Efficient expression and purification of porcine circovirus type 2 virus-like particles in Escherichia coli

Journal of Biotechnology 220 (2016) 78–85

Contents lists available at ScienceDirect

Journal of Biotechnology journal homepage: www.elsevier.com/locate/jbiotec

Efficient expression and purification of porcine circovirus type 2 virus-like particles in Escherichia coli Pei-Ching Wu a , Tzu-Yu Chen a , Jiun-Ni Chi a , Maw-Sheng Chien b,∗∗ , Chienjin Huang a,∗ a Graduate Institute of Microbiology and Public Health, College of Veterinary Medicine, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 40227, Taiwan, ROC b Graduate Institute of Veterinary Pathobiology, College of Veterinary Medicine, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 40227, Taiwan, ROC

a r t i c l e

i n f o

Article history: Received 20 November 2015 Received in revised form 11 January 2016 Accepted 14 January 2016 Available online 18 January 2016 Keywords: PCV2 capsid protein Virus-like particles Size exclusion chromatography Vaccine

a b s t r a c t Porcine circovirus type 2 (PCV2) capsid (Cap) protein has been successfully used as a vaccine to control porcine circovirus associated disease (PCVAD). Most PCV2 subunit vaccines are recombinant Cap protein expressed in baculovirus/insect cell expression system, but using this eukaryotic system is laborious and expensive. In our previous study, full-length of PCV2Cap protein expressed in Escherichia coli formed virus-like particles (VLPs). This expression system has the advantages of being relatively simple and inexpensive. In this study, we constructed a recombinant plasmid containing the full-length codon-optimized cap (ORF2) gene to improve high-level expression of recombinant Cap protein (rCap) with no changed amino acids. The highly water-soluble rCap protein was purified by a single-column, high-throughput fractionation procedure based on size exclusion chromatography. Yield was 10 mg per 200 ml bacterial culture. The rCap protein self-assembled into VLPs of diameter 25–30 nm that contained exogenous nucleic acids. The immunogenicity of PCV2 VLPs was analyzed by immunizing mice. VLP-immunized mice mounted specific immune responses to PCV2. Thus, expression of rCap in E. coli was feasible for large-scale production of PCV2 VLPs, which could potentially be used for a VLP-based PCV2 vaccine. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Porcine circovirus type 2 (PCV2) is a circular single-stranded DNA virus belonging to the genus Circovirus and the family Circoviridae (Allan and Ellis, 2000). PCV2 is considered to be the etiologic agent responsible for porcine circovirus associated disease (PCVAD) that frequently affects growing pigs at 5–18 weeks of age. Pigs with PCVAD may show clinical symptoms of progressive weight loss, pallor, weakness, jaundice, diarrhea, respiratory distress, lymphadenopathy and enlarged lymph nodes (Meng, 2013). PCV2 can cause severe immunosuppressive effects in swine, which become more susceptible to subsequent or concomitant microbial infections (Vincent et al., 2007). PCV2 can cross the placenta and infect the fetus, leading to birth defects and miscarriage or stillbirth. Consequently, PCVAD has a serious impact on swine industry

∗ Corresponding author. Fax: +886 4 22851741. ∗∗ Corresponding author. E-mail addresses: [email protected] (M.-S. Chien), [email protected], [email protected] (C. Huang). http://dx.doi.org/10.1016/j.jbiotec.2016.01.017 0168-1656/© 2016 Elsevier B.V. All rights reserved.

production worldwide, causing economic losses (Opriessnig et al., 2010; Segalés et al., 2005). The PCV2 virion is small and non-enveloped with icosahedral T = 1 symmetry. It has a diameter of 20.5 nm and is formed from 60 capsid protein subunits (Crowther et al., 2003). The capsid protein, Cap, encoded by open reading frame 2 (ORF2) of PCV2 is the sole structural protein of the viral coat (Nawagitgul et al., 2000). The Cap protein contains type-specific epitopes, is highly immunogenic and reacts strongly with serum from PCV2-infected pigs (Fenaux et al., 2003; Mahé et al., 2000). Recombinant Cap proteins have been expressed in eukaryotic expression systems and they assemble spontaneously to form VLPs that mimic the native virus morphology yet are devoid of viral nucleic acids (Liu et al., 2008). As a vaccine candidate, VLPs are safe and efficient at triggering both humoral and cell-mediated immune responses in host immune cells (Antonis et al., 2006; Ludwig and Wagner, 2007; Noad and Roy, 2003). Therefore, Cap protein is the preferred target for developing new recombinant VLP-based vaccines against PCV2. Many studies have supported vaccination as a management strategy for preventing the impact of PCVAD and improving pig growth at PCVAD-affected farms (Chae, 2012; Fachinger et al., 2008; Horlen et al., 2008; Kixmöller et al., 2008; Martelli et al.,

