-ME. These unboiled and boiled TSPs of the agroinfiltrated leaves were separated by SDS-polyacrylamide gel electrophoresis, and the separated TSPs were transferred to a HybondTM C nitrocellulose membrane (Amersham Pharmacia Biotech, Piscataway, USA) using a Trans-Blot® SD Semi-Dry Transfer Cell (Bio-Rad, Hercules, USA).
The membrane was incubated with 5% skim milk in TBST (Tris-buffered saline with 0.05% Tween-20) to block non- specific antibody binding. After washing three times with TBST, CTB-Mfa1 fusion proteins in the TSPs were detected by incubating the membrane with rabbit anti-cholera toxin (CT) antibodies (Sigma, St. Louis, USA) or mouse polyclonal antibodies against recombinant Mfa1 proteins that were purified from E. coli transformed with mfa1 gene-containing pRSET expression vector [22]. The membrane was washed three times with TBST and incubated for 2 h with 1:5,000- diluted goat anti-rabbit IgG or goat anti-mouse IgG-alkaline phosphatase (AP) conjugate (Sigma). The membrane was washed twice with TBST, then once with TMN buffer (100 mM Tris-Cl, pH 9.5, 5 mM MgCl2, and 100 mM NaCl), and incubated with BCIP/NBT color development solution (Sigma) to detect the immune complexes on the membrane.
2.5. GM1-ELISA
The quantification and biological activity of CTB-Mfa1 fusion proteins in the agroinfiltrated leaves were measured Fig. 1. Construction of plant expression vectors. The mfa1 gene encoding 53-kDa major fimbrillin variant of minor fimbriae (Mfa1) from Porphyromonas gingivalis A7A1-28 was divided into three segments (mfa1-1, mfa1-2, and mfa1-3), each of which was fused to the 3' end of ctb gene coding for CTB. The primer sequences listed were designed to amplify mfa1-1, mfa1-2, and mfa1-3 genes (A). Each ctb-mfa1 fusion gene was constructed into a plant expression vector under the control of the duplicated 35S Cauliflower mosaic virus gene promoter (pDu35S) and nopaline synthetase terminator (Nos-T). Endoplasmic reticulum (ER)-retention sequence SEKDEL (Ser- Glu-Lys-Asp-Glu-Leu) was located at the C-terminus of the fusion gene. Hinge region sequence GPGP (Gly-Pro-Gly-Pro) was inserted between CTB gene and Mfa1 fragment gene. The T-DNA sequence was flanked by right and left borders (RB and LB). The plant expression vectors containing ctb-mfa1-1, ctb-mfa1-2, and ctb-mfa1-3 were designated pTKP29, pTKP30, and pTKP31, respectively.
Neomycin phosphotransferase II (NPTII) gene was under the control of 35S Cauliflower mosaic virus gene promoter (p35S) and terminator (35S-T) (B).
of 1% BSA in PBS, incubated at 37°C for 2 h, and washed three times with PBST. Two-fold serial dilutions of commercial recombinant CTB (Sigma) starting from 10 ng protein in 100 µL per well were added to the wells to make a standard curve. Along with the wells of serially diluted CTB, the other wells were loaded with the TSPs extracted from the agroinfiltrated or non-agroinfiltrated leaves and incubated at 37°C for 2 h. The wells were washed with PBST, incubated with 1:5,000-diluted rabbit anti-CT antibodies in PBS at 37°C for 2 h, and then washed three times with PBST. The wells were incubated with 1:5,000- diluted goat anti-rabbit IgG conjugated with AP in PBS and washed three times with PBST. AP substrate solution (Sigma) was added to each well (100 µL) and allowed the color to develop at room temperature in the dark. The absorbance measurement of the color was made at 405 nm on a SpectraCountTM (Packard Instrument Co., Downers Grove, IL) ELISA reader. The expression levels of CTB- Mfa1 fusion proteins with ability to bind to the GM1- ganglioside were estimated according to the standard curve of CTB.
