Human papillomavirus L1 protein expressed in Escherichia coli self-assembles into virus-like particles that are highly immunogenic

Virus Research 220 (2016) 97–103

Contents lists available at ScienceDirect

Virus Research journal homepage: www.elsevier.com/locate/virusres

Human papillomavirus L1 protein expressed in Escherichia coli self-assembles into virus-like particles that are highly immunogenic Yumei Chen a,b , Yunchao Liu b , Gaiping Zhang a,b,c,∗ , Aiping Wang d,∗ , Ziming Dong a,∗ , Yanhua Qi d , Jucai Wang c , Baolei Zhao c , Ning Li d , Min Jiang d a

College of Basic Medical Science, Zhengzhou University, Zhengzhou 450001, China Henan Provincial Key Laboratory of Animal Immunology, Henan Academy of Agricultural Sciences, Zhengzhou 450002, China c College of Animal Science and Veterinary Medicine, Henan Agricultural University, Zhengzhou 450002, China d School of Life Sciences, Zhengzhou University, Zhengzhou 450001, China b

a r t i c l e

i n f o

Article history: Received 17 February 2016 Received in revised form 17 April 2016 Accepted 19 April 2016 Available online 20 April 2016 Keywords: Human papillomavirus Bacterial expression Virus-like particles Splenic lymphocytes Neutralizing antibody

a b s t r a c t HPV vaccines based on L1 virus-like particles (VLPs) provided a high degree of protection against HPVs infection. In this study, the codon optimized HPV16 L1 gene were sub-cloned into five procaryotic expression vectors (pET-28a, pET-32a, pGEX-4T-2, pE-sumo and pHSIE), and fused with different protein tags. No recombinant proteins were expressed in pET-28a-L1 and pHSIE-L1, and the proteins expressed by pET-32a-L1 plasmid with TRX-tag were in the form of inclusion body. Only SUMO-tagged and GSTtagged L1 proteins expressed by pE-Sumo-L1 or pGEX-4T-L1 were soluble. The yield of SUMO-L1 protein reached 260 mg/L fermentation medium in shake flask. After SUMO tags were eliminated, a 90% purity of L1 proteins was generated by ion-exchange and Ni-NTA affinity chromatography. The purified HPV16 L1 protein self-assembled into virus-like particles (VLPs) and showed a haemagglutination activity. High titers specific and neutralizing antibodies were detected in HPV 16 L1VLPs vaccinated mice. Cytokines such as IFN-␥ and IL-2 showed significant higher in VLPs vaccinated mice compared with negative control (p < 0.05, p = 0.055). Thus, the expression of recombinant HPV16 L1 VLPs in Escherichia coli was feasible, which could potentially be used for a VLP-based HPV vaccine. © 2016 Published by Elsevier B.V.

1. Introduction Human papillomavirus (HPV) is an epithelial cell specific virus, which is correlated with human pathogens (de Villiers, 2013). Until now, more than 200 subtypes of HPV have been identified. Some of them classified as high-risk HPV types are related to cancers (especially cervical cancer), while low-risk types are widely detected in sexually transmitted disease worldwide, such as genital wart cases in both man and woman (Anhang et al., 2004). HPV 16, a high-risk type, has been detected in about 54% of malignant cells of cervical cancers (Munoz et al., 2003; Tommasino, 2014). Recent studies have shown that HPV16 may associate with other epithelial cancers, such as nasopharyngeal carcinoma and head and neck cancer (Maxwell et al., 2015). HPV is a non-enveloped, double stranded circular DNA virus with a virion particle of 55–60 nm in diameter, which present as

∗ Corresponding authors at. College of Basic Medical Science, Zhengzhou University, Zhengzhou 450001, China. E-mail addresses: [email protected] (G. Zhang), [email protected] (A. Wang), [email protected] (Z. Dong). http://dx.doi.org/10.1016/j.virusres.2016.04.017 0168-1702/© 2016 Published by Elsevier B.V.