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85

2011). Vaccination of sows reduces viremia and systemic PCV2 loads and increases PCV2-specific neutralizing antibodies in the colostrum that provide maternal protection to offspring against PCV2 (Kurmann et al., 2011). Presently, there are four commercialized vaccines that are effective against PCV2 infection on the international market: Circovac® (Merial S.A.S., Lyon, France), Suvaxyn PCV2 One Dose (Fort Dodge Animal Health, Fort Dodge, USA), Ingelvac CircoFLEX® (Boehringer Ingelheim Vetmedica Inc., St. Joseph, USA) and Circumvent PCV (Intervet Inc., Millsboro, USA). All these commercially available PCV2 vaccines are based on Cap protein. They are subunit or inactivated vaccines and most were developed using a recombinant baculovirus/insect cell expression system. However, the expression level of recombinant Cap protein is limited and obtaining this protein is laborious and relatively expensive for veterinary applications. In recent years, a prokaryotic expression system in Escherichia coli that is simple, inexpensive and provides high yields has been used to express large amounts of Cap protein. However, 5 clustering of rare codons at the nuclear localization signal (NLS) of the ORF2 gene can cause premature termination of the nascent polypeptide chain and impede the production of Cap protein in E. coli cells. Deletion of the NLS (Trundova and Celer, 2007), fusion to a tag protein (Liu et al., 2001; Yin et al., 2010; Zhou et al., 2005) or expression of certain coding regions of the ORF2 gene (Wu et al., 2008) have been used to circumvent difficulties in producing Cap protein in E. coli. A recent report demonstrated that full-length Cap can be expressed in E. coli by adapting codon usage with an additional polyhistidine tag at both ends of ORF2. However, the produced protein failed to self-assemble into VLPs (Marcekova et al., 2009). In our previous study, a full-length ORF2 gene with four optimized codons near the 5 end was successfully expressed using a modified E. coli expression system. The resulting Cap proteins self-assembled into VLPs (Wu et al., 2012). In this study, a full-length ORF2 gene with codon-optimized for E. coli was synthesized and inserted into the pET24a (+) expression vector to improve expression of recombinant Cap protein (rCap). A large amount of soluble rCap protein was obtained and purified by a single-step purification procedure using size exclusion chromatography (SEC). We studied the structure of assembled rCap proteins using transmission electron microscope (TEM). Immunogenicity of the rCap VLPs was determined by immunizing BALB/c mice and antibody and interferon (IFN)-␥ secreting cell (SC) responses were evaluated for the potential application of the protein as a VLP-based vaccine candidate.

2. Materials and methods

79

Darmstadt, Germany) via the NdeI/XhoI sites. The recombinant expression plasmid pET24a-rORF2 was expected to express Cap protein without the flanking N-terminal T7 tag and C-terminal His tag. Accuracy of PCV2 ORF2 coding sequences was verified by restriction enzyme digestion and DNA sequence analysis.

2.2. Expression of recombinant Cap protein in E. coli Recombinant pET24a-rORF2 plasmid was transformed into E. coli BL21 (DE3) and a single colony was grown in Luria-Bertani (LB) medium with 50 ␮g/ml kanamycin at 37 ◦ C overnight with shaking at 220 rpm. The culture was diluted 1:20 in 200 ml fresh LB medium and incubated at 37 ◦ C at 220 rpm until optical density at 600 nm (OD600 ) was 0.6–0.8. Isopropyl ␤-d-thiogalactopyranoside (IPTG) was added to a final concentration of 1 mM with incubation for 3 h at 37 ◦ C. Cells were harvested by centrifugation at 6000 × g for 10 min at 4 ◦ C. Cell pellet was resuspended in 10 ml of phosphate buffered saline (PBS, 137 mM NaCl, 2.7 mM KCl, 10 mM Na2 HPO4 , 2 mM KH2 PO4 , pH 6.8) and sonicated on ice for 30 cycles of 10 s pulses at 15 s intervals using a VCX-750 Vibra Cell Ultrasonic Processor (Sonics & Materials Inc., Newtown, USA) at 25% amplitude. Lysates were divided into supernatant and pellet by centrifugation at 10,000 × g for 20 min at 4 ◦ C. Pellets were resuspended in PBS at a volume equal to the supernatant. Expression and solubility of rCap protein were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western blots.

2.3. SDS-PAGE and Western blots Specimens to be analyzed were resuspended in an equal volume of 2× sample buffer (125 mM Tris–HCl, pH 6.8, 20% glycerol, 4% SDS, 10% ␤-mercaptoethanol, 0.25% bromophenol blue) and heated to 100 ◦ C for 10 min. Proteins were subsequently separated on 12% polyacrylamide gels. Separating gels were stained with Coomassie brilliant blue R-250 (Thermo Fisher Scientific Inc., Carlsbad, USA) or transferred by electroblotting onto PolyScreen PVDF transfer membranes (NEN Life Science Products Inc., Boston, USA) using Trans-Blot® SD Semi-Dry Electrophoretic Transfer Cell (Bio-Rad, Hercules, USA) according to the manufacturer’s instructions. Membranes were treated with blocking solution (PBS containing 5% skim milk) for 1 h and an appropriate dilution of swine immune serum (P94) specific to PCV2Cap overnight at 4 ◦ C and goat anti-swine IgG (H + L) horseradish peroxidase conjugate (Zymed Laboratories Inc., South San Francisco, USA) for 45 min at 37 ◦ C. Membranes were soaked in 1-StepTM Ultra TMB-Blotting Solution (Thermo Fisher Scientific Inc., Carlsbad, USA) for color development.