2.6. Oral immunization with agroinfiltrated N. benthamiana leaves
In order to obtain large amounts of plant leaf tissues containing CTB-Mfa1 fusion proteins for oral immunization, vacuum infiltration was performed [18]. Briefly, the A.
tumefaciens cells transformed with ctb-mfal-1, -2, and -3 were grown for 2 days in 5 mL of LB containing appropriate antibiotics at 28°C, transferred into 200 mL of LB, and further incubated at 28°C for 1 day. The bacterial cells were harvested by centrifugation and suspended in 10 L of MES infiltration buffer to a final OD600 of 0.2. Six- week-old N. benthamiana plants were placed upside down on a shelf, and their leaves were submerged in the MES infiltration buffer containing the bacterial cells in a vacuum chamber. Vacuum pressure of 0.1 mPa was applied for 2 min, and then the release valve was slowly opened to release the vacuum. The plants vacuum-infiltrated with A.
tumefaciens were further grown for 7 days under greenhouse conditions with a 16-h continuous lighting per day, and the agroinfiltrated leaves were harvested. The leaves were freeze-dried in a lyophilizer (EYELA FDU-2100, Tokyo Rikakikai Co., Japan) and then ground into fine powders
leaves containing CTB-Mfa1-1, -2 or -3 (approximately 100 µg of each fusion protein in 2 mL of PBS per mouse, which was yielded from 216 mg, 144 mg, 216 mg, and 139 mg of lyophilized leaf samples for control, CTB-Mfa1-1, CTB-Mfa1-2, and CTB-Mfa1-3, respectively) once a week for 6 weeks according to the method previously described [24,25]. Blood and fecal samples were obtained from the mice immediately before the first feeding and 3 days after the feeding at week 4 and 6 as previously described [26].
The collected blood samples were stored at room temperature for 1 h and chilled on ice for 1 h. Sera were collected by centrifuging the blood samples at 13,000 × g for 10 min at 4oC and stored at -70oC until use. Each fecal sample (200 mg) was dissolved in 1 mL of PBS with 0.01%
sodium azide. Insoluble materials of the fecal samples were removed by centrifuging at 13,000 × g for 10 min at 4oC.
The supernatants (soluble fecal extracts) were stored at -70oC for further analyses.
2.7. Detection of antibodies in sera and fecal extracts In order to measure the levels of serum IgG and fecal IgA antibodies to CTB and Mfa1 proteins, ELISA was performed as described in our previous study [27] with some modifications. Briefly, commercial recombinant CTB and recombinant Mfa1 proteins (Mfa1-1, Mfa1-2, and Mfa1-3) purified from E. coli were dissolved in bicarbonate buffer. Wells of a microtiter plate were coated with 100 µL (0.8 µg) per well of the proteins and incubated at 4oC overnight. After washing three times with PBST, the wells were blocked with 300 µL per well of 1% BSA in PBS, incubated at 37oC for 2 h, and then washed three times with PBST. The sera and fecal extracts were diluted 1:100 and 1:4 with PBS, respectively, loaded into the protein-coated wells of the plate, and incubated at 37oC for 2 h. The wells were washed three times with PBST, incubated with 100 µL of 1:10,000-diluted anti-mouse IgG (Sigma) or anti-mouse IgA (Sigma) conjugated with AP for 2 h at 37oC, and washed three times with PBST. AP substrate solution (100 µL) was added to each well to develop color for 10 min at room temperature in dark conditions. The levels of the antibodies were determined by measuring the absorbance of the color developed in the wells at 405 nm using the ELISA reader.