T = 7 icosahedral symmetry (Munger et al., 2004). The capsid of HPV contains 72 copies of pentamers which consisted by 5 L1 proteins, and finally arrange in an icosahedral particle with L2 protein together (Buck et al., 2008). The HPV L1 protein can self-assemble into virus-like particles (VLPs), which exhibited morphologically and immunologically features like native virions (Buck et al., 2013). Since the absent of infectious genome, the VLP could not infect and lead to disease. Therefore, HPV VLPs have been approved by FDA (USA Food and Drug Administration) as safe and effective vaccine for cervical cancers and other related diseases control (McKee et al., 2015; Shaw, 2013). Although yeast and baculovirus expression systems can produce HPV16 VLPs with high immunogenicity (Hanumantha Rao et al., 2011; Zhang et al., 2010), some defects, such as big technical difficulties, high cost and low yield, limit the widely use of these systems, especially in poor countries and regions. Therefore, looking for a new and more efficient approach to prepare HPV L1 protein is very important to reduce the cost of HPV vaccine and expand the availability of the vaccine. Bacterial expression system might be a reasonable choice to solve this problem, due to its high level on protein expression and rapid growth in relatively inexpensive culture medium (Zhang et al., 2010). How-

98

Y. Chen et al. / Virus Research 220 (2016) 97–103

Table 1 Primers used in this article. Primers

Sequences (5 –3 )a

P1 P2 P3 P4 P5

TTGAATTCATGTCTCTGTGGCTGCCGTCTGAG AATGCGGCCGCTTACAGTTTACGTTTTTTACG TTGGTCTCTAGGTATGTCTCTGTGGCTGCCG AATCTCGAGTTACAGTTTACGTTTTTTACG TTGGATCCATGTCTCTGTGGCTGCCGTCTGAG

a The sequence underline were the restriction enzyme sites used for inserting amplified fragments into the expression vector.

ever, the production of HPV L1 protein in bacteria also has a few technical matters, like mainly in existed as inclusion body, low immunogenicity, etc. (Chen et al., 2001; Ma et al., 2007). Fusion protein approach has been shown to be an effective method to solve the protein inclusion body in Escherichia coli expression system. Proteins, such as thioredoxin (TRX), glutathione-stransferase (GST), maltose binding protein (MBP) and small ubiquitin like modifier protein (SUMO), are commonly used to promote the correct folding and stability of recombinant protein in E. coli (Xu et al., 2014; Zhang et al., 2015). In this study, we produced HPV L1 protein from E. coli expression system and VLPs were self-assembled. The VLPs shown equivalent biological activity and immunogenicity as generated from eukaryotic expression system. The results indicated the VLPs generated from E. coli felicitously present the antigen epitopes of HPV 16 capsid and have a potential as a cost-effective vaccine candidate for HPV prevention. 2. Materials and methods 2.1. Construction of HPV16 L1 expression plasmid

a flask containing fresh LB broth (1 l) and cultured under 37 ◦ C with shaking (220 rpm). Until the optical absorbance OD600 reached 1.0, 0.3 mM IPTG was used to induce target protein for 20 h at 20 ◦ C. After centrifugation, cell pellets were harvested and disrupted by high pressure homogenizer. SP SepharoseTM Fast Flow column (GE, China) was equilibrated with PB buffer containing 0.2 M NaCl. The clarified cell lysates were pumped onto the columns at a flow rate of 0.5 ml/min. After washing with 20 beds of PB buffer (0.4 M NaCl), SUMO-L1 proteins were eluted by PB buffer with 0.4–1.0 M NaCl. All the fractions were collected and analyzed by SDS-PAGE and Western Blot. Then, the SUMO tag was eliminated by incubating the purified product with SUMO protease (Solarbio, Shanghai) at 30 ◦ C for 4 h. Then, all the samples were loaded onto a Ni-NTA Sepharose FF (GE, China) and the flow through samples were collected. Besides, endotoxin was cleared away by using the affinity matrix of modified polymyxin B (PMB) (GenScript, China) and determined by LAL based colorimetric assay (GenScript, China). The concentration of purified fraction was calculated using Micro BCATM protein assay kit (Thermo Scientific, USA) following the manufacturer’s protocol. 2.3. Assembly and characterization of HPV 16 VLPs After dialyzing against PB buffer containing 0.5 M NaCl for 24 h, HPV 16 L1 protein was placed onto a 200 mesh carbon-coated copper grid for negatively stained with 1% phosphotungstic acid (PTA). Observation was carried out by using a transmission electron microscope (TEM, JEM-1400) at 80 KV. Size distribution of HPV VLPs was analyzed by dynamic light scattering (DLS) and a computerized inspection system (Malvern, UK). Hemagglutination (HA) and hemagglutination inhibition (HAI) assays were performed as previously described (Bazan et al., 2009).

