TC-1 cells, generated by the co-transformation of C57BL/6 mouse lung epithelial cells with HPV16 E6 and E7 oncogenes and the human ras oncogene (21), were purchased from the Cancer Hospital, Chinese Academy of Medical Sciences (ATCC no. CRL-2785). TC-1 cells were cultured in RPMI-1640 medium (HyClone, Logan, UT, USA), containing 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA), 100 U/ml penicillin, and 100 U/ml streptomycin (Beyotime, Shanghai, China). R10 medium, used for culturing BMDCs, contains standard RPMI-1640 plus 50 nM β-mercaptoethanol (Biosharp, Hefei, China). All cells were cultured at 37°C in humidified air supplemented with 5% CO₂.

Specific pathogen-free (SPF) 6- to −8-week-old female C57BL/6 mice were maintained at the Laboratory Animal Center of Wuhan Institute of Virology, Chinese Academy of Sciences (CAS). All mouse studies were performed according to Regulations of the Administration of Affairs Concerning Experimental Animals in China (WIVA25201304), and the protocols were reviewed and approved by the Laboratory Animal Care and Use Committee at the Wuhan Institute of Virology, CAS. Briefly, all the mice were fed in an independent ventilated cage (IVC) and the IVCs were kept within an experimental animal barrier environment. The food was sterilized by Co⁶⁰-irradiation and the water was sterilized using an autoclave. We replenished the food and water twice or three times a week.

mE7 was generated from the oncogenic E7 gene of HPV16 (GenBank accession no. AF125673) synthesized at Sangon Biotech (Shanghai, China) and a linker encoding 15-amino acids containing two N-X-S/T motifs (GGGGSNGTGGGNASC) with two potential N-linked glycosylation sites. In contrast to mE7, E7 was generated from the site-directed mutagenesis of the two potential N-linked glycosylation sites in the linker region. NGT and NAS were conservatively mutated to AGT and AAS, respectively.

The mE7 and E7 were cloned into plasmid pPICZαA. Subsequently, recombinant plasmids pPICZαA-mE7 and pPICZαA-E7 were linearized using SacI (NEB, Ipswich, MA, USA), purified, transfected into electrocompetent Pichia pastoris KM71 (pulsed 1.5 kV, 200 Ω, and 25 µF) and subsequently incubated for 2 h in yeast extract peptone dextrose medium (YPD). Recombinant clones were selected on YPD agar plates, containing 75 µg/ml Zeocin (Invitrogen, San Diego, CA, USA), and identified using colony PCR. The expression and purification of recombinant protein in KM71 was conducted according to Lam et al (15).

BMDCs were generated according to Lutz et al (22). Briefly, bone marrow cells from the tibiae and femurs of one 8-week-old C57BL/6 mouse were cultured in R10 medium supplemented with 200 U/ml rmGM-CSF (PeproTech, Rocky Hill, NJ, USA). The cells were fed with fresh medium on days 3, 6, and 8 and matured with fresh R10 medium containing 100 U/ml rmGM-CSF and 500 U/ml TNF-α (PeproTech) on day 10.

The mature BMDCs were cultured in 6-cm plates and co-incubated with 10 µg/ml mE7 or E7 in the presence or absence of 10 mg/ml yeast mannans (Sigma, St. Louis, MO, USA) for 45 min. Subsequently, the cells were collected and washed using PBS. The collected cells were lysed using RIPA lysis buffer (Beyotime), and centrifuged at 10,000 × g for 5 min to remove the debris. The supernatants were precipitated with 4 volumes of acetone, incubated at −20°C overnight, centrifuged at 13,000 × g for 15 min and subsequently analyzed by western blotting.

T cell polarization assays were performed following the method of Satchidanandam et al (23). Briefly, the mE7- or E7-preincubated BMDCs were cultured for 24 h and then treated with 50 µg/ml mitomycin C (Sigma) for 1 h followed by extensive washing. The above BMDCs were co-cultured in a 96-well plate with one million splenocytes from C57BL/6 mice immunized with mE7, E7 or PBS, at the BMDC:splenocyte ratio of 1:10 for 72 h. The supernatants were collected for IFN-γ measurement using capture ELISA (BioLegend, San Diego, CA, USA), according to the protocol for Mouse IFN-γ ELISA MAX Standard Sets.

The 6- to 8-week-old C57BL/6 mice were grouped and subcutaneously (s.c.) vaccinated with 40 µg mE7, 40 µg E7 or 100 µl PBS in the presence of 5 µg adjuvant MPL (Sigma, L6895) on days 0, 5 and 10. On day 15, the mice were sacrificed by cervical dislocation under aseptic conditions. Eyeball enucleation was conducted for blood collection. The sera and splenocytes were collected and harvested from blood.