2.1. Construction of the recombinant expression plasmid The entire PCV2 ORF2 gene optimized for E. coli codon usage was designed based on the wild-type ORF2 sequence (GenBank accession no. AY885225). The modified nucleotide sequence was predicted by the Codon Adaptation Tool (JCAT, http://www.jcat.de, Technical University of Braunschweig, Germany), a bioinformatics tool to adapt codon usage to most sequenced prokaryotic organisms and selected eukaryotic organisms. The adaptation sequence was synthesized (Integrated DNA Technologies Inc., Coralville, USA) and sequence alignment of the codon-optimized ORF2 (rORF2) and wild-type ORF2 (wt-ORF2) is in Fig. 1. The rORF2 fragment was amplified by polymerase chain reaction (PCR) using primers rORF2/NdeI (5 AGTCATATGACCTACCCGCGTCG-3 , underlined is NdeI site) and rORF2/XhoI (5 -GTTCTCGAGTTACGGTTTCAGCGG-3 , underlined is XhoI site). The amplified rORF2 DNA fragment was subcloned into the pET24a (+) expression vector (Novagen, Merck KGaA,

2.4. Purification of rCap protein VLPs by size exclusion chromatography Soluble rCap protein self-assembled to VLPs and particles were purified by SEC using a HiPrep 16/60 Sephacryl S-500 High Resolution column (GE Healthcare Bio-Sciences AB, Uppsala, Sweden). The gel filtration column was equilibrated with one-half column volume of distilled water and two column volumes of PBS. Protein samples were loaded at 4% column volume and separated at a flow rate of 0.5 ml/min using an ÄKTAFPLCTM system (GE Healthcare Ltd., Chalfont St. Giles, UK). Cap protein was eluted with 140 ml PBS and fractions were collected in 10 ml. Fraction aliquots were analyzed by 12% SDS-PAGE and Western blotting. Protein concentrations were determined by Bradford assay (Bio-Rad, Hercules, USA) using bovine serum albumin (BSA) as a standard. Fractions containing purified protein were pooled and stored at −20 ◦ C.

80

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85

Fig. 1. Nucleotide sequence alignment between the wild-type (GenBank accession no. AY885225) and optimized ORF2 gene. Full-length ORF2 of PCV2 (702 bp) was optimized for E. coli codon usage. Optimized codons are boxed and changed nucleotides are in red. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2.5. Transmission electron microscopy (TEM) Purified rCap protein was adsorbed onto a carbon-coated copper grid for 3 min at room temperature. Grids were dried gently using filter paper and stained with 2% uranyl acetate for 40 s. Excess liquid was removed with filter paper, and samples were observed under a JEOL JEM 1400 TEM (JEOL Ltd., Tokyo, Japan) at an accelerating voltage of 120 kV. 2.6. Extraction of nucleic acids from PCV2 VLPs Encapsidation of exogenous nucleic acid into E. coli-derived PCV2 s was verified by extracting nucleic acids using QIAamp DNA Mini Kits (Qiagen, Venlo, Netherlands). Purified rCap VLPs (200 ␮l) were treated with 20 units DNase I (NEB, Ipswich, USA) to remove attached DNA. RNA and DNA were co-purified by QIAamp Mini spin columns, and eluted in 50 ␮l elution buffer. 15 ␮l nucleic acid extract was further treated with 2 or 10 units DNase I at 37 ◦ C for 10 min. Extracted nucleic acids were examined by pulsed-field gel electrophoresis with 1% agarose gels. 2.7. Immunization of mice A total of 9 specific-pathogen-free (SPF) female BALB/c mice (6 weeks old) were randomly allotted to an immunized group (n = 6) or control group (n = 3). Mice were inoculated

intraperitoneally with 200 ␮l containing 30 ␮g purified rCap protein or 200 ␮l PBS (control group) at a 1:1 volume ratio of water-in-oil-in-water emulsion with the adjuvant Montanide ISA201 (SEPPIC, Paris, France) at 2-week intervals. Blood samples were collected at 0, 10, 24, 35, 39 days after immunization and sera were analyzed for PCV2Cap-specific antibody responses using an indirect enzyme-linked immunosorbent assay (ELISA). For each group, three mice were euthanized 39 days after immunization. Sera and splenocytes were collected for serum neutralization tests and IFN-␥ enzyme-linked immunospot (ELISpot) assays. 2.8. Indirect ELISA ELISA plates (96-well NUNC MaxiSorpTM , Thermo Fisher Scientific, Waltham, USA) were coated with 3.5 ␮g/ml baculovirus/insect cell-produced PCV2 VLPs in 50 ␮l coating buffer (14.2 mM Na2 CO3 , 34.9 mM NaHCO3 , 3.1 mM NaN3 , pH 9.6) overnight at 4 ◦ C. Wells were washed with 200 ␮l PBST (PBS containing 0.05% Tween-20) and blocked with 100 ␮l blocking buffer (PBS containing 1% BSA) at 37 ◦ C for 2 h. After washing with PBST, wells received 50 ␮l tested mouse serum (1:800 diluted) in blocking buffer and were incubated at 37 ◦ C for 1 h. After washing, 50 ␮l of 4000-fold-diluted goat anti-mouse IgG (H + L) horseradish peroxidase conjugate (Zymed Laboratories Inc., South San Francisco, USA) in blocking buffer was added into wells and incubated at 37 ◦ C for 1 h. Plate were washed with PBST three times followed by washing with PBS

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85

twice and addition of 50 ␮l ready-to-use ABTS single-solution chromogen/substrate (Zymed Laboratories Inc., San Francisco, USA) and incubated at room temperature for 15 min in the dark. An equal volume of 0.01% sodium azide in 0.1 M citric acid was added to stop the reactions. Optical density at 405 nm (OD405 nm ) was measured using a MRX II Model 96 Well Plate Reader (DYNEX Technologies Inc., Chantilly, USA). Samples were analyzed in duplicate and OD405 nm of duplicates was averaged. Negative controls were SPF mouse sera and positive controls were hyperimmune mouse sera specific to PCV2 using mean OD405 nm values. The cut-off value was the mean OD405 nm value for all SPF mice sera plus three standard deviations (SD).

81

3. Results 3.1. High-level expression of full-length codon-optimized ORF2 gene in E. coli The recombinant expression plasmid pET24a-rORF2 containing the full-length codon-optimized ORF2 gene was constructed for efficient expression of Cap proteins in E. coli. High-level production of soluble 27 kDa molecular weight was seen in the supernatant fraction by SDS-PAGE. Specificity was confirmed by Western blot using anti-PCV2 swine immune serum (Fig. 2). 3.2. Purification of rCap-protein VLPs by SEC

2.9. Serum neutralization test Replicates of two-fold dilutions starting at 1:2 of heatinactivated (30 min, 56 ◦ C) serum samples were mixed with 200 TCID50 PCV2 virus stock in microtitration plates and incubated at 37 ◦ C for 1 h. After addition of 1 × 104 porcine kidney (PK-15) cells and incubation at 37 ◦ C for 72 h, infected cells were washed once with PBS, fixed with 10% formalin solution (Thermo Fisher Scientific, Waltham, USA) for 10 min, and washed with PBS three times. Cells were sequentially treated with 1% BSA/PBS, with an appropriate dilution of P94 swine immune serum, and with 1000-fold dilution of Alexa Fluor® 488 AffiniPure Goat Anti-Swine IgG (H + L) (Jackson ImmunoResearch Inc., West Grove, USA). Cells were visualized using an Olympus IX70 inverted fluorescence phase contrast microscope. Neutralization titers are expressed as the reciprocal of the highest dilution of serum sample that caused 50% neutralization. 2.10. ELISpot assays PCV2-specific IFN-␥ SC frequencies were quantified to evaluate immune cell activation in splenocytes by ELISpot assays. MultiScreenTM 96-well assay plates (EMD Millipore, Billerica, USA) were coated with 100 ␮l 1:60 diluted IFN-␥ capture antibody (R&D Systems Inc., Minneapolis, USA) at 4 ◦ C overnight. After washing with PBS three times, 200 ␮l blocking buffer (1% BSA and 5% sucrose in PBS) was added to each well and incubated at room temperature for 2 h before 100 ␮l of RPMI 1640 medium (Gibco® , Invitrogen Life Technologies, Grand Island, USA) containing 5 × 105 mouse splenocytes were added to each well with 100 ␮l baculovirus/insect cell-produced PCV2 VLPs (40 ␮g/ml) at 37 ◦ C in a 5% CO2 incubator for 24 h. Incubation with 100 ␮l Concanavalin A (40 ␮g/ml, Sigma–Aldrich, St. Louis, USA) was the positive control and RPMI 1640 medium was the negative control. Wells were treated with 100 ␮l 1:60 diluted IFN-␥ detection antibody (R&D Systems Inc., Minneapolis, USA) at 4 ◦ C overnight, and100 ␮l 1:60 diluted streptavidin-AP (R&D Systems Inc., Minneapolis, USA) for 2 h at room temperature and 100 ␮l BCIP/NBT solution (R&D Systems Inc., Minneapolis, USA) in the dark for 30 min at room temperature. Spot numbers were determined with a camera and analyzed using Image-Pro Plus image analysis software (Media Cybernetics Inc., Rockville, USA). The number of spots in negative controls was subtracted from counts of spot-forming cells in stimulated wells. Data are expressed as numbers of IFN-␥ SC per 2 × 105 splenocytes. 2.11. Statistical analysis Statistical differences among groups were assessed by ANOVA (Graphpad prism version 5, GraphPad Software Inc., La Jolla, USA). All differences were considered significant at P < 0.05(*), <0.01(**), <0.001(***) and <0.0001(****), respectively.

The soluble rCap proteins assembled into VLPs and were purified through a HiPrep 16/60 Sephacryl S-500 High Resolution column connected to an ÄKTAFPLCTM system. SDS-PAGE and Western blotting were performed on 14 fractions (Fig. 3). Cap VLPs were mainly present in fractions 6 through 9, and the fractions 6 and 7, which had higher purity, were pooled for further use. The yield of purified Cap proteins was approximately 10 mg per 200 ml bacterial culture and rCap protein purity was as high as 95% after gel-filtration chromatography. 3.3. E. coli-derived rCap protein packaged exogenous nucleic acids and assembled spontaneously into VLPs To determine whether PCV2 rCap self-assembled into VLPs in E. coli, the morphology of purified rCap proteins without additional tags was examined by TEM. Purified rCap proteins appeared as virion-like structures with diameters from approximately 25 to 30 nm (Fig. 4a). These structures were not detected in cell lysates from E. coli BL21 (DE3) (data not shown). Based on the TEM results, E. coli nucleic acids were hypothesized to be encapsidated by the rCap proteins, because the size of E. coli-derived VLPs was larger than authentic PCV2 capsids. E. coli nucleic acids were extracted from rCap VLPs and analyzed on a 1% agarose gel. Before extraction, DNase I was used to remove any remaining cellular DNA, which might adhere to the exterior surface of the particles. Nucleic acids from nuclease-treated particles were treated with 2 or 10 units of DNase I and analyzed by agarose gel electrophoresis (Fig. 4b). The amount of cellular nucleic acid extracted from nuclease-treated VLPs progressively decreased with increasing nuclease treatment. The results indicated that rCap VLPs in E. coli contained exogenous DNA molecules. 3.4. VLPs elicited both humoral and cell-mediated immune responses to PCV2 in immunized mice Six SPF mice were immunized intraperitoneally with purified rCap VLPs at 2-week interval three times, control group mice (n = 3) received PBS. Serum samples were collected to evaluate the humoral immune response by indirect ELISA and neutralizing antibody assays. At 39 days after immunization, three mice of each group were euthanized and splenocytes were collected to analyze the cell-mediated immune response by IFN-␥ ELISpot assay. Specific IgG antibody responses in mice against PCV2 were detected by indirect ELISA using baculovirus-derived PCV2 VLPs as antigen (Fig. 5). The cut-off value for the mean OD405 nm for all SPF mouse sera with 3 SD was 0.166 and all mice tested negative for specific PCV2 antibodies at 0 and 10 days after primary immunization. Cap-specific antibodies were induced in rCap VLP-immunized mice after booster immunization. Significantly increased antibody titers were appeared at 35 and 39 days after immunization, when mice received a second booster immunization. In contrast, PCV2 antibody titers in the control group were not detectable at all times. All

82

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85

Fig. 2. Expression of rCap in E. coli. Rcombinant plasmid pET24a-rORF2 (lanes 1–2) or pET24a (lanes 3–4) transformed E. coli BL21 (DE3) cells were cultured and induced with 1 mM IPTG for 3 h at 37 ◦ C. Cells were harvested and disrupted by sonication. Total lysate were clarified by centrifugation and divided into supernatant (S) and pellet (P) fractions, followed by 12% SDS-PAGE (a) and Western blot analysis with swine immune serum (b). Predicted rCap protein is indicated by an arrow.

Fig. 3. Purification of rCap protein by size-exclusion chromatography. Soluble E. coli expressed rCap proteins were fractionated for into 14 fractions (lanes 1–14) after gel filtration followed by 12% SDS-PAGE (a) and Western blotting with swine immune serum (b). E. coli transformed with pET24a was the control.

Fig. 4. TEM image of rCap VLPs (a) and agarose gel analysis of DNA extracted from VLPs (b). Purified VLPs were negatively stained and observed by TEM. Scale bar is 20 nm. VLPs were treated with DNase I (lane 1) or untreated (lane 4) followed by nucleic acid extraction. The extracted nucleic acids from lane 1 were further digested with 2 units (lane 2) or 10 units DNase I (lane 3).

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85

83

4. Discussion

Fig. 5. Time-course for development of antibodies to PCV2 after rCap-VLP immunization of mice as detected by indirect ELISA. Immunized mice received 200 ␮l with 30 ␮g purified rCap VLPs at 2-week-intervals. Control mice received 200 ␮l PBS. PCV2Cap-specific antibodies were detected by indirect ELISA using baculovirus/insect cell-produced PCV2 VLPs as antigen. ELISA results are represented as OD405 nm values with data as the means ± SD. Asterisk indicates antibody titers of the immunized group were significantly higher than the control group (**P < 0.01; ****P < 0.0001).

experimental mice remained healthy during the study (data not shown). This result suggested that E. coli-derived VLPs elicited a specific immune response against PCV2. Serum samples were collected from three mice in each group at 39 days after immunization and titers of neutralizing antibodies against PCV2 were determined in PK15 cells. Immunized mice induced 1:8–1:12 neutralizing antibody titers against PCV2, significantly higher than the non-specific neutralization seen with control sera (P < 0.001) (Fig. 6a). Immunogenicity by indirect ELISA and neutralizing antibody assays demonstrated that immunization with rCap VLPs elicited humoral immune responses in mice. The cellular immune response induced by rCap VLPs was characterized by measuring IFN-␥ SC responsiveness in vitro stimulated with baculovirus/insect cell-produced PCV2 VLPs (Fig. 6b). Immunization with rCap VLPs triggered strong cell activation to induce IFN-␥ (P < 0.05). Taken together, these results showed that rCap VLPs were an efficient antigen for eliciting both humoral and cellmediated immune responses against PCV2.

This study used the entire PCV2 ORF2 nucleotide sequence optimized for E. coli codon usage and a pET24a (+) expression system. The pET expression system has high-efficiency translation and is one of the most commonly used systems for cloning and in vivo expression of recombinant proteins in E. coli. Overexpression of recombinant proteins in E. coli has a tendency to aggregate into insoluble inclusion bodies. In the Western blot analysis, the majority of expressed rCap proteins (27 kDa) were present at the supernatant fraction and the minor bend with approximately 55 kDa in size might be the dimmer form of rCap (Fig. 2B). We could obtain significantly increased yields of water-soluble pET24a-expressed rCap protein in E. coli BL21 (DE3) cells leading to formation of VLPs. Full-length codon optimization compared to our previous study with four codons optimized at 5 end (Wu et al., 2012) resulted in higher level of rCap expression (data not shown). Advantages to this system are that it is easy to maintain, simple to scale-up and less expensive to use a bacterial expression system to express recombinant proteins than eukaryotic expression systems (Yin et al., 2010). Marcekova et al. (2009) synthesized an entire codon-optimized PCV2 ORF2 sequence but it is not expressed in E. coli unless the optimized-ORF2 gene has a polyhistidine tag at the N-terminus. However, a fusion protein tag might influence Cap protein assembly and protease cleavage of the fusion tag with purification could be cumbersome and costly. In contrast, we deleted the T7 and His-tag sequence from the pET24a (+) vector by PCR and successfully expressed rCap protein without a fusion tag in E. coli. The rCap proteins self-assembled into VLPs that could be simply separated from unrelated bacterial proteins by size exclusion chromatography using a HiPrep 16/60 Sephacryl S-500 High Resolution column. This single-step gel filtration chromatography approach is convenient and suitable for obtaining substantial quantities of rCap protein at high purity. This system has the potential to be used for studies or serological diagnostic assays and a production-scale PCV2 vaccination program. Recently, Trible et al. (2012) proposed a decoy mechanism for PCV2 that is based on the recognition of different structural forms of Cap. In general, antibodies are directed against the whole virion (VLP) which results in inducing neutralizing antibodies production. In contrast, the recognition of monomer Cap or Cap fragments results in the production of non-neutralizing antibodies against the immunodominant decoy epitope (amino acids 169–180)

Fig. 6. Serum neutralizing antibody response (a) and frequency of PCV2-specific IFN-␥ SC (b) after immunization with VLPs. Neutralizing antibody titers against PCV2 were detected at 39 days after immunization (a). Levels of IFN-␥ SC in splenocytes of immunized and control mice at 39 days after immunization were analyzed upon in vitro re-stimulation with 4 ␮g baculovirus/insect cell-produced PCV2 VLPs for 24 h by ELISpot assays (b). Data are means ± SD for three independent experiments. Asterisk indicates that immunized group is statistically different than the control group (*P < 0.05; ***P < 0.001).

84

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85

eventually leading to disease progression (Trible et al., 2012). Therefore, spontaneous formation of VLPs by rCap proteins is critical to designing PCV2 vaccines since the decoy epitope is buried within the VLP structure. Electron microscopy revealed that the morphology of particles formed by rCap proteins in E. coli resembled VLPs (Fig. 4a). The rCap VLPs were found approximately 25–30 nm in diameter, which is larger than PCV2 virion particles. VLPs in yeast expressing optimized Cap protein are also variable, ranging from 14.9 to 30 nm (mean 25.9 nm) in diameter (Bucarey et al., 2009). Some E. coli-produced VLPs were less homogeneous in size and shape than native PCV2 virions, possibly because of misfolding or lack of post-translational modification machinery in bacterial expression systems. A similar observation suggested that differences may be due to the absence of minor structural proteins or to slightly impaired DNA regulation (Fenaux et al., 2002). Extraction of E. coli-expressed rCap VLPs demonstrated that DNA molecules were encapsidated in the particles (Fig. 4b). This is in good agreement with our hypothesis that the Cap N-terminal portion, which is enriched with basic amino acids, preferentially binds to the bacterial genomic DNA, and therefore, formation of larger VLPs due to packaging with a variety of bacterial DNA molecules. The encapsidation of PCV2Cap proteins is required further investigation. PCV2 vaccines are highly effective and are widely used to control PCV2 infection (Lyoo et al., 2012). Studies on the effectiveness of experimental or commercial vaccines in SPF or conventional pigs shows the induction of virus-specific antibodies and IFN-␥ SC associated with reduction of viremia, shedding and viral load in tissues but with subclinical outcomes (Chae, 2012; Ferrari et al., 2014; Kekarainen et al., 2010). Therefore, development of a safe, cost-effective, efficient vaccine for protecting pigs against PCV2 infection has become a necessary strategy. In this study, a mouse model was used to evaluate E. coli-derived rCap VLPs for potential applications in PCV2 vaccines. PCV2 vaccines that have been efficacy-tested in mice provide protection to swine (Kim et al., 2009; Opriessnig et al., 2009). Immunization of mice with rCap VLPs induced a potent humoral immune response as demonstrated by the elicitation of PCV2-specific and neutralizing antibodies (Figs. 5 and 6 a). The onset of humoral immunity as total specific and neutralizing antibodies upon PCV2 natural or experimental infection is an important response to counteract the onset of clinical signs (Ferrari et al., 2014). Immunization with rCap VLPs significantly induced the secretion of IFN-␥ in mice splenocytes, indicating a potent cell-mediated immune response (Fig. 6b). The high levels of IFN-␥ secreting cells induced by VLP-immunized mice splenocytes suggested a specific Th1-dependent immune response that potentiated the activation of cytotoxic effector cells to eliminate infected cells and promoted virus-specific antibody production (Gerner et al., 2009). Thus, PCV2 VLPs efficiently produced by expressing rCap protein in E. coli followed by single-step gel-filtration purification elicited both antibody and T cell responses against PCV2 in mice. Production of rCap VLPs in E. coli might be an alternative strategy to producing a new generation of vaccines against PCV2 infections with advantages of easy preparation and low cost. Further investigation of E. coli-derived rCap VLPs for anti-PCV2 vaccines for pigs is currently in progress.

References Allan, G.M., Ellis, J.A., 2000. Porcine circoviruses: a review. J. Vet. Diagn. Invest. 12, 3–14. Antonis, A.F., Bruschke, C.J., Rueda, P., Maranga, L., Casal, J.I., Vela, C., Hilgers, L.A., Belt, P.B., Weerdmeester, K., Carrondo, M.J., Langeveld, J.P., 2006. A novel recombinant virus-like particle vaccine for prevention of porcine parvovirus-induced reproductive failure. Vaccine 24, 5481–5490.

˜ Bucarey, S.A., Noriega, J., Reyes, P., Tapia, C., Sáenz, L., Zuniga, A., Tobar, J.A., 2009. The optimized capsid gene of porcine circovirus type 2 expressed in yeast forms virus-like particles and elicits antibody responses in mice fed with recombinant yeast extracts. Vaccine 27, 5781–5790. Chae, C., 2012. Commercial porcine circovirus type 2 vaccines: efficacy and clinical application. Vet. J. 194, 151–157. Crowther, R.A., Berriman, J.A., Curran, W.L., Allan, G.M., Todd, D., 2003. Comparison of the structures of three circoviruses: chicken anemia virus, porcine circovirus type 2, and beak and feather disease virus. J. Virol. 77, 13036–13041. Fachinger, V., Bischoff, R., Jedidia, S.B., Saalmüller, A., Elbers, K., 2008. The effect of vaccination against porcine circovirus type 2 in pigs suffering from porcine respiratory disease complex. Vaccine 26, 1488–1499. Fenaux, M., Halbur, P.G., Haqshenas, G., Royer, R., Thomas, P., Nawagitgul, P., Gill, M., Toth, T.E., Meng, X.J., 2002. Cloned genomic DNA of type 2 porcine circovirus is infectious when injected directly into the liver and lymph nodes of pigs: characterization of clinical disease, virus distribution, and pathologic lesions. J. Virol. 76, 541–551. Fenaux, M., Opriessnig, T., Halbur, P.G., Meng, X.J., 2003. Immunogenicity and pathogenicity of chimeric infectious DNA clones of pathogenic porcine circovirus type 2 (PCV2) and nonpathogenic PCV1 in weanling pigs. J. Virol. 77, 11232–11243. Ferrari, L., Borghetti, P., De Angelis, E., Martelli, P., 2014. Memory T cell proliferative responses and IFN-␥ productivity sustain long-lasting efficacy of a Cap-based PCV2 vaccine upon PCV2 natural infection and associated disease. Vet. Res. 45, 44. Gerner, W., Käser, T., Saalmüller, A., 2009. Porcine T lymphocytes and NK cells—an update. Dev. Comp. Immunol. 33, 310–320. Horlen, K.P., Dritz, S.S., Nietfeld, J.C., Henry, S.C., Hesse, R.A., Oberst, R., Hays, M., Anderson, J., Rowland, R.R., 2008. A field evaluation of mortality rate and growth performance in pigs vaccinated against porcine circovirus type 2. J. Am. Vet. Med. Assoc. 232, 906–912. Kekarainen, T., McCullough, K., Fort, M., Fossum, C., Segalés, J., Allan, G.M., 2010. Immune responses and vaccine-induced immunity against porcine circovirus type 2. Vet. Immunol. Immunopathol. 136, 185–193. Kim, T., Toan, N.T., Seo, J., Jung, B., Lee, J., Lee, B., 2009. Bordetella bronchiseptica aroA mutant as a live vaccine vehicle for heterologous porcine circovirus type 2 major capsid protein expression. Vet. Microbiol. 138, 318–324. Kixmöller, M., Ritzmann, M., Eddicks, M., Saalmüller, A., Elbers, K., Fachinger, V., 2008. Reduction of PMWS-associated clinical signs and co-infections by vaccination against PCV2. Vaccine 26, 3443–3451. Kurmann, J., Sydler, T., Brugnera, E., Buergi, E., Haessig, M., Suter, M., Sidler, X., 2011. Vaccination of dams increases antibody titer and improves growth parameters in finisher pigs subclinically infected with porcine circovirus type 2. Clin. Vaccine Immunol. 18, 1644–1649. Liu, Q., Tikoo, S.K., Babiuk, L.A., 2001. Nuclear localization of the ORF2 protein encoded by porcine circovirus type 2. Virology 285, 91–99. Liu, L.J., Suzuki, T., Tsunemitsu, H., Kataoka, M., Ngata, N., Takeda, N., Wakita, T., Miyamura, T., Li, T.C., 2008. Efficient production of type 2 porcine circovirus-like particles by a recombinant baculovirus. Arch. Virol. 153, 2291–2295. Ludwig, C., Wagner, R., 2007. Virus-like particles-universal molecular toolboxes. Curr. Opin. Biotechnol. 18, 537–545. Lyoo, K.S., Joo, H.S., Davies, P.R., Han, J.H., 2012. Comparison of porcine circovirus type 2 (PCV2) infection in light and heavy pigs of market age on farms with routine PCV2 vaccination. Can. J. Vet. Res. 76, 221–223. Mahé, D., Blanchard, P., Truong, C., Arnauld, C., Le Cann, P., Cariolet, R., Madec, F., Albina, E., Jestin, A., 2000. Differential recognition of ORF2 protein from type 1 and type 2 porcine circoviruses and identification of immunorelevant epitopes. J. Gen. Virol. 81, 1815–1824. Marcekova, Z., Psikal, I., Kosinova, E., Benada, O., Sebo, P., Bumba, L., 2009. Heterologous expression of full-length capsid protein of porcine circovirus 2 in E. coli and its potential use for detection of antibodies. J. Virol. Methods 162, 133–141. Martelli, P., Ferrari, L., Morganti, M., De Angelis, E., Bonilauri, P., Guazzetti, S., Caleffi, A., Borghetti, P., 2011. One dose of a porcine circovirus 2 subunit vaccine induces humoral and cell-mediated immunity and protects against porcine circovirus-associated disease under field conditions. Vet. Microbiol. 149, 339–351. Meng, X.J., 2013. Porcine circovirus type 2 (PCV2): pathogenesis and interaction with the immune system. Annu. Rev. Anim. Biosci. 1, 43–64. Nawagitgul, P., Morozov, I., Bolin, S.R., Harms, P.A., Sorden, S.D., Paul, P.S., 2000. Open reading frame 2 of porcine circovirus type 2 encodes a major capsid protein. J. Gen. Virol. 81, 2281–2287. Noad, R., Roy, P., 2003. Virus-like particles as immunogens. Trends. Microbiol. 11, 438–444. Opriessnig, T., Patterson, A.R., Jones, D.E., Juhan, N.M., Meng, X.J., Halbur, P.G., 2009. Limited susceptibility of three different mouse (Mus musculus) lines to porcine circovirus-2 infection and associated lesions. Can. J. Vet. Res. 73, 81–86. Opriessnig, T., Prickett, J.R., Madson, D.M., Shen, H.G., Juhan, N.M., Pogranichniy, R.R., Meng, X.J., Halbur, P.G., 2010. Porcine circovirus type 2 (PCV2)-infection and re-inoculation with homologous or heterologous strains: virological, serological, pathological and clinical effects in growing pigs. Vet. Res. 41, 31. Segalés, J., Allan, G.M., Domingo, M., 2005. Porcine circovirus diseases. Anim. Health Res. Rev. 6, 119–142. Trible, B.R., Suddith, A.W., Kerrigan, M.A., Cino-Ozuna, A.G., Hesse, R.A., Rowland, R.R., 2012. Recognition of the different structural forms of the capsid protein

P.-C. Wu et al. / Journal of Biotechnology 220 (2016) 78–85 determines the outcome following infection with porcine circovirus type 2. J. Virol. 86, 13508–135014. Trundova, M., Celer, V., 2007. Expression of porcine circovirus 2 ORF2 gene requires codon optimized E. coli cells. Virus Genes 34, 199–204. Vincent, I.E., Balmelli, C., Meehan, B., Allan, G., Summerfield, A., McCullough, K.C., 2007. Silencing of natural interferon producing cell activation by porcine circovirus type 2 DNA. Immunology 120, 47–56. Wu, P.C., Chien, M.S., Tseng, Y.Y., Lin, J., Lin, W.L., Yang, C.Y., Huang, C., 2008. Expression of the porcine circovirus type 2 capsid protein subunits and application to an indirect ELISA. J. Biotechnol. 133, 58–64. Wu, P.C., Lin, W.L., Wu, C.M., Chi, J.N., Chien, M.S., Huang, C., 2012. Characterization of porcine circovirus type 2 (PCV2) capsid particle assembly and its application to virus-like particle vaccine development. Appl. Microbiol. Biotechnol. 95, 1501–1507.

85

Yin, S., Sun, S., Yang, S., Shang, Y., Cai, X., Liu, X., 2010. Self-assembly of virus-like particles of porcine circovirus type 2 capsid protein expressed from E. coli. Virol. J. 7, 166. Zhou, J.Y., Shang, S.B., Gong, H., Chen, Q.X., Wu, J.X., Shen, H.G., Chen, T.F., Guo, J.Q., 2005. In vitro expression, monoclonal antibody and bioactivity for capsid protein of porcine circovirus type II without nuclear localization signal. J. Biotechnol. 118, 201–211.