3. Results
3.1. Expression of CTB-Mfa1 fusion proteins in the agroinfiltrated N. benthamiana leaves
A. tumefaciens containing each ctb-mfa1 fusion construct was infiltrated into N. benthamiana leaf tissues. In the immunoblot analysis using anti-CT antibodies, the expression of CTB-Mfa1-1 (deduced molecular weight of the pentamer is ~136 kDa) and CTB-Mfa1-3 (~140 kDa) in the plant leaf tissues 7 days after the agroinfiltration was found to be strong, whereas CTB-Mfa1-2 (~141 kDa) was very weak (Fig. 2). To check the gene silencing of the target gene expression, A. tumefaciens containing the gene encoding TBSV p19 [28], a suppressor of gene-silencing, was co- infiltrated with each ctb-mfa1 fusion gene. The expression of all the fusion gene constructs in the agroinfiltrated leaves together with TBSV p19 gene was increased when compared with that in the agroinfiltrated leaves without the suppressor gene; under the conditions of co-agroinfiltration with TBSV p19, all the fusion proteins were expressed at similar levels (Fig. 2A). The immunoblot analysis also revealed that when the agroinfiltrated leaves were unboiled, strongly reacting multiple bands (oligomeric and polymeric forms) of the CTB-Mfa1 fusion proteins with a wide range of high molecular masses (Fig. 2A) were detected by anti- CT. In contrast, the fusion proteins in the boiled samples of the agroinfiltrated leaves appeared to be disassembled into monomeric forms since the same antibody recognized a single distinct band with a molecular mass of ~35 kDa (Fig. 2B).
To obtain antibodies against Mfa1 protein fragments, each fragment gene was introduced into pRSET vector, which was subsequently transformed into E. coli. The resulting three different Mfa1 fragment proteins were purified from E. coli. Deduced molecular weights of Mfa1-1, -2 and -3 were 15.6 kDa, 16.5 kDa, and 16.3 kDa (plus ~3-kDa N- terminal 6xHis tag coding sequence of pRSET vector for each fragment), respectively. Polyclonal antibodies were developed in mice immunized with the purified bacterial recombinant Mfa1-1 and Mfa1-3 proteins (100 µg). In contrast, an attempt to raise antibodies against recombinant Mfa1-2 protein (100 µg) in mice continuously failed since the mice were all dead after immunization with the protein.
Anti-Mfa1-1 antibodies obtained from the mice showed cross-reactivity with the purified recombinant Mfa1-3 (Fig. 3B). Using anti-Mfa1-1 polyclonal antibodies in the immunoblot analysis, the production of CTB-Mfa1-1 in the agroinfiltrated leaves was confirmed (Fig. 3B).
3.2. Quantification and biological activity of CTB- Mfa1fusion proteins produced in the agroinfiltrated leaves
Expression levels of CTB-Mfa1 fusion proteins in TSPs of the agroinfiltrated leaves with the fusion gene and TBSV p19 gene simultaneously were measured by GM1-ELISA, comparing the binding activity of the TSPs with the standard curve that was generated from CTB binding to GM1. The expression levels of CTB-Mfa1-1, CTB-Mfa1-2, and CTB-Mfa1-3 in the agroinfiltrated leaves were determined to be 0.69, 0.46, and 0.72% of TSPs, respectively. In Fig. 2. Immunoblot detection of CTB-Mfa1 fusion proteins expressed in agroinfiltrated leaves using anti-CT antibodies. The protein extracts were prepared from the leaf tissues of Nicotiana benthamiana 7 days after agroinfiltration with Agrobacterium tumefaciens containing ctb-mfa1-1, ctb-mfa1-2, and ctb-mfa1-3 fusion genes. The CTB-Mfa1 fusion proteins were detected using anti-CT antibodies as the primary antibody without boiling (A) and with boiling (B). Lane M is pre-stained protein ladder (Fermentas, Glen Burnie, USA).
Lane PC is bacterial CTB as a positive control. Lane NC is protein extract of non-agroinfiltrated leaf as a negative control. Lanes 1-3 are protein extracts of the agroinfiltrated leaves without TBSV suppressor gene p19. Lanes 4-6 are protein extracts of the agroinfiltrated leaves with TBSV suppressor gene p19. Lanes 1 and 4, 2 and 5, and 3 and 6 are protein extracts of the agroinfiltrated leaves expressing CTB-Mfa1-1, CTB-Mfa1-2, and CTB-Mfa1-3, respectively. P denotes the assembled polymeric CTB-Mfa1 fusion proteins and M denotes the disassembled monomeric fusion proteins.
contrast, TSPs of the non-agroinfiltrated leaves failed to show the GM1-binding activity (Fig. 4).