The gene encoding HPV16 L1 protein (GenBank No. AACO9292) was optimized according to the codon usage in E. coli, allowing some deviations and synthesized by GenScript (Nanjing, China) (Grote et al., 2005; Wu et al., 2007). As shown in Fig. 1, five primers (Table 1) were synthesized to subclone L1 gene into procaryotic expression vectors. Briefly, Primer P1 and P2 were used to subclone the L1 gene into the pHSIE vector downstream of intein coding sequence (Wang et al., 2012). Primers P3 and P4 were used to generate a SUMO fusion protein and the P4 and P5 primer were used to insert the L1 encoding gene into pET-28a, pET-32a and pGEX-4T-2, respectively. The recombinant plasmids, named as pE-SUMO-L1, pHSIE-L1, pET-32a-L1, pET-28a-L1 and pGEX-4T-L1, were transformed into E. coli BL21 (DE3), respectively and further confirmed by restriction analysis and sequencing. Single positive colony was selected and inoculated to Luria–Bertani (LB) medium with 50 ␮g/ml kanamycin or 80 ␮g/ml ampicillin. With shaking at 37 ◦ C until OD600 reached 0.6, the culture were induced by adding 0.5 mM isopropyl-␤-d-1-thiogalactopyranoside (IPTG) and incubated at 20 ◦ C for 12 h. Then, cells were collected by centrifugation at 12000g and resuspended in 50 mM phosphate buffer (PB, pH 7.0). All the cells were lysed by ultrasonic (pulses of 5 s with 10 s off during 8 min) at 30% intensity in iced water bath and clarified by centrifugation at 12000g for 20 min. The supernatant and precipitation fractions were analyzed, respectively, by SDS-PAGE and Western Blot (using anti-HIS6 tag monoclonal antibody) as previously described (Xu et al., 2014).

The antiserum titer was detected by using indirect ELISA (Xu et al., 2014). ELISA plates were coated with 1 ␮g of HPV L1 VLPs per well in 0.1 M CBS at 4 ◦ C overnight. After blocked with 5% nonfat milk, serum samples were added to the plate and incubated at 37 ◦ C for 1 h. HRP-conjugated goat anti-mouse IgG diluted in 5% non-fat milk (1:5000) was incubated in the plate at 37 ◦ C for 30 min followed by six washes with PBST. The colorimetric reaction was performed by incubating all plates with 70 ␮l substrate solutions for 8 min at room temperature. Then, after stopping by 70 ␮l of 2 N H2 SO4 , absorbance was read at 450 nm.

2.2. Expression and purification of HPV16 L1 protein

2.6. Pseudovirus neutralization assay

The single clone of E. coli BL21(DE3) carrying plasmid pESumoL1 was picked from LB agar plate supplemented with 100 ␮g/ml ampicillin, and cultured overnight in LB broth (10 ml) at 37 ◦ C. The overnight starter cultures (10 ml) were then inoculated into

HPV16 pseudovirus was produced according to previous reports (Buck et al., 2004; Day et al., 2013). A plasmid encoding green fluorescent protein (GFP) was packaged in the pseudovirus as a reporter gene. Pseudovirus neutralization assays were performed

2.4. Immunization of animals Animals care and study procedure were following the guideline of the Animal Research Ethics Board of Zhengzhou University. HPV16 L1 VLPs were adsorbed on alhydrogel by mixing them for 30 min at 25 ◦ C. Four-week-old female Kunming mice were randomly assigned to three groups each consisting of four mice. One group was subcutaneously inoculated with PBS as a negative control; the other two groups were immunized with 5 ␮g or 20 ␮g VLPs mixture, respectively. The boost was performed at 14 days after the first vaccination. Blood samples were collected at 0, 7, 14, 21, 28, 37 and 60 days post first immunization and sera were prepared for detection. 2.5. ELISA for antibody detection

Y. Chen et al. / Virus Research 220 (2016) 97–103

99

Fig. 1. The schematic graph of constructed plasmids. The five expression vectors have been constructed based on the vector backbones of pHSIE, pE-Sumo, pET-28a, pET-32a and pGEX-4T-2. His6 tag has been attached to the N-terminal of the fusion protein except the plasmid pGEX-4T-L1 has a GST tag. The stop codon (TAA) has been added in front of the Not I or Xho I restriction enzyme site. * The Bsa I site of enzyme cleavage was disappeared after the plasmid being constructed.

