Recombinant Production of a Novel Fusion Protein: Listeriolysin O Fragment Fused to S1 Subunit of Pertussis Toxin

Background: Some resources have suggested that genetically inactivated PTs bear a more protective effect than chemically inactivated products. This study aimed to produce new version of PT, by cloning an inactive PTS1 in a fusion form with N-terminal half of the LLO pore-forming toxin. Methods: Deposited pdb structure file of the PT was used to model an extra disulfide bond. Codon-optimized ORF of the PTS1 was used to make recombinant constructs of PTS1 and LLO-PTS1 in the pPSG-IBA35 vector. The recombinant PTS1 and LLO-PTS1 proteins were expressed in BL21 DE3 and SHuffle T7 strains of E. coli and purified by affinity chromatography. Cytotoxic effects of the recombinant proteins were examined in the MCF-7 cell line. Results: The purity of the products proved to be more than 85%, and the efficiency of the disulfide bond formation in SHuffle T7 strain was higher than BL21 DE3 strain. No cytotoxicity of the recombinant proteins was observed in MCF-7 cells. Soluble recombinant PTS1 and LLO-PTS1 proteins were produced in SHuffle T7 strain of E. coli with high efficiency of disulfide bonds formation. Conclusion: The LLO-PTS1 with corrected disulfide bonds was successfully expressed in E. coli SHuffle T7 strain. Due to the safety for human cells, this chimeric molecule can be an option to prevent pertussis disease if its immunostimulatory effects would be confirmed in the future.

INTRODUCTION ertussis is a highly contagious respiratory illness caused by Bordetella pertussis bacteria [1] . This disease has been controlled utilizing two types of vaccines: the wP and aP ones [2] . The aP vaccines are subunit vaccines that contain pure and inactive components of B. pertussis cells [3,4] , including detoxified pertussis toxin, filamentous hemagglutinin, pertactin, fimbrial antigens and so on. However, the protectivity of the aP vaccines has not been adequate so that several outbreaks has been reported from different countries since 2012 [5,6] . The causes of the weakness of aP vaccines are yet unclear because it is not known whether the re-emergence of pertussis is due to the vaccine waning immunity or to the fundamental differences in the nature of the immune response caused by aP vaccines compared with the wP vaccines or natural infection [7] . After the presentation of toxin-mediated theory about pertussis disease by Margaret Pittman in 1978 [8] , it was suggested that a suitable inactivated and immunogenic PT is necessary P 34 Iran. Biomed. J. 25 (1): [33][34][35][36][37][38][39][40] and sufficient for a pertussis vaccine [9] . Moreover, a former study reported that fimbria, similar to pertactin, is unnecessary for vaccine-induced immunostimulation [10] . PT is composed of two subunits: A subunit with ADP-ribosyltransferase activity (S1 subunit), which is the main antigen for protection, and B oligomer, which is responsible for the toxin attachment to the target cells and is an important protective antigen in all available pertussis vaccines [11] . Currently, the toxoid components in aP vaccines are chemically inactivated by aldehyde agents. However, more immunogenic PTs are inactivated genetically and are expected to have a longer protective effect than current products, because of the better preservation of the antigenic epitopes [9] . The use of adjuvants in vaccine formulation is also important. Researchers believe that adjuvants can play a critical role in improving immunogenicity in next-generation aP vaccines [12,13] . LLO molecule has been the subject of many research studies, and its role as an adjuvant against several infection and cancer models has been clearly illustrated. In fact, it has been identified that LLO molecule can act as an adjuvant either in chimeric form or as a separately expressed molecule [14][15][16][17][18] .
In this study, a new mutant of PTS1 carrying an extra disulfide bond was designed. This chimeric molecule (LLO-PTS1) was then expressed in BL21 DE3 and SHuffle T7 strains of E. coli and characterized further.