3.3. Serum IgG and fecal IgA antibodies in orally immunized mice
For immunization, powders of the lyophilized agroinfiltrated leaf tissues containing approximately 100 µg of the fusion proteins were mixed with PBS and fed to mice orally six times at one-week intervals. Serum and fecal samples were collected from the mice, and levels of serum IgG and fecal IgA antibodies were measured by ELISA using CTB, and recombinant Mfa1-1, -2, and -3 proteins as antigens.
Serum IgG antibodies to CTB were elevated in mice orally
immunized with the agroinfiltrated leaf powder containing CTB-Mfa1-1 fusion protein, but mice immunized with the leaf powders possessing CTB-Mfa1-2 and CTB-Mfa1-3 failed to elicit IgG antibodies to CTB (Fig. 5B). Strongly reactive for Mfa1-1 IgG antibodies were elicited in sera from mice immunized with the leaf powder possessing CTB-Mfa1-1 fusion protein. While, sera from mice immu- nized with the leaf powders containing CTB-Mfa1-2 and CTB-Mfa1-3 were found to slightly elevate IgG antibodies to the recombinant Mfa1-2 and Mfa1-3, respectively (Fig. 5C).
Levels of fecal IgA antibodies to CTB were elevated in all mice and were especially high in mice immunized with the leaf powder containing CTB-Mfa1-1 (Fig. 5D). On the Fig. 4. Quantification and biological activity of CTB-Mfa1 fusion proteins in agroinfiltrated leaves. The expression level (A) and biological activity (B) of CTB-Mfa1 fusion proteins produced in the agroinfiltrated leaves were confirmed via GM1-ELISA 7 days after the agroinfiltration. GM1-ELISA was conducted in triplicate using microtiter plates coated with GM1-gangliside, a receptor molecule for biologically active CTB. The error bars represent the standard deviation of the means. *P < 0.005.
Fig. 3. Immmunoblot detection of CTB-Mfa1 fusion proteins expressed in agroinfiltrated leaves using anti-Mfa1 antibodies. The mfa1-1, mfa1-2, and mfa1-3 genes were introduced into pRSET vector, which was subsequently transformed into E. coli. Mfa1-1, Mfa1-2, and Mfa1-3 proteins expressed in Escherichia coli were purified by Ni-NTA column (A). Lane M is pre-stained protein ladder. Lanes 1-3 are Mfa1-1, Mfa1-2, and Mfa1-3 protein purified from E. coli, respectively. The purified recombinant Mfa1 protein was injected into mice, and sera collected from mice immunized with Mfa1-1 detected CTB-Mfa1-1 fusion protein in the agroinfiltrated leaves and cross-reacted with the purified recombinant Mfa1-3 protein (B). Lane 1 is Mfa1-1 protein purified from E. coli. Lane 2 is protein extract from non- agroinfiltrated leaf as a negative control. Lanes 3 and 4 are protein extracts from the agroinfiltrated leaves expressing CTB-Mfa1-1 without boiling and with boiling, respectively. Lane 5 is Mfa1-3 protein purified from E. coli.
other hand, fecal IgA antibody responses to the recombi- nant Mfa1-1 were strong in mice immunized with the leaf powder possessing CTB-Mfa1-1, but the responses to the Mfa1-2 and Mfa1-3 were weak in mice immunized with the leaf powders possessing CTB-Mfa1-2 and CTB-Mfa1-3, respectively (Fig. 5E). In general, the serum and fecal antibodies were more strongly elicited in the mice at week 6 than week 4.
Immunoblot was conducted to check the immunoreactivity of the sera from mice immunized with the agroinfiltrated leaves expressing CTB-Mfa1 fusion proteins. The sera from mice immunized with CTB-Mfa1-1 and Mfa1-3 recognized the monomeric (56 kDa; 53-kDa Mfa1 protein plus ~3-kDa N-terminal 6xHis tag coding sequence of pRSET vector)
and polymeric forms of 53-kDa recombinant protein obtained from E. coli [18] (Fig. 6).