Fig. 2. The expression and solubility of fusion proteins were analyzed on 12% SDS-PAGE gel. (a) M, protein molecular weight markers. E. coli cell lysates, Lane 1, pHSIE-L1; lane 2, pE-SUMO-L1; lane 3, pET-28a-L1; lane 4, pET-32a-L1; lane 5, pGEX-4T-L1; lane 6, BL21 (DE3) stain without plasmid transformed. (b) The solubility of fusion protein. Lane 1, 2, 3, the insoluble lysate fraction of E. coli containing plasmid pE-SUMO-L1, pET-32a-L1 and pGEX-4T-L1, respectively; lane 4, 5, 6, the soluble lysate fraction of E. coli containing plasmid pE-Sumo-L1, pET-32a-L1 and pGEX-4T-L1, respectively. (c) The optimization of fermentation in shake flask (pE-SUMO-L1). Lane 1, the soluble lysate fraction; lane 2, the insoluble lysate fraction. (d) The purified HPV 16 L1 protein after eliminating SUMO tag. Lane 1, the fraction of wash buffer, lane 2, the purified HPV L1 protein; Western Blot analysis HPV 16 L1 protein with anti-HPV16 L1 monoclonal antibody.

as described before with few modification (Buck et al., 2004; Zhang et al., 2010). The 293FT cells were incubated in 48-well plate with a density of 1.5 × 104 cells/well at 37 ◦ C for 6 h. Diluted mice sera was mixed with 50 ␮l of HPV16 pseudovirus in DMEM (containing 100 TCID50 ) in triplicate, and pseudovirus mixed with DMEM media were prepared as control. After incubated with 293FT cells at 37 ◦ C for 1 h, the mixtures were placed on 293FT cells and incubated at 37 ◦ C for 72 h. Cells were harvested and detected by flow cytometry (FCM, BD, USA). The percent of infection inhibition was estimated as described previously (Zhang et al., 2010).

(200 meshes). Lymphocyte separation medium (Solarbio, Shanghai) was used to isolate the splenic lymphocytes as instruction manual described. The splenic lymphocytes were plated in 96-well plates at a density of 5 × 104 cells/well in DMEM supplemented with 10% FBS and stimulated with 10 ␮g/ml sterile HPV VLPs. The plates were incubated at 37 ◦ C in a 5% CO2 atmosphere for 60 h. Levels of IFN-␥, IL-2, IL-4 were measured by using commercially available ELISA kits (BD, USA).

2.7. Cytokines detection

2.8. Statistics

After 60 days of the first immunization, the mice were killed by cervical dislocation and the spleens were isolated as previously described (Guo et al., 2012). Briefly, the spleens were harvested under sterile conditions, and filtered with a sterile nylon sieve

Data were presented as the means ± standard deviation of three or more independent experiments. The statistical significance was assessed by two-tailed Student’s t-test by SPSS13.0 software (p < 0.05).

100

Y. Chen et al. / Virus Research 220 (2016) 97–103

Fig. 3. Characterization of assembled HPV 16 L1 VLPs. (a) Transmission electron microscope (TEM) images of HPV VLPs, VLPs were stained with 2% phosphomolybdic acid, grids were air-dried and examination with a JEM-1400 operated at 80 kV. The scale bar represents 200 nm. (b) Dynamic light scattering (DLS) detects the size distribution of VLPs. (c) HPV L1 VLPs agglutinate mouse erythrocytes. 2-fold serial diluted samples were incubated 1% mouse erythrocytes for 3 h at 4 ◦ C and photographed. 1, 2, 3, 4, 5 represents sample from 200 ng to 12.5 ng per well. HA, HPV L1 VLPs; NC, PBS control.

3. Results 3.1. Expression and purification of HPV L1 protein As shown in Fig. 1, the codon optimized HPV L1 gene (GenBank No. KT948784) was inserted into five vectors. HPV L1 fused with SUMO, GST and TRX tag were effectively expressed in E. coli, while His6 -SUMO-Intein-L1 and His6 −L1 proteins were undetected (Fig. 2a). Among these expressed proteins, TRX-L1 was expressed as inclusion body, while GST-L1 and SUMO-L1 were mainly expressed in soluble forms (Fig. 2b). As shown in Fig. 2, the expression levels of soluble GST-L1 and SUMO-L1 fusion proteins accounted for approximately 15% and 20% of the totals supernatant proteins, respectively. Results from Western Blot were further confirmed the soluble expression of GST-L1 and SUMO-L1 in E. coli (Supplementary Fig. S1). After a preliminary optimization in shake flasks, the yield of SUMO-L1 protein reached about 260 mg/L fermentation medium (Fig. 2c). The soluble SUMO-L1 protein was purified by loading the clarified cell lysates on SP column, and eluted with gradient of 0.4–1.5 M sodium chloride in PB as described before. The majority of the SUMO-L1 protein was eluted in fractions from 0.6 M to 0.8 M sodium chloride in PB (Supplementary Fig. S2a). After removed SUMO tags, the target protein was presented as a distinct 55 kD and which is consistent with theoretical calculated molecular weight of HPV16 L1 (Supplementary Fig. S2b). Results from SDS-PAGE and Western Blot showed that the purity of HPV L1 was approximately 90% (Fig. 2d). The content of endotoxin was 20.1 EU/mg in purified HPV L1 protein. The final yield of purified HPV L1 protein was about 56 mg/L of the fermentation medium in shake flasks. 3.2. Assembly and characterization of HPV VLPs To further analyze the morphology of HPV L1 protein, electron microscopy and DLS analyses were performed. After dialyzed against PB containing 0.5 M NaCl, the samples were applied onto carbon-coated grids and visualized by TEM. The recombinant HPV16 L1 proteins self-assembled into some structures of 50–60 nm in diameter, which were similar to the size of HPV virions (Fig. 3a). Same result was also detected by DLS (Fig. 3b). In