MATERIALS AND METHODS
The E. coli DH5α and BL21 DE3 strains were procured from the Pasteur Institute of Iran (Tehran).
SHuffle®T7 competent cells were prepared from New England Biolabs Inc., USA. DpnI restriction enzyme was obtained from Thermo Fisher Scientific (USA), and other chemicals were provided by molecular biology grade providers (Sigma and Merck Co., Germany).

Disulfide bond prediction
1PRT.pdb file, MODIP software of the National Centre for Biological Sciences Integrated Web ServerV1.0.1 [19] , and Disulfide By Design software version 2.12 [20] were used to model, generate files and analyze the new desired disulfide bond in the structure of the pertussis S1 enzymatic subunit. Threedimensional graphics were generated using YASARA molecular graphics suite [21] .

Gene synthesis
The mutant gene of PTS1 was codon-optimized and chemically synthesized by Pishgaman Gene Transfer Co. (Iran). The synthetic fragment was received in the pUC57 cloning vector. Mutant PTS1 used in the present study had the following mutations: R9K, R13L, and E129G.

Recombinant construct preparation
A set of primers (PerF and PerR) was applied for the amplification of the PTS1 gene. Each primer had a 5' arm that was complementary to each side of the vector backbone, as illustrated in Figure 1. The PCR amplification of the PTS1 fragment was carried out using the above primers to result in PTS1 contain complementary ends to the pPSG-IBA35 expression    (Fig. 2). In addition, the PTS1 gene was amplified with a set of primers, at the forward with a 5' arm complementary to the last codons of LLO and the linker, and at the reverse with a 5' arm hybridizing to the vector backbone. The PCR amplification of LLO and PTS1 genes with these primers were performed to synthesize two gene fragments. The fragments were fused together using a SOEing PCR protocol, thus leading to a PCR product of about 1500 bp size harboring arms complementary to the pPSG-IBA35 vector backbone. This PCR product was used in a quick-change PCR to synthesize the whole vector with LLO-PTS1 fusion gene inserted between the two arms. DpnI-treated PCR products were transformed into chemically competent DH5α bacterial cells using the heat shock method. A single colony from each transformation reaction was cultured in 5-ml LB broth containing 50 μg.mL -1 of ampicillin at 37°C for 16 hours. The recombinant plasmids were extracted by column method (QIAprep Spin Miniprep kit, Qiagen TM , Germany), and the sequence of the construct was confirmed by DNA sequencing.

Recombinant protein expression and purification
Twenty nanograms of the purified plasmids were used to transform BL21 DE3 and SHuffle T7 competent cells by the heat shock method. Multiple colonies of BL21 DE3 and SHuffle T7 transformants were cultured in a LB medium containing 50 μg.mL -1 of ampicillin at 37 °C and 28 °C, respectively, induced by the addition of 0.1 mM of IPTG when OD 600nm reached approximately 0.7. Expression of the recombinant proteins PTS1 and LLO-PTS1 was checked in a number of colonies. The fresh aliquots of recombinant protein expressing clones were used for the inoculation of another 100 mL of the LB medium. Clones were grown until the OD 600nm of 0.7, and the expression was induced by adding 0.1 mM of IPTG. After an overnight induction, the bacteria were harvested by centrifugation at 4500 ×g at room temperature for 10 min. Bacterial pellet was bead-beaten 15 times for 20 seconds with a bench-top vortex with maximum shaking speed (2000 rpm) with 30-second intervals of cooling on ice. Lysis buffer contained Tris (100 mM; pH 8.0), NaCl (300 mM), and glycerol (10%). The soluble fraction was centrifuged, and the supernatant was loaded onto a column filled with Ni-NTA agarose resin (Qiagen) to purify the His-tagged recombinant protein. Equilibration and washing buffers of the chromatography were the same as the lysis buffer. Elution buffer was also the same as the lysis buffer, except for the 300 mM of imidazole. Eluted fractions were analyzed by 12.5% SDS-PAGE for the presence of the recombinant protein, and fractions containing the recombinant protein were dialyzed against the phosphate buffer (pH 8.0) containing 50 mM of NaCl.