Four million splenocytes from each immunized mouse were cultured for 48 h in 24-well plates in the presence of 10 µg/ml E7 or phorbol-12-myristate-13-acetate (PMA) plus ionomycin (Dakewe Biotech Co., Shenzhen, China). The supernatants were collected, and the concentrations of IFN-γ, IL-2 and TNF-α were measured using capture ELISA (BioLegend), according to the protocols for Mouse IFN-γ/IL-2/TNF-α ELISA MAX Standard Sets.

The sera were collected from each immunized mouse at 5 days after the last immunization to determine specific IgG using ELISA. Polyvinylchloride 96-well plates (Nunc, Rochester, NY, USA) were coated with E7 at 2 µg/ml in PBS and incubated overnight at 4°C. The plates were subsequently blocked with 10% FBS in PBS, washed, and incubated with the two-fold diluted serums, followed by detection with HRP-conjugated Affinipure Goat Anti-Mouse IgG (Proteintech, Wuhan, China).

One million splenocytes from each immunized mouse and PBMCs from each group were cultured for 24 h in 96-well plates in the presence of 10 µg/ml E7 or PMA plus ionomycin as positive control. For surface staining, cells were incubated with the following antibodies (BD Biosciences, Franklin Lakes, NJ, USA): PE-Cy7 anti-CD3e (145-2C11), FITC anti-CD4 (RM4-5), PE anti-CD8a (53-6.7) and 7-AAD Viability Staining Solution (BioLegend). For intracellular IFN-γ staining, APC anti-IFN-γ (XMG1.2, BD Biosciences) was used. All the staining procedures were according to BD Biosciences protocols. Data were acquired using a BD FACS flow cytometer and analyzed by FlowJo software (Tree Star, Ashland, OR, USA).

Sixteen million splenocytes from each immunized mouse were cocultured with one million mitomycin C-pretreated TC-1 cells in RPMI-1640 supplemented with 10% FBS, 50 µg/ml concanavalin A (Con A; Sigma) and 20 U/ml IL-2 (PeproTech) at 37°C in 5% CO₂. After 5 days, the viable splenocytes were collected and used as effector cells, and TC-1 cells were used as target cells. The Pierce LDH Cytotoxicity assay kit (Thermo Fisher Scientific, Waltham, MA, USA) was used to measure the effector cells against TC-1 at ratios of 2.5:1 and 5:1 according to the manufacturer's instructions. Specific lysis was calculated as percent specific lysis = [(Experimental value - Effector cell spontaneous control - Target cell spontaneous control)/(Target cell maximum control - Target cell spontaneous control)] ×100.

On day 0, the 6- to 8-week-old C57BL/6 mice were injected (s.c.) with 2×10⁵ TC-1 cells in the right flank. Vaccination with mE7, E7 and PBS with MPL was performed on days 3, 8, and 13. Tumor growth was measured using a caliper, and tumor size was calculated as volume = (length × width²)/2 (24). For survival analysis, the mice were considered as dead when the tumor reached 1000 mm³. When the tumor size of the immunized mice exceeded 1000 mm³, the mice were sacrificed and dissected. Tumors were separated by ophthalmic scissors and tweezers, and kept in formalin. Tumor-free mice vaccinated with mE7 were challenged with 2×10⁵ TC-1 cells again on day 31. The percentage of tumor-free mice was recorded.

Statistical analysis was performed using Prism software version 6.0 (GraphPad, San Diego, CA, USA). All data are presented as the means ± SD. The cytokine concentration, antibody titers and FACS results were assessed using an unpaired two-tailed t-test. Cytotoxicity assays and tumor sizes were assessed using two-way ANOVA multiple comparisons. Survival percentages were assessed using the log-rank test. P-values of <0.05 were considered to indicate statistically significant differences.

As previously reported, Pichia pastoris has been widely used to express mannosylated recombinant proteins (25). To express a mannosylated recombinant E7 (defined as mE7), a 15-amino acid linker containing two N-X-S/T motifs representing potential N-linked glycosylation sites was fused to the C-terminal of HPV16 E7. As a control, E7 was subjected to the site-directed mutagenesis of the two potential N-linked glycosylation sites in the linker region of mE7. NGT and NAS were conservatively mutated to AGT and AAS, respectively (Fig. 1A).

The mE7 and E7 recombinant proteins were efficiently expressed by Pichia pastoris KM71 in secreted forms, and mE7 showed a molecular weight between 22 and 40 kDa, while E7 showed a molecular weight of ~22 kDa (Fig. 1B), indicating the considerable glycosylation of mE7 in KM71 and the unglycosylation of E7, since the unglycosylated E7 expressed from BL21 (DE3) cells was 22 kDa (data not shown). The optimum time for mE7 expression (72 h) was longer than that for E7 (36 h).

To detect whether mannosylated proteins are more efficiently taken up by DCs via MR-mediated endocytosis, BMDCs were incubated with equal mE7 or E7 in the presence or absence of yeast mannans, which block MR and DC-SIGN (26). BMDCs endocytosed more mE7 than E7, and yeast mannans inhibited the uptake of mE7, while E7 was not apparently affected (Fig. 2A). As shown in Table I, the relative amount of mE7 ranged from 2.25 (without mannans) to 1.02 (with mannans) while that of E7 ranged from 1.38 (without mannans) to 1.25 (with mannans). This result indicates that MR or DC-SIGN plays a dominant role in the uptake of mannosylated proteins by BMDCs, and mannosylation can promote the uptake of mE7 by BMDCs, which is crucial to activate downstream immune responses. Subsequently, we co-cultured antigen-uptake BMDCs with splenocytes of immunized mice for 72 h. As a results, mE7-uptake BMDCs induced more IFN-γ secretion by splenocytes than E7, while the levels of IFN-γ were reduced with mannans, respectively (Fig. 2B). Taken together, mannosylation enhances the ability of BMDCs to take up mE7 and increases T cell polarization.

We measured the E7-specific antibody titers in the sera of immunized mice using ELISA. The results showed that vaccination with mE7 developed higher antibody titer than E7 (Fig. 3A).

To assess the level of cytokines generated by vaccination with mE7, the splenocytes from vaccinated mice were harvested and stimulated with E7 or a positive control PMA plus ionomycin. Subsequently, the supernatants were collected and measured using ELISA. Under equal doses of mE7 and E7, vaccination with mE7 induced more cytokines, including IFN-γ, IL-2 and TNF-α, than vaccination with E7 (Fig. 3B-D). Compared with E7, mE7 induced almost six times more IFN-γ (Fig. 3B), which is an important activator of macrophages, inducer of MHC II expression and indicator of Th1 responses. Compared with E7, mE7 also induced approximately twice as much IL-2 (Fig. 3C), which stimulates the differentiation of regulatory T cells, promotes T cell growth and NK cytolytic activity, and mediates activation-induced cell death (27). In addition, compared with E7, mE7 induced approximately twice as much TNF-α (Fig. 3D), which induces fever, apoptotic cell death, cachexia, inflammation and inhibition of tumorigenesis. Moreover, the E7-specific IFN-γ-secreting CD8⁺ T cells in the spleen and PBMCs were analyzed by flow cytometry (Fig. 3E and F). Compared with E7, mE7 induced more E7-specific IFN-γ-secreting CD8⁺ T cell within the spleen and PBMCs. These results suggested that vaccination with mE7 generated stronger Th1 responses which are crucial for tumor regression and considerable E7-specific antibody responses.

The above results that vaccination with mE7 promoted Th1 responses (Fig. 3B-F) represents potent E7-specific T cell responses. To examine whether vaccination with mE7 primes more effective E7-specific cytotoxic CD8⁺ T cells than E7, the splenocytes of vaccinated mice were collected and cocultured with mitomycin C-pretreated TC-1 cells for 5 days. Viable effector cells were measured for cytotoxic activity against TC-1 cells. As shown in Fig. 4, the effector cells from mice immunized with mE7 showed stronger cytolytic effects on TC-1 cells (65.4%) than those from mice immunized with E7 (19.8%) and PBS (27.9%) when the effector:target ratio was 5:1.

To assess the antitumor activity generated using the mE7 vaccine, we performed an in vivo tumor challenge assay with HPV16 E6 and E7-expressing TC-1 cells. C57BL/6 mice were injected with 2×10⁵ TC-1 cells in the right flank. After 3 days, tumor-challenged mice (5 per group) were vaccinated with three doses of mE7, E7 or PBS in the presence of MPL every 5 days. Tumor growth was measured using a caliper every 2 or 3 days. As shown in Fig. 5A, the mice vaccinated with mE7 showed lower average tumor sizes than mice vaccinated with E7 and PBS. All mice vaccinated with mE7 eliminated tumors at 26 days after tumor challenge, while only 20% of the mice vaccinated with E7 eliminated tumors at more than 37 days after tumor challenge. All mice vaccinated with PBS failed to inhibit tumor growth and were sacrificed after 40 days, when the tumor sizes reached 1000 mm³. Additionally, all mice vaccinated with mE7 survived after 51 days. In contrast, 40% of the mice vaccinated with E7 and 100% of the mice vaccinated with PBS died after 51 days (Fig. 5B and C). After 31 days, the mice vaccinated with mE7 were rechallenged with 2×10⁵ TC-1 cells, and no mice developed tumors until 51 days (Fig. 5C), indicating immunological memory. At 40 and 51 days, the mice vaccinated with PBS, mE7 and E7 were sacrificed and dissected, and 5 PBS tumors and 3 E7 tumors were observed (Fig. 5D). Taken together, mE7 represents an effective therapeutic vaccine against TC-1 tumors.