4. Discussion
In this study, agroinfiltration of N. benthamiana leaves with A. tumefaciens based on the binary vector was employed successfully to induce quick and high levels of transient expression of 53-kDa Mfa1 protein of P. gingivalis in the leaves. CTB is often used to increase antigen uptake by immune cells and to improve mucosal immune responses by serving as a mucosal carrier for the fused antigens [29].
However, there is a size limit for antigens when they are Fig. 5. Detection of serum IgG and fecal IgA in immunized mice. Each mouse in the experimental and control groups (5 mice each group) was gavaged six times at an interval of one week with 2 mL of PBS containing powders of the lyophilized agroinfiltrated leaves expressing CTB-Mfa1 fusion protein (approximately 100µg of the fusion protein) or powders of non-agroinfiltrated leaves. Blood and fecal samples were collected immediately before the first feeding and 3 days after the feeding at week 4 and 6 (A). The mouse sera and fecal samples were diluted 100-fold and 4-fold, respectively, and used to measure the levels of serum IgG (B, C) and fecal IgA (D, E) antibodies against CTB (B, D) and the recombinant Mfa1 proteins purified from Escherichia coli (C, E). The error bars represent the standard deviation of the means. *P < 0.001, **P < 0.01, ***P < 0.05, significantly different from the value of non-agroinfiltrated leaf- gavaged mice (WT).
fused to CTB because CTB must be assembled into a pentameric form to function as an antigen carrier. In the present study, therefore, the gene encoding Mfa1 protein of P. gingivalis was divided into three fragments, and each of the mfa1 gene fragments was then fused to the 3’ end of ctb gene. Their expression levels reached 0.69, 0.46, and 0.72% of TSPs for CTB-Mfa1-1, CTB-Mfa1-2, and CTB- Mfa1-3 fusion proteins, respectively, in co-agroinfiltrated leaves with TBSV p19-coding gene, which is a gene- silencing suppressor gene. CTB-Mfa1-2 fusion protein showed a lower level of the expression as compared with the other fusion proteins when agroinfiltrated only with ctb-mfa1-2 gene. A higher expression level of CTB-Mfa1- 2 was obtained by agroinfiltration together with TBSV P19 gene, indicating that the low expression level of CTB- Mfa1-2 was due to silencing of the gene.
In our previous study, a fusion protein of CTB-FimA(C- terminal 1/3) was expressed in potato tubers and observed to be assembled into biologically active oligomeric and polymeric forms that strongly reacted with anti-native FimA antibodies, suggesting a potential use of FimA fused to CTB as a vaccine for periodontitis [30]. Like CTB-FimA, the plant-produced CTB-Mfa1 fusion proteins in the present study were observed to be assembled into oligomeric and polymeric forms as revealed by immunoblot showing multiple bands of high molecular mass proteins. GM1-ELISA demonstrated that the CTB-Mfa1 fusion proteins were able
fusion proteins with a wide range of high molecular masses were detected by immunoblot. This immunoreactive band pattern was similar to that reported in our previous study of transgenic rice calli expressing a fusion protein of M cell- targeting peptide ligand-consensus Dengue virus envelope protein domain III [31]. The appearance of multiple oligomeric and polymeric bands of the CTB-Mfa1 fusion proteins other than a band of pentameric form detected in the immunoblot analysis can be explained, at least in part, by the signal peptide, which could be either processed or unprocessed depending on how far the fusion proteins are being translocated. In fact, our previous study reported that consensus Dengue virus envelope protein in a rice callus showed several bands, but when it was secreted into the rice suspension culture, it was detected as a single band only [31]. Plant-based glycosylation may be another possibility that can explain the presence of proteins with different molecular weights. Plant-produced CTB with a single glycosylation site substitution from Asn4 to Gln4 appeared as a single monomeric band, but the CTB protein without the substitution showed two monomeric bands [32].
