Recombinant HPV16 E7 assembled into particles induces an immune response and specific tumour protection administered without adjuvant in an animal model

Freshly streaked bacterial colonies, containing the E7 plasmid [30], were inoculated in 25 ml LB medium (DIFCO) and grown to saturation overnight (O/N) at 37°C. The culture was then inoculated in 500 ml LB, and grown until the culture density reached OD₆₀₀ = 0.6. The His-E7 protein was induced by the addition of 1 mM IPTG (A.G. Scientific, Inc) for 3 h. The culture was harvested and centrifuged for 30 min in a Sorval centrifuge at 6000 rpm in GSA rotor. The bacterial pellet was lysed for 30 min in a rotator at room temperature (R/T) in a denaturing buffer (40 ml) containing 8 M urea (MP Biomedicals, Inc), 10 mM NaH₂PO₄, 10 mM Tris-HCl pH 8, 300 mM NaCl, 1 mM DTT, (Sigma-Aldrich), 1% Triton-X 114 (Sigma-Aldrich) and 1% Triton X-100 (Buffer B mod). To break the DNA, the lysate was sonicated for 60 min in the pulsed mode (50% on/off pulse; effective sonication time, 30 min) using an ultrasonic processor (Vibra-Cell 400, Sonics). The lysate was clarified in a Sorval centrifuge for 20 min at 10.000 rpm in a SS34 rotor. The supernatant was incubated for 30 min with 4 ml of 50% slurry NiNTA resin (QIAGEN) at RT. To reduce the endotoxin content, the E7-NiNTA agarose suspension was collected in a 50 ml tube, extensively washed in batch and spun down in a centrifuge at 500 × g. The E7-NiNTA was sequentially washed in Buffer B (pH 8, without detergents) containing 10% glycerol (100 ml), 20% ethanol (100 ml) and 60% isopropanol (200 ml). The isopropanol washes were alternated with cold 10 mM Tris-HCl washes (200 ml) [34]. The last sequential washes were performed using 500 ml Buffer C (8 M urea, 10 mM NaH₂PO₄, 10 mM Tris-HCl pH 6.3). The protein was eluted by gravity-flow in several 2 ml fractions from packed E7-Ni-NTA using 1 M Imidazole (Sigma-Aldrich) in Buffer B. After an analytical Coomassie stained SDