UV-C irradiation as an effective tool for sterilization of porcine chimeric VP1-PCV2bCap recombinant vaccine

Ultraviolet irradiation is an effective method of virus and bacteria inactivation. The dose of UV-C light necessary for baculovirus inactivation by measurement of fluorescent GFP protein produced by baculovirus expression system after the irradiation of baculovirus culture in doses ranging from 3.5 to 42 J/m2 was determined. At a dose of 36.8 J/m2, only 0.5% of GFP-expressing cells were detected by flow cytometry and confocal microscopy. The stability of purified VP1-PCV2bCap protein produced by baculovirus expression system was analyzed after the irradiation at doses ranging from 3.5 to 19.3 J/m2. Up to the dose of 11 J/m2, no significant effect of UV-C light on the stability of VP1-PCV2bCap was detected. We observed a dose-dependent increase in VP1-PCV2bCap-specific immune response in BALB/c mice immunized by recombinant protein sterilized by irradiation in dose 11 J/m2 with no significant difference between vaccines sterilized by UV-C light and filtration. A substantial difference in the production of VP1-PCV2bCap specific IgG was observed in piglets immunized with VP1-PCV2bCap sterilized by UV-C in comparison with protein sterilized by filtration in combination with the inactivation of baculovirus by binary ethylenimine. UV-C irradiation represents an effective method for vaccine sterilization, where commonly used methods of sterilization are not possible.

www.nature.com/scientificreports/In this work, we describe the purification of chimeric VP1-PCV2bCap recombinant protein complexes by affinity chromatography and analysis of the stability and immunogenicity of this protein after the irradiation by UV-C light and effect of UV-C light on the degradation of baculovirus as a possible contaminant of recombinant vaccines produced by BEVS.VP1-PCV2bCap represents a modern and safe immunizing agent capable of inducing a strong humoral immune response against PCV2 infection with significant neutralizing activity.Moreover, this chimeric antigen containing VP1-PCV2bCap protein sequences based on the mouse polyomavirus (MPyV) has a so-called DIVA (differentiating infected from vaccinated animals) vaccine potential, which can induce an immune response against the mouse polyoma VP1 protein and distinguish between PCV2 naturally infected and vaccinated animals.

Results
SF9 cells were infected with recombinant baculovirus containing Green Fluorescent Protein gene (Bac-GFP) irradiated by UV-C light at doses ranging from 3.5 to 42 J/m 2 .The effect of UV light on the viability of Bac-GFP was evaluated by monitoring GFP-fluorescent SF9 cells by flow cytometry (Fig. 1A) and confocal microscopy (Fig. 1B), where the fluorescent SF9 cells corresponded to cells infected by surviving Bac-GFP capable to express GFP protein.We detected Bac-infected cells expressing GFP protein in cultures treated with Bac-GFP irradiated by UV-C at doses ranging from 3.5 to 32.1 J/m 2 .The percentage of Bac-GFP-positive cells gradually decreased with increasing levels of UV-C irradiation.The result was comparable for both methods.In addition to decreasing counts of fluorescent SF9 cells with increasing dose of UV-C irradiation, we also observed gradual reduction of fluorescence intensity that probably corresponded to structural changes of GFP protein after higher doses of irradiation.Bac-GFP irradiation at a dose of 36.8J/m 2 UV-C light resulted in the detection of less than 0.5% of Bac-GFP-infected cells and no infected cells were observed after Bac-GFP irradiation at 42 J/m 2 .The effect of UV-C light on baculovirus DNA damage at the molecular level was verified by real-time PCR in samples of DNA isolated from SF9 cells infected with Bac-GFP and irradiated by UV-C light at doses ranging from 3.5 to 42 J/m 2 (Fig. 2).The amount of DNA gradually decreased with the dose of UV-C from 3.5 to 27.5 J/m 2 and dramatically reduced at the dose of 32.1 J/m 2 and higher.
To determine the virulence of baculovirus after microfluidization, which is commonly used method for cell lysis and further isolation of protein of interest, we applied the lysates from Bac-GFP infected SF9 cells to noninfected SF9 cells.Infectivity of Bac-GFP after the lysis by microfluidization was analyzed by flow cytometry.Less than 1% of positive cells were detected in samples of cells infected by Bac-GFP diluted 10 ×, 100 × or 1.000 × (Fig. 3).Less than 0.5% of positive cells were detected after the process of microfluidization complemented with DENARASE treatment and purification across the CoNTA column (data not shown).
VP1-PCV2bCap protein was expressed in SF9 cells and purified by CoNTA affinity chromatography in the high purity (Fig. 4A).We were able to obtain 2 mg of purified protein from 300 mL of cell culture.The stability of purified protein after the irradiation by UV-C light in dose 3.5 J/m 2 , 7.2 J/m 2 , 11 J/m 2 , and 19.3 J/m 2 was analyzed by SDS-PAGE and western blot with antibody detecting His-tag of purified protein.The protein stayed relatively stable after the irradiation at a dose up to 11 J/m 2 of UV-C.Further increase of UV-C dose led to partial breakdown of the protein (Fig. 4B).Therefore, the dose of 11 J/m 2 of UV-C that can substantially decrease the viability of baculovirus and does not affect VP1-PCV2bCap protein stability was applied to sterilize VP1-PCV2bCap protein mixture which was further tested on animals.
The immunogenicity of the protein mixture sterilized by UV-C irradiation in dose of 11 J/m 2 was analyzed by ELISA 3 weeks after the immunization of mice by antigen containing 5 µg, 10 µg, or 25 µg of purified VP1-PCV2bCap protein supplemented with Emulsigen (10%) as an adjuvant.Immunogenicity of VP1-PCV2bCap sterilized by irradiation was compared to VP1-PCV2bCap sterilized by filtration and administered at the same doses (Fig. 5A).High amounts of specific antibodies were detected in all immunized mice 3 weeks after immunization.The levels of specific IgG antibodies were increased in a protein dose-dependent manner with the most robust immune response in mice vaccinated with 25 µg of purified VP1-PCV2bCap protein.No significant differences were observed between groups of mice immunized with protein mixture sterilized by UV-C irradiation compared to mice immunized with protein sterilized only by filtration (Fig. 5B).
The efficacy of VP1-PCV2bCap protein sterilized either by UV-C (11 J/m 2 ) or by binary ethylenimine (BEI) in combination with filtration was then verified in the pigs by the prime/boost immunization of 6-week-old piglets.Humoral immune response was monitored for 6 weeks (Fig. 6A).While the immune response after the first dose was poor with no difference between piglets immunized with VP1-PCV2bCap irradiated and VP1-PCV2bCap filtered, the levels of specific antibodies increased markedly 3 weeks after the booster.Interestingly, VP1-PCV2bCap protein mixture sterilized by UV-C irradiation elicited significantly higher levels of specific IgG antibodies than protein mixture inactivated by BEI and sterilized by filtration (Fig. 6B).
Production of effective neutralizing antibodies was determined by neutralization assay based on the infection of porcine PK-15 kidney cells by mixture of PCV2 Stoon 1010 virus inoculum with inactivated piglet sera collected at 6 weeks after the first immunization.As positive control PK-15 cells treated with serum of control group of piglets was used.Sera of piglets immunized by mixture of purified protein sterilized by UV-C irradiation showed a better neutralization effect as sera of piglet immunized by protein mixture sterilized by filtration after the inactivation by BEI (Fig. 7).

Discussion
UV-C irradiation is simple, effective, and safe sterilization method to dispose of bacteria and viruses 7 .This method was tested for the inactivation of many human and animal viruses.The minimum dose of irradiation used for inactivation of microorganisms depends on the morphological features of microorganisms and probably on the distance from the source of irradiation.As a rule, nonenveloped viruses are more resistant to ultraviolet irradiation than enveloped viruses 8,9 .However, this is not a strict guideline, for example, porcine parvovirus (PPV) a nonenveloped virus is highly sensitive to UV-C requiring only 2.4 J/m 2 dose of irradiation for complete www.nature.com/scientificreports/inactivation 10 , whereas porcine circovirus (PCV) another nonenveloped virus requires about 24 J/m 211 .Baculoviruses represent double stranded DNA viruses highly resistant to UV-C irradiation due to their ability to repair genomes during replication processes within the host cells 12 .In the case of Bac-GFP, a dose of 36.8J/m 2 completely inactivated baculoviruses, and no expression of GFP protein was detected by flow cytometry and confocal microscopy.Similar results were observed after the real-time PCR analysis.We detected a gradually decreased amount of baculovirus DNA up to dose 27.5 J/m 2 of irradiation and dramatic decreased concentration of DNA at the dose of 32.1 J/m 2 and higher.This is in agreement with previously published data, which describes false-positive signals by real-time PCR when the damage of the viral genome is less than 1:1 13 .Real-time PCR is also not capable to distinguish between infectious and noninfectious viral genomes 14 .Importantly, we showed that the virulence of baculovirus is radically decreased after microfluidization step during the protein isolation process.However, sterilization of the final product by an appropriate method is necessary to inactivate the residual baculovirus which could potentially be present in the protein mixture.
Irradiation by UV-C light leads to the degradation of pathogens and nucleic acids with less damage of surface antigenic epitopes of produced protein and minimal changes in immune response to the vaccine in comparison to gamma radiation or chemical sterilization 1 .Germicidal UV light (200-300 nm) produces photodimeric lesions between adjacent pyrimidine nucleotides in deoxy and ribonucleic acids of bacteria, viruses, and protozoa.These changes block transcription and replication processes causing the inactivation of the microorganisms 13 .
Many structural and/or functional modifications of proteins caused by UV-C irradiation lead to the breaking of disulfide bonds, cross-linking, aggregation, fragmentation, oxidation, or deamination of proteins, and often a subsequent reduction in the activity of proteins 5 .In the case of our chimeric protein VP1-PCV2bCap, increasing the dose of UV-C light led to partial degradation of the protein detected by lowered band density in western blot analysis.No fragments of chimeric protein evaluated by anti-His tag antibody were detected.www.nature.com/scientificreports/However, minimal damage to chimeric protein was observed after irradiation with the dose of 11 J/m 2 UV-C light.In contrast to the irradiation method, sterilization by filtration led in our case to the loss of considerable amount of purified chimeric protein due to its high molecular weight (365 kDa) and retention of the protein by filter (Supplementary information, Fig. S3).In addition, our data suggest that a substantial part of baculoviruses are degraded after the lysis of the cells by microfluidization.No live baculovirus was observed in elution after the purification through the CoNTA column and so a minimal dose of UV-C irradiation is necessary for the total sterilization of the final product.To compare the immunological effectiveness of VP1-PCV2bCapprotein mixture sterilized by UV-C irradiation vs. filtration, BALB/c mice were immunized in the one-step vaccination protocol.The immune response was dose-dependent, but no significant differences were observed between the two types of diversely sterilized protein mixtures.In the case of in vivo study on piglets, we used UV-C  irradiation or filtration in combination with BEI for the complete inactivation of residual baculoviruses in a protein mixture.We observed a significant increase of specific antibody titers in the group of piglets immunized by protein mixture sterilized by UV-C irradiation 3 weeks after the booster, but a markedly lower response in the group vaccinated with VP1-PCV2bCap sterilized by filtration in combination with BEI inactivation.BEI has been widely tested for the inactivation of viral vaccines [15][16][17] .Valero et al. described the lower neutralization effect of antibodies produced by vaccine inactivated by BEI 15 .This is in agreement with our in vivo study on piglets and the results from the neutralization assay, where we observed a better neutralization effect of antibodies in sera of piglets immunized by protein mixture sterilized by UV-C irradiation in comparison to those vaccinated by protein mixture sterilized by filtration after the inactivation by BEI.We assume that a greater neutralization effect may be due to an overall higher level of IgG specific antibodies in the sera of piglets immunized with VP1-PCV2bCap protein mixture sterilized by UV-C irradiation.The neutralization effect can also correlate with other immune responses.Comparison of T cell and B cell-mediated immune responses to immunization and their contribution to protection would provide more information about the effectiveness of vaccines.Still, UV-C light irradiation represents a promising sterilization strategy for the preparation of recombinant protein vaccines applicable in veterinary practice.

Conclusion
Sterilization of recombinant protein vaccine by UV-C irradiation is a promising strategy for the preparation of large chimeric protein antiviral vaccines, where filtration, pasteurization, alkylation, and treatment with detergents are not possible due to specific properties of chimeric proteins.UV-C irradiation is an effective method of baculovirus inactivation and an appropriate method of sterilization as a key step for the production of safe chimeric recombinant vaccines.

Preparation of recombinant plasmid with GFP (Bac-GFP)
Construction of recombinant plasmid was described previously 18 .Briefly, C-terminal fragment of VP3 protein was amplified by PCR (forward primer: 5′-CAT CAG CGA GCT CAG GGT ACTC-3′ and reverse primer: 5′-TTA GAG GAT CCT TAG AGA CGC CGC TT-3′).The PCR fragment and pEGFP-C2 plasmid (Clontech, Palo Alto, CA) were cut with restriction enzymes Sac I and BamH I and ligated together to generate pEGFP-t-VP3 plasmid.EGFP-t-VP3 sequence was amplified by PCR (forward primer: 5′-AGA TAG GAT CCA CCA TGG TGA GCA AG-3′ and reverse primer: 5′-TTA GAG GAT CCT TAG AGA CGC CGC TT-3′).The PCR product was cut with restriction enzymes Bgl II and BamH I and inserted into the pAcDB3/VP1 plasmid (gift from Drs. T. Ramqvist and T. Dalianis, Karokinska Institut, Department of Oncology and Pathology, Sweden).Recombinant baculovirus Bac-GFP was prepared and purified by plaque assays as was described previously 19 .

Purification of chimeric protein from SF9 insect cells
Insertion of sequence of porcine circovirus 2b (PCV2b) capsid protein into pFastBacI -VP1 DA Ʌ 7 transfer vector, preparation of recombinant baculovirus, generation of high-titer viral stocks and protein production were described previously 6,20 .SF9 cell pellet containing VP1-PCV2bCap recombinant protein was resuspended in native lysis buffer (20 mmol L −1 Tris-HCl, 150 mmol L −1 NaCl, pH 7.6) supplemented with cOmplete, EDTA-free Protease Inhibitor Cocktail (Merck) and lysed by Microfluidizer Processor Cell Disruptor Homogenizer (New Life Scientific) cooled on ice.Cell lysate was centrifuged at 10.000 µg for 30 min and supernatant was treated with 180 U DENARASE (c-LEcta GmbH) and 2 µl of 1 M MgCl 2 per ml of supernatant and incubated for 1 h at 37 °C.Supernatant was centrifuged at 10.500 µg for 10 min and chimeric VP1-PCV2bCap protein was purified under native conditions using CoNTA agarose (Merck) according to the manufacturer's instructions.The purified VP1-PCV2bCap was further used for evaluation of irradiation dose-response effect, protein analysis, and immunization of the animals.The irradiation was performed as a final step after protein purification to inactivate residual baculovirus.

Analysis of the VP1-PCV2bCap stability after the irradiation by Steripen UltraLight
Purified VP1-PCV2bCap protein (60 mL) in concentration 0.2 mg/mL was irradiated by Steripen UltraLight in dose 3.5 J/m 2 , 7.2 J/m 2 , 11 J/m 2 , and 19.3 J/m 2 of UV-C light according to the manufacturer's instructions.VP1-PCV2bCap proteins were analyzed by SDS-PAGE and Western blot.

SDS-PAGE and Western blot analysis
The cell lysate, purified protein with or without irradiation were separated on 10% SDS-PAGE stained with Coomassie Brilliant Blue R-250 or blotted onto polyvinylidene difluoride (PVDF) membrane (Millipore, USA).The PVDF membrane was blocked overnight with 3% non-fat milk and developed with Anti-6X His tag HRP antibody (ABCAM, UK; dilution 1:1.000) followed by the detection using GE Healthcare Amersham ECL Prime Western Blotting Detection Reagent (Thermo Fisher, USA) and analyzed with Azure Biosystems C300 (Azure Biosystems, Inc., USA) and cSeries Capture Software.

Immunization of animals
Experiments were conducted according to the principles enunciated in the Guide for the Care and Use of Laboratory Animals issued by the Czech Society for Laboratory Animal Science and reviewed by the Ethical committee

Figure 2 .
Figure 2. The effect of UV-C light on baculovirus DNA damage.DNA was isolated from SF9 cells infected with Bac-GFP irradiated by UV-C in doses 3.5 J/m 2 , 7.2 J/m 2 , 11 J/m 2 , 15 J/m 2 , 19.3 J/m 2 , 23.2 J/m 2 , 27.5 J/ m 2 , 32.1 J/m 2 , 36.8 J/m 2 and 42 J/m 2 after the incubation for 36 h.The copy number of DNA was analyzed by real-time PCR.Positive control (PC)-SF9 cells infected by Bac-GFP without irradiation.Negative control (NC)uninfected SF9 cells.

Figure 4 .
Figure 4. Analysis of the VP1-PCV2bCap stability after the UV-C irradiation.(A) Purified protein was separated on SDS-PAGE followed by Coomassie Brilliant Blue staining (SDS-PAGE) or analyzed by Western blot (WB) detected by Anti-6X His tag HRP antibody.L cell lysate, P purified protein.(B) Purified VP1-PCV2bCap protein irradiated by UV-C light at a dose of 3.5 J/m 2 , 7.2 J/m 2 , 11 J/m 2 , and 19.3 J/m 2 , separated on SDS-PAGE (UP) and analyzed by western blot developed with Anti-6X His tag HRP antibody (DOWN).The amount of irradiated protein was partially decreased after 11 J/m 2 and each subsequent UV-C irradiation.0 J/ m 2 -VP1-PCV2b before irradiation.

Figure 5 .
Figure 5.Comparison of VP1-PCV2bCap-specific antibody response in mice induced by vaccine containing purified protein sterilized by UV-C irradiation or filtration.Six groups of mice were immunized subcutaneously with mixture of purified VP1-PCV2bCap protein in combination with Emulsigen (10%) sterilized by UV-C irradiation in dose 5 µg, 10 µg or 25 µg or by filtration in the same doses.Control mice were injected with a sterile buffer.Sera were collected directly before immunization and 3 weeks after the immunization.(A) Immunization scheme.(B) Levels of VP1-PCV2bCap-specific IgG antibodies at different time points (D0-day 0/before immunization, 3w-3 weeks) and doses of antigen.*P < 0.05, **P < 0.01, ***P < 0.001 compared to the respective control before immunization (D0) vs. after immunization at 3 weeks.Mean ± SEM.One-way ANOVA with Bonferroni post hoc test.

Figure 6 .
Figure 6.Comparison of VP1-PCV2bCap-specific antibody response in piglets induced by mixture of purified protein sterilized by UV-C irradiation or filtration after the BEI inactivation.Piglets (n = 4) were immunized intramuscularly twice (prime at day 0 and booster at 3 weeks) with 50 µg of purified VP1-PCV2bCap protein alone (without sterilization) or VP1-PCV2bCap in combination with Emulsigen (10%; Emul) sterilized by UV-C irradiation or filtration after the inactivation by BEI.Control piglets (n = 2) were injected only with a sterile buffer.Sera were collected before vaccination, at 3 weeks (before booster) and 6 weeks after the first vaccination.(A) Immunization scheme.(B) Levels of VP1-PCV2bCap-specific IgG antibodies at different time points.*P < 0.05, **P < 0.01 compared to the non-immunized control at 6 weeks.# P < 0.05 indicates significant difference between VP1-PCV2bCap protein (without inactivation) vs. VP1-PCV2bCap + Emulsigen 10% with UV-C irradiation at 6 weeks.Mean ± SEM.One-way ANOVA with Bonferroni post hoc test.

Figure 7 .
Figure 7.The neutralization effect of VP1-PCV2bCap antibodies from piglets immunized with VP1-PCV2bCap protein mixture on PCV2 infected PK-15 cells.The cells were stained with anti-PCV2Cap H9 Alexa Fluor-647 (red) and the nuclei were stained with Hoechst (blue).The representative overlay images (pink) from 3 independent repetitions in 6 areas per well.(A) Positive control cells were infected with PCV2 Stoon 1010 virus and treated with sera of control group.(B) The cells infected with PCV2 Stoon 1010 virus treated with sera of piglets immunized with purified VP1-PCV2bCap protein mixture sterilized by UV-C irradiation.(C) The cells infected with PCV2 Stoon 1010 virus treated with sera of piglets immunized with purified VP1-PCV2bCap protein mixture sterilized by filtration after the BEI inactivation.(D) Non-infected cells were used as a negative control.(E) PCV2 positive cells were counted and the inhibition ratio of PCV2 infection was compared.***P < 0.001 differences between positive control cells infected with PCV2 Stoon 1010 virus treated with sera of control group and cells infected with PCV2 Stoon 1010 virus treated with sera of piglets immunized withVP1-PCV2bCap protein mixture supplemented with Emulsigen (10%) UV-C sterilized or sterilized by filtration after the BEI inactivation.ANOVA with Bonferroni post hoc test.