Therefore, glycosylation is a contributing factor involved in size variation of proteins. The bands appeared below the band of the monomeric CTB-Mfa1 fusion protein might be considered as truncated forms of the CTB-Mfa1 fusion protein.
Powders of the lyophilized agroinfiltrated leaves containing the fusion proteins were administered orally to mice. The mice immunized with the agroinfiltrated leaf powder possessing CTB-Mfa1-1 fusion protein elicited stronger immune responses of serum IgG and fecal IgA antibodies to recombinant Mfa1-1 protein and CTB when compared with the mice immunized with powders of CTB-Mfa1-2 and CTB-Mfa1-3 fusion proteins. When the purified recombinant Mfa1-1 protein was injected into mice, strong antibody responses against Mfa1-1 were elicited in the mice, but Mfa1-2 and Mfa1-3 proteins failed to induce significant immune responses. These results clearly indicate that Mfa1-1 segment possesses more and/or stronger immunogenic determinants than Mfa1-2 and Mfa1-3. In our previous study, FimA was divided into two fragments (FimA1 and FimA2) and fused to CTB. The CTB-FimA2 fusion protein purified from E. coli elicited a higher level of immune responses in mice than the CTB-FimA1 fusion protein, suggesting that FimA2 segment (C-terminal half) Fig. 6. Immunoreactivity of sera from CTB-Mfa1 fusion protein-
immunized mice with recombinant 53-kDa protein. Recombinant 53-kDa protein purified from Escherichia coli was used to detect the immunoreactivity of serum IgG from mice immunized with the agroinfiltrated leaf powder expressing CTB-Mfa1-1 (A) and CTB-Mfa1-3 (B). Lane M is pre-stained protein ladder. Lane 1 is boiled sample of recombinant 53-kDa protein purified from E. coli. Lane 2 is unboiled sample of 53-kDa recombinant protein.
may contain more immunogenic determinants [22]. FimA and 53-kDa Mfa1 proteins are known to be strong vaccine candidates for P. gingivalis infection-associated periodontitis.
However, since the proteins of long, full-length may not be functionally expressed in plants when fused to CTB, these immunodominant epitopes of FimA and 53-kDa Mfa1 proteins, therefore, should be considered for the development of periodontitis vaccines.
An attempt to raise antibodies against the recombinant Mfa1-2 protein in mice continuously failed since the mice were dead after immunization with the protein. This observation suggests that the Mfa1-2 (151-300 amino acids) region contains a toxic domain, which may induce a pathologic change in host tissue and thus contribute to the virulence and pathogenic characteristics of P. gingivalis.
Mfa1-2 needs to be further investigated for its possible toxicity.
The immunoblot analysis was carried out to check the immunoreactivity of the sera from mice immunized with the agroinfiltrated leaves expressing CTB-fusion proteins.
The sera from mice immunized with the agroinfiltrated leaves containing CTB-Mfa1-1 and CTB-Mfa1-3 proteins recognized the monomeric (proteins denatured under boiling conditions) and polymeric forms (proteins assembled into a native-like minor fimbrial structure under non-boiling conditions) of recombinant 53-kDa protein. The sera from mice immunized with the agroinfiltrated leaves of CTB- Mfa1-1 showed a stronger immunoreactivity with the recombinant protein when compared with the sera from mice immunized with the agroinfiltrated leaves containing CTB-Mfa1-3. These results suggest that the plant-produced CTB-Mfa1-1 fusion protein may be a better vaccine candidate for P. gingivalis-induced periodontitis.
In conclusion, mfa1 gene fragments encoding three segments of Mfa1 from P. gingivalis were fused to ctb gene, and the ctb-mfa1 fusion genes were transformed into A. tumefaciens, which was then used for the transformation of N. benthamiana leaves. The plant-produced CTB-Mfa1 fusion proteins were demonstrated to have the biological activity to recognize GM1-ganglioside, an intestinal epithelial cell receptor for CTB, and elicited serum IgG and fecal IgA antibodies to CTB and Mfa1 antigen proteins when administered to mice orally. These results suggest that the plant-produced CTB-Mfa1 fusion proteins, especially CTB- Mfa1-1(N-terminal 1/3), can be used as vaccine candidates to prevent P. gingivalis infection-associated periodontitis.
Acknowledgements
This research was supported by Basic Science Research Program through the National Research Foundation of
Korea (NRF) grant (NRF-2016R1D1A1B03932450) funded by the Korea government (MSIP; Ministry of Science, ICT
& Future Planning) and Research Base Construction Fund Support Program funded by Chonbuk National University in 2018.
References
1. Loesche, W. J. and N. S. Grossman (2001) Periodontal disease as a specific, albeit chronic, infection: diagnosis and treatment. Clin.
Microbiol. Rev. 14: 727-752.
2. Kim, J. and S. Amar (2006) Periodontal disease and systemic conditions: a bidirectional relationship. Odontology. 94: 10-21.
3. Lamont, R. J. and H. F. Jenkinson (1998) Life below the gum line: pathogenic mechanisms of Porphyromonas gingivalis.
Microbiol. Mol. Biol. Rev. 62: 1244-1263.
4. Lamont, R. J. and H. F. Jenkinson (2000) Subgingival colonization by Porphyromonas gingivalis. Oral Microbiol.
Immunol. 15: 341-349.
5. Holt, S. C., L. Kesavalu, S. Walker, and C. A. Genco (1999) Virulence factors of Porphyromonas gingivalis. Periodontol.
2000. 20: 168-238.
6. O-Brien-Simpson, N. M., P. D. Veith, S. G. Dashper, and E. C.
Reynolds (2003) Porphyromonas gingivalis gingipains: the molecular teeth of a microbial vampire. Curr. Protein. Pept. Sci.
4: 409-426.
7. Enersen, M., K. Nakano, and A. Amano (2013) Porphyromonas gingivalis fimbriae. J. Oral Microbiol. 5: 20265.
8. Kokeguchi, S., K. Kato, H. Kurihara, F. Nishimura, and Y.
Murayama (1990) Purification and characterization of two major outer membrane proteins from Porphyromonas gingivalis. Dent.
Jpn. (Tokyo) 27: 29-34.
9. Kurihara, H., F. Nishimura, T. Nakamura, M. Nakagawa, I.
Tanimoto, Y. Nomura, S. Kokeguchi, K. Kato, and Y. Murayama (1991) Humoral immune response to an antigen from Porphyromonas gingivalis 381 in periodontal disease. Infect.
Immun. 59: 2758-2762.
10. Guo, S., K. Takahashi, S. Kokeguchi, S. Takashiba, D. F. Kinane, and Y. Murayama (2000) Antibody responses against Porphyromonas gingivalis infection in patients with early-onset periodontitis. J. Clin. Periodontol. 27: 769-777.
11. Arai, M., N. Hamada, and T. Umemoto (2000) Purification and characterization of a novel secondary fimbrial protein from Porphyromonas gingivalis strain 381. FEMS Microbiol. Lett.
193: 75-81.
12. Nagano, K., Y. Hasegawa, Y. Yoshida, and F. Yoshimura (2015) A major fimbrilin variant of Mfa1 fimbriae in Porphyromonas gingivalis. J. Dent. Res. 94: 1143-1148.
13. Takaiwa, F., Y. Wakasa, H. Tagagi, and T. Hioi (2015) Rice seed for delivery of vaccines to gut mucosal immune tissues. Plant Biotechnol. J. 13: 1041-1055.
14. Sack, M., T. Rademacher, H. Spiegel, A. Boes, S. Hellwig, J.
Drossard, E. Stoger, and R. Fischer (2015) From gene to harvest:
insights into upstream process development for the GMP production of a monoclonal antibody in transgenic tobacco plants. Plant Biotechnol. J. 13: 1094-1105.
15. Karg, S. R. and P. T. Kallio (2009) The production of biopharmaceuticals in plant systems. Biotechnol. Adv. 27: 879- 894.