Fig. 4. HPV L1 specific immune responses in mice. Groups of mice (n = 4) were immunized with 5 ␮g or 20 ␮g HPV L1 VLPs, and PBS as control. The specific IgG antibody titers of serum from 60 days post inactivation mice were determined by end point dilution VLP-ELISA (Sera diluted from 1:100 to 1:51200). In control mouse, anti-HPV 16 L1 VLP specific IgG titers were less than 100. Data were expressed as the mean ± SE of values, and statistical significance among groups was compared.

addition, HA test showed that the HPV VLPs have HA activity on mouse erythrocytes (Fig. 3c).

3.3. Level of specific antibody To evaluate the immunogenicity of HPV VLPs obtained from E. coli, Kunming mice were immunized with HPV VLPs adsorbed with alum adjuvant. Results from direct ELISA showed that the titer of specific antibody against HPV VLPs increased as time and two sharp increases were detected at boost and at 37 days post inoculation (Supplementary Fig. S3). The antibody titers were significant different (P < 0.05) between both VLPs inoculated groups and PBS group. High dose group showed relatively higher titers of specific antibodies than low dose group (p = 0.194, Fig. 4).

Y. Chen et al. / Virus Research 220 (2016) 97–103

101

Fig. 5. Neutralisation assays of HPV 16 pseudovirus by sera from vaccinated mice. Three groups of mice (n = 4) were vaccinated with 5 ␮g, 20 ␮g HPV L1 VLPs, or PBS as control on 0 day and 14 day, and bled until 60 days. (a) HPV 16 pseudovirions contains reporter plasmid was incubated with diluted serum (1: 3000) from vaccinated mice at 4 ◦ C for 1 h. The mixture was add into monolayer 239FT cells and photographed by fluorescence microscope after 48 h. PBS, sera from mice vaccinated with PBS; Control, pseudovirions incubated with PBS. (b) 293FT cells were detected by FCS, percent infection inhibitions were calculated and expressed as the mean ± SE (n = 4). The statistical significance among groups was compared. (c) Hemagglutination inhibition assay (HAI). 5 ␮g, 20 ␮g, and PBS represent sera from 5 ␮g, 20 ␮g VLPs and PBS vaccinated mouse; 1– 5, seras diluted from 1: 200 to 1:3200.

Fig. 6. Immune related cytokines of mice were quantitatively measured using spleniclymphocytes. Three groups of mice (n = 4) were vaccinated with 5 ␮g, 20 ␮g HPV L1 VLPs, and PBS on 0 day and 14 day. Spleens were harvested 60 days after first immunization. The splenic lymphocytes were isolated and then stimulated with 10 ␮g/ml VLP. The concentrations of cytokines in the culturesupernatants were calculated by sandwich ELISA. 5 ␮g, 20 ␮g, the dose of VLPs vaccinated mice; PBS, PBS immunized mice. Data were shown as the mean ± SE of values. Statistical significance among groups was compared.

3.4. Level of neutralizing antibody in mice HPV16 pseudovirus neutralization assays were performed as described before. Specific neutralizing antibodies were detected in sera from HPV VLPs immunized mice, but not in mice injected with PBS (Fig. 5a). The results from FCM showed significant difference in sera percentage inhibition rate between VLPs immunized and non-immunized groups (P < 0.05, Fig. 5b). However, no difference was found between high and low dose groups (p = 0.575). Besides,

HA induced by HPV pseudovirus was inhibited by sera from VLPs inoculated mice (Fig. 5c). 3.5. Cytokine production As showed in Fig. 6, the production of IFN-␥ and IL-2 was significantly higher in both immunized groups compared with the PBS control. However, no significant differences were detected in the production of IL-4 among these three groups.

102

Y. Chen et al. / Virus Research 220 (2016) 97–103

4. Discussion HPVs were detected in more than 99% of cervical cancers, and 50% of them had relationship with HPV16 infection (Hallez et al., 2004; Tommasino, 2014). Data from the International Agency for Research on Cancer (IARC) showed that more than 288,000 women are killed by cervical cancer worldwide each year and at least 80% of cervical cancer deaths occur in developing countries (Giuliano et al., 2011), which make the appearance of an efficiently vaccine become more urgently. Many kinds of viral capsid proteins generated from eukaryotic expression systems could effectively self-assembly, such as hepatitis B virus, HIV gag protein, and HPV L1 protein (Garcea and Gissmann, 2004). But the cost and technical difficulties of those expression systems are relatively higher. For both research and commercial purposes, the expression of recombinant protein in bacteria is more simple and economical compared with in eukaryotic expression systems. However, there are also challenges in expression of foreign protein in bacterial, such as low yield, mis-folding, low immunogenicity and insolubility. In addition, the codon usage frequency and RNA secondary structures of foreign gene could also affect the yield of heterologous protein expressed in E. coli (Burgess-Brown et al., 2008; Kim and Lee, 2006). Therefore, we optimized the HPV 16 L1 gene according to the codon usage frequencies of E. coli without any change in amino acid, and rationalized mRNA structure to avoid the rho-independent transcription terminators as well as ribosome binding sites. The solubility and properly folding of heterologous protein expressed in E. coli could be affected by many factors, such as the fusion tags, molecular chaperonin, the synthesis rate of target protein, and so on. In this study, different tags were constructed in five plasmids and utilized to enhance the folding and solubility of recombinant HPV 16 L1 proteins in E. coli. However, as shown in Fig. 2, His6 -SUMO-Intein-L1 and HIS6 -L1 were not expressed and His6 -TRX tag fusion L1 was expressed as inclusion body. Only HPV16 L1 protein with a SUMO-tag or a GST-tag was detected in the soluble fraction of cell lysate, but the yield of GST fusant was relatively lower than SUMO fusant. In additon, the soluble HPV L1 protein fusion with GST-tag was also described before (Chen et al., 2001). This result indicated that HPV L1 protein expression in E. coli was affected by fusion tags, and SUMO-tag can also improve the solubility of L1 protein effectively as GST-tag does. After optimization of fermentation conditions for SUMO-L1 production, the expression levels of soluble SUMO-L1 fusion protein reached approximately 260 mg/L fermentation medium. Considering the high rate of protein expression and cheap medium, E. coli system exhibits a significant cost advantage in preparation of HPV16 L1 protein. Considering the fusion protein may interfere with the selfassembly of HPV L1 protein, the SUMO-tag was eliminated to generate a L1 protein without any extra amino acid sequences. The yield of the purified protein was about 56 mg/L, which was significantly higher than the yield of HPV from yeast and insect (Silvia Boschi et al., 2009; Zheng et al., 2004). The L1 protein with a native N-terminus self-assembled into HPV16 VLPs and this fact was confirmed by TEM and DLS (Fig. 3). It exhibited as aspherical particle with approximately 55 nm in diameter, which was similar as HPV16 VLPs generated from eukaryotic expression systems. The HA activity of HPV VLPs depended on the appropriate conformational epitopes (Roden et al., 1995). Results from HA test demonstrated that the L1 VLPs generated from E. coli possessed appropriate spatial conformation. These results further confirmed that the SUMO-tag could effectively improve the properly folding of HPV L1 protein. Animal experiments were designed to evaluate the value of recombinant L1 VLPs as a vaccine candidate against HPV infection. The result of direct ELISA suggested that the HPV L1 VLPs could

effectively induce a specific humoral response (Fig. 4). Since HPVs cannot be grown in tissue culture, infectious HPV pseudoviruses, encapsidating a reporter GFP plasmid, were developed for neutralization assays. Significant infection inhibition by sera generated from L1 VLPs immunized mice was detected by fluorescence microscope and FCM (Fig. 5). Simultaneously, HA induced by HPV 16 pseudoviruses was inhibited by sera from immunized mice. The results suggested the recombinant L1 VLPs was able to induce an effective and specific antibody response and also a significantly viral neutralization, which suggest that the epitopes on the recombinant L1 VLPs generated from E. coli were properly presented and were able to protect against HPV infection by eliciting humoral immune response. Besides humoral immune response, the antigenspecific cellular immune response also plays an important role in conferring the protection against challenge viruses of HPV. Results from ELISA showed that L1 VLPs immunized groups elicited a significantly higher expression of antigen-specific IFN-␥ and IL-2 in splenic lymphocytes than control group (Fig. 6). But the production of IL-4, which was secreted by Th2 cells, had no difference. These results indicated that the HPV L1 VLPs from E. coli induced an effective cellular immune response. In this study, we established a high-efficient recombinant E. coli expression system for the production of HPV 16 L1 protein, which showed a great potential as a candidate for HPV vaccine. The generated L1 proteins correctly self-assembled into HPV VLPs, which was able to efficiently elicit both humoral and cellular immunities against HPV infection and made E. coli expressed L1 VLPs a potential HPV vaccine candidate for industrial purpose. The works in this study indicated the prospect and advantages of E. coli in production of HPV L1 protein, which may also help to reduce the cost of L1 protein production. Competing interests The authors declare that they have no competing interests. Acknowledgments We are grateful to Prof. John T. Schiller for providing plasmid (pSheLL 16) to produce HPV16 pseudovirus and Prof. Zhou Enmin providing the vector pHSIE. This study was supported by the grant from program of the National Nature Science Foundation of China (No.31472177), the China Postdoctoral Science Foundation (No.2013M541980), Major Program of Science and Technology in Henan (No.141100110100) and Henan Province Scientific and technological innovation talent program in Colleges (No.14HASTIT027). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.virusres.2016.04. 017. References Anhang, R., Goodman, A., Goldie, S.J., 2004. HPV communication: review of existing research and recommendations for patient education. CA Cancer J. Clin. 54, 248–259. Bazan, S.B., de Alencar Muniz Chaves, A., Aires, K.A., Cianciarullo, A.M., Garcea, R.L., Ho, P.L., 2009. Expression and characterization of HPV-16 L1 capsid protein in Pichia pastoris. Arch. Virol. 154, 1609–1617. Buck, C.B., Pastrana, D.V., Lowy, D.R., Schiller, J.T., 2004. Efficient intracellular assembly of papillomaviral vectors. J. Virol. 78, 751–757. Buck, C.B., Cheng, N., Thompson, C.D., Lowy, D.R., Steven, A.C., Schiller, J.T., Trus, B.L., 2008. Arrangement of L2 within the papillomavirus capsid. J. Virol. 82, 5190–5197. Buck, C.B., Day, P.M., Trus, B.L., 2013. The papillomavirus major capsid protein L1. Virology 445, 169–174.

Y. Chen et al. / Virus Research 220 (2016) 97–103 Burgess-Brown, N.A., Sharma, S., Sobott, F., Loenarz, C., Oppermann, U., Gileadi, O., 2008. Codon optimization can improve expression of human genes in Escherichia coli: a multi-gene study. Protein Expr. Purif. 59, 94–102. Chen, X.S., Casini, G., Harrison, S.C., Garcea, R.L., 2001. Papillomavirus capsid protein expression in Escherichia coli: purification and assembly of HPV11 and HPV16 L1. J. Mol. Biol. 307, 173–182. Day, P.M., Thompson, C.D., Schowalter, R.M., Lowy, D.R., Schiller, J.T., 2013. Identification of a role for the trans-golgi network in human papillomavirus 16 pseudovirus infection. J. Virol. 87, 3862–3870. de Villiers, E.M., 2013. Cross-roads in the classification of papillomaviruses. Virology 445, 2–10. Garcea, R.L., Gissmann, L., 2004. Virus-like particles as vaccines and vessels for the delivery of small molecules. Curr. Opin. Biotechnol. 15, 513–517. Giuliano, A.R., Palefsky, J.M., Goldstone, S., Moreira Jr, E.D., Penny, M.E., Aranda, C., Vardas, E., Moi, H., Jessen, H., Hillman, R., 2011. Efficacy of quadrivalent HPV vaccine against HPV infection and disease in males. N. Engl. J. Med. 364, 401–411. Grote, A., Hiller, K., Scheer, M., Münch, R., Nörtemann, B., Hempel, D.C., Jahn, D., 2005. JCat: a novel tool to adapt codon usage of a target gene to its potential expression host. Nucleic Acids Res. 33, W526–W531. Guo, H.C., Feng, X.M., Sun, S.Q., Wei, Y.Q., Sun, D.H., Liu, X.T., Liu, Z.X., Luo, J.X., Hong, Y., 2012. Immunization of mice by hollow mesoporous silica nanoparticles as carriers of porcine circovirus type 2 ORF2 protein. Virol. J. 9, 1–10. Hallez, S., Simon, P., Maudoux, F., Doyen, J., Noel, J.C., Beliard, A., Capelle, X., Buxant, F., Fayt, I., Lagrost, A.C., Hubert, P., Gerday, C., Burny, A., Boniver, J., Foidart, J.M., Delvenne, P., Jacobs, N., 2004. Phase I/II trial of immunogenicity of a human papillomavirus (HPV) type 16 E7 protein-based vaccine in women with oncogenic HPV-positive cervical intraepithelial neoplasia. Cancer Immunol. Immunother. 53, 642–650. Hanumantha Rao, N., Baji Babu, P., Rajendra, L., Sriraman, R., Pang, Y.Y., Schiller, J.T., Srinivasan, V.A., 2011. Expression of codon optimized major capsid protein (L1) of human papillomavirus type 16 and 18 in Pichia pastoris; purification and characterization of the virus-like particles. Vaccine 29, 7326–7334. Kim, S., Lee, S.B., 2006. Rare codon clusters at 5’-end influence heterologous expression of archaeal gene in Escherichia coli. Protein Expr. Purif. 50, 49–57. Ma, Z., Chen, B., Zhang, F., Yu, M., Liu, T., Liu, L., 2007. Increasing the expression levels of papillomavirus major capsid protein in Escherichia coli by N-terminal deletion. Protein Expr. Purif. 56, 72–79. Maxwell, J.H., Grandis, J.R., Ferris, R.L., 2015. HPV-Associated head and neck cancer: unique features of epidemiology and clinical management. Annu. Rev. Med. 67, 91–101.