Disulfide bond determination
Free SH groups of recombinant proteins purified from BL21 DE3 and SHuffle T7 E. coli strains were examined in a denatured condition (pre-incubating proteins with 6 M of GuHCl for 15 min) using Ellman's reagent. Concentrations of recombinant protein and 5,5′-Dithiobis(2-nitrobenzoic acid) were 2 μM and 1 mM in a cuvette of 300-μl volume, respectively. The absorbance of samples at 412 nm was measured, and the number of free SH groups was calculated for the recombinant PTS1 and LLO-PTS1 proteins purified from BL21 DE3 and SHuffle T7 strains, as stated before [22] .

Cell culture and MTT assay
Human cell line MCF-7 was purchased from the Pasteur Institute of Iran (Tehran). Cells were cultured and maintained in RPMI-1640 medium supplemented with 10% FBS, streptomycin and penicillin (1%), pH 7.4, in a humidified atmosphere of 95% air plus 5% CO 2 at 37 °C. The anti-proliferative activity of PTS1 and LLO-PTS1 proteins in different concentrations was measured by the MTT assay. Cell viability of each well was calculated as [A 570 of treated cells/A 570 of control untreated cells] × 100%.

Disulfide bond prediction
The structure of the PTS1 was extracted from 1PRT.pdb file deposited in the PDB database. After an energy minimization step, using the MODIP program and Disulfide By Design, the best disulfide bridges were analyzed, and one of which was selected to introduce into the PTS1 gene. The distance of the disulfide bond partners from each other in the linear sequence of the protein, ΔG, and dihidral bond is among the parameters concerned with the selection of the best disulfide bond. The potential disulfide bond selected in the present study was Phe53-Asn197 with the numbering of the amino acids based on the first methionine of the PTS1 signal sequence (Fig. 3).

Fusion gene and recombinant construct preparation
Designed PTS1 gene was chemically synthesized and delivered in the pUC57 cloning vector. PCR of the gene using primers harbouring 5' arms complementary to the LLO gene fragment and the pPSG-IBA35 vector backbone was carried out. In addition, the amplification of the coding sequence of the first 241 amino acids of LLO toxin was performed with designed primers. Quick-change PCR with amplified PTS1 gene was performed, lanes A and C in Figure 4, to produce recombinant construct pPSG-PTS1. Moreover, the SOEing PCR led to the the megaprimer of about 1500 bp, as illustrated in Figure 5 (lane D). Quick-change PCR using this megaprimer was conducted to produce the recombinant construct pPSG-LLO-PTS1.

Expression and purification
Recombinant N-terminal His-tagged PTS1 protein and LLO-PTS1 fusion protein were produced in BL21 DE3 and SHuffle T7 strains of E. coli in parallel using 0.1 mM concentration of IPTG, as an inducer. The purity and size of the recombinant proteins produced in BL21 DE3 and SHuffle T7 strains were the same (lanes D and F; Fig. 4). Yield of the total purified protein was 23.2 mg.L -1 and 13.1 mg.L -1 for the PTS1 protein and 19.4 and 17.4 mg.L -1 for the LLO-PTS1 fusion protein in the two strains, respectively, after the overnight induction of the recombinant protein expression.

Disulfide bond determination of the recombinant proteins
The SH groups titration in denaturing condition indicated that the disulfide bonds of the recombinant mutant PTS1 protein purified from SHuffle T7 strain have been formed with high efficiency. The ratio of the free SH groups for each molecule was calculated, which proved to be 0.21 and less than 0.05 for proteins purified from BL21 DE3 and SHuffle T7 strains, respectively. Hence, the efficiency of disulfide bond formation in the two proteins was approximately 80% and over 94%, respectively. The same results were procured for the LLO-PTS1 protein.

Cell culture and MTT assay
Serial dilutions of recombinant PTS1 protein and LLO-PTS1 fusion protein were applied to the cultured MCF-7 cells in RPMI 1640 media supplemented with 10% FBS. As shown in Figure 6, the purified recombinant PTS1 protein was not cytotoxic to MCF-7 cell line. The 5-fluorouracil was also tested as the positive control.

DISCUSSION
E. coli has become the most common host for the recombinant protein expression. It is used readily and is economical. In recent years, SHuffle T7 strain of E. coli has been introduced for the cytoplasmic expression of recombinant proteins having disulfide bonds. It benefits from two main features: oxidative environment of cytoplasm and the presence of DsbC chaperone in the cytoplasm [23] . A few recombinant proteins have successfully been produced in SHuffle T7 cytoplasm [24,25] . In the present study, recombinant PTS1 and its fusion form, LLO-PTS1, were produced in the SHuffle strain. Because the proteins produced in the present study were dialyzed in a buffer free of any oxidant and reducing agent after elution from the affinity column, it seems that the more oxidant environment of SHuffle T7 cytoplasm can be responsible for the higher efficiency of disulfide bond formation in the protein obtained from this strain.
Chemically inactivated toxoid vaccines possess several adverse effects, thereby demanding for the use of alternative strategies to improve safety. Efficacy of the aP vaccines in generating a Th2, instead of Th1, immune response against pertussis antigens is also the matter of controversy [26][27][28][29] . Subunit vaccines are attracting interest due to the lack of native toxin problems. Using new subunit vaccines has been discussed elsewhere [30] . Higher stability of the subunit vaccines in the formulations prepared for use in populations in comparison with the live vaccines or even high-cost acellular vaccines is another advantage of these substances [31] .
Lee and colleagues [32] have found that the fusion of the S1, S3, and filamentous haemagglutinin can be expressed as a single recombinant protein. In addition, the produced recombinant antigen was immunogenic in mice, but the antibody titers against the pertussis toxin subunits showed to be weak. In another study, recombinant Mycobacterium bovis BCG was utilized as a host for the production of PTS1 protein. This mode of administration provoked a good cellular immunity in a mouse model, but the level of humoral immunity against the pertussis toxin antigen, as the main virulence factor of pertussis, was low [33] . Diphtheria, tetanus, and pertussis toxin molecule subunits have already been expressed in tobacco and carrot in a soluble form. It has been indicated that these plant-derived proteins could elicit a specific antibody response in animal models [34] . Although there are several advantages in large-scale protein production in plants over the conventional biofactories like the animal cell lines or yeast and bacteria cells, there are yet some challenges in the industrial extraction and purification steps. Recombinant pertussis toxin subunits have also been expressed and purified separately to form oligomers in vitro. They bear the potential to form in vitro assemblies, which are antigenic, and the antibodies they provoke have been displayed to have inhibitory activities against pertussis toxin [35] .
Herein, we have produced a new form of genetically inactivated PTS1 using a cloning strategy, which successfully produced two forms of inactivated PTS1 toxin (PTS1 and fusion LLO-PTS1 proteins) in E. coli BL21 DE3 and SHuffle T7 strains. However, the rate of the disulfide bond in SHuffle T7 strain showed to be more efficient. The analysis of the safety of the proteins produced against a cell line model revealed that both recombinant proteins are safe and non-toxic for the human cell culture model. Our results demonstrated that the recombinant proteins PTS1 and LLO-PTS1 can be expressed in SHuffle T7 strain cytoplasm and purified in a soluble form with disulfide bonds with high efficiency. Overall, the LLO-PTS1 fusion protein can be formulated as a new generation of the adjuvants to be used as a novel immunestimulator in the next research studies of pertussis disease.

ACKNOWLEDGMENTS
This study is a part of the Ph.D. dissertation of Hossein Forghani and financially supported by Iran University of Medical Sciences, Tehran, Iran (grant no. 25818) and Shahid Sadoughi University of Medical Sciences, Yazd, Iran. The LLO-PTS1, a novel fusion protein, was registered as a patent in the https://iripo.ssaa.ir/.