16. Chebolu, S. and H. Daniell (2009) Chloroplast-derived vaccine antigens and biopharmaceuticals: expression, folding, assembly and functionality. Curr. Top Microbiol. Immunol. 332: 33-54.
19. Kim, T. G., N. Q. D. Tien, M. S. Yang, J. H. Moon, and J. Y. Lee (2016) Production of monoclonal antibodies against 53-kDa protein of Porphyromonas gingivalis in transgenic rice cell suspension. Plant Cell Tiss. Organ. Cult. 126: 387-397.
20. Kim, T. G., M. Y. Kim, and M. S. Yang (2010) Cholera toxin B subunit-domain III of dengue virus envelope glycoprotein E fusion protein production in transgenic plants. Protein Expr.
Purif. 74: 236-241.
21. Van Haute, E., H. Joos, M. Maes, G. Warren, M. Van Montagu, and J. Schell (1981) Intergeneric transfer and exchange recombination of restriction fragments cloned in pBR322: a novel strategy for the reversed genetics of the Ti plasmids of Agrobacterium tumefaciens. EMBO J. 2: 411-417.
22. Kim, T. G., N. X. Huy, M. Y. Kim, D. K. Jeong, Y. S. Jang, M. S.
Yang, W. H. R. Langridge, and J. Y. Lee (2009) Immunogenicity of a cholera toxin B subunit Porphyromonas gingivalis fimbrial antigen fusion protein expressed in E. coli. Mol. Biotechnol.
41: 157-164.
23. Kim, T. G. and W. H. R. Langridge (2003) Assembly of cholera toxin B subunit full-length rotavirus NSP4 fusion protein oligomers in transgenic potato. Plant Cell Rep. 21: 884-890.
24. Kong, Q., L. Richter, Y. F. Yang, C. J. Arntzen, H. S. Mason, and Y. Thanvala (2001) Oral immunization with hepatitis surface antigen expressed in transgenic plants. Proc. Natl. Acad. Sci.
USA. 98: 11539-11544.
25. Yoshida, T., E. Kimura, S. Koike, J. Nojima, E. Futai, N.
Sasagawa, Y. Watanabe, and S. Ishiura (2011) Transgenic rice
Escherichia coli heat-labile toxin B subunit in transgenic rice callus. Mol. Biotechnol. 44: 14-21.
28. Voinnet, O., Y. M. Pinto, and D. C. Baulcombe (1999) Suppression of gene silencing: a general strategy used by diverse DNA and RNA viruses of plants. Proc. Natl. Acad. Sci. USA. 96:
14147-14152.
29. Sun, J. B., J. Holmgren, and C. Czerkinsky (1994) Cholera toxin B subunit: an efficient transmucosal carrier-delivery system for induction of peripheral immunological tolerance. Proc. Natl.
Acad. Sci. USA. 91: 10795-10799.
30. Shin, E. A., J. Y. Lee, T. G. Kim, Y. K. Park, and W. H. R.
Langridge (2006) Synthesis and assembly of an adjuvanted Porphyromonas gingivalis fimbrial antigen fusion protein in plants. Protein Expr. Purif. 47: 99-109.
31. Kim, T. G., M. Y. Kim, N. X. Huy, S. H. Kim, and Y. S. Yang (2013) M cell-targeting ligand and consensus dengue virus envelope protein domain III fusion protein in transgenic rice calli.
Mol. Biotechnol. 54: 880-887.
32. Yuki, Y., M. Mejima, S. Kurokawa, T. Hiroiwa, Y. Takahashi, D.
Tokuhara, T. Nochi, Y. Katakai, M. Kuroda, N. Takeyama, K.
Kashima, M. Abe, Y. Chen, U. Nakanishi, T. Masumura, Y.
Takeuchi, H. Kozuka-Hata, H. Shibata, M. Oyama, K. Tanaka, and H. Kiyono (2013) Induction of toxin-specific neutralizing immunity by molecularly uniform rice-based oral cholera toxin B subunit vaccine without plant-associated sugar modification.
Plant Biotechnol. J. 11: 799-808.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