103

McKee, S.J., Bergot, A.S., Leggatt, G.R., 2015. Recent progress in vaccination against human papillomavirus-mediated cervical cancer. Rev. Med. Virol. 25, 54–71. Munger, K., Baldwin, A., Edwards, K.M., Hayakawa, H., Nguyen, C.L., Owens, M., Grace, M., Huh, K., 2004. Mechanisms of human papillomavirus-induced oncogenesis. J. Virol. 78, 11451–11460. Munoz, N., Bosch, F.X., de Sanjose, S., Herrero, R., Castellsague, X., Shah, K.V., Snijders, P.J., Meijer, C.J., 2003. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N. Engl. J. Med. 348, 518–527. Roden, R.B., Hubbert, N.L., Kirnbauer, R., Breitburd, F., Lowy, D.R., Schiller, J.T., 1995. Papillomavirus L1 capsids agglutinate mouse erythrocytes through a proteinaceous receptor. J. Virol. 69, 5147–5151. Shaw, A.R., 2013. Human papillomavirus vaccines six years after approval. Annu. Rev. Med. 64, 91–100. Silvia Boschi, B., Agtha, D.A.M.C., Karina Araújo, A., Aurora Marques, C., Garcea, R.L., Paulo Lee, H., 2009. Expression and characterization of HPV-16 L1 capsid protein in Pichia pastoris. Arch. Virol. 154, 1609–1617. Tommasino, M., 2014. The human papillomavirus family and its role in carcinogenesis. Semin. Cancer Biol. 26, 13–21. Wang, Z., Li, N., Wang, Y., Wu, Y., Mu, T., Zheng, Y., Huang, L., Fang, X., 2012. Ubiquitin-intein and SUMO2-intein fusion systems for enhanced protein production and purification. Protein Expr. Purif. 82, 174–178. Wu, G., Dress, L., Freeland, S.J., 2007. Optimal encoding rules for synthetic genes: the need for a community effort. Mol. Syst. Biol. 3, 134. Xu, J., Guo, H.C., Wei, Y.Q., Dong, H., Han, S.C., Ao, D., Sun, D.H., Wang, H.M., Cao, S.Z., Sun, S.Q., 2014. Self-assembly of virus-like particles of canine parvovirus capsid protein expressed from Escherichia coli and application as virus-like particle vaccine. Appl. Microbiol. Biotechnol. 98, 3529–3538. Zhang, T., Xu, Y., Qiao, L., Wang, Y., Wu, X., Fan, D., Peng, Q., Xu, X., 2010. Trivalent human papillomavirus (HPV) VLP vaccine covering HPV type 58 can elicit high level of humoral immunity but also induce immune interference among component types. Vaccine 28, 3479–3487. Zhang, J., Lv, X., Xu, R., Tao, X., Dong, Y., Sun, A., Wei, D., 2015. Soluble expression rapid purification, and characterization of human interleukin-24 (IL-24) using a MBP-SUMO dual fusion system in Escherichia coli. Appl. Microbiol. Biotechnol. 99, 6705–6713. Zheng, J., Ma, J., Yang, X.F., Liu, H.L., Cheng, H.W., Si, L.S., Wang, Y.L., 2004. Highly efficient and economical baculovirus expression system for preparing human papillomavirus type16 virus-like particle. Acta Biochim. Biophys. Sin. 36, 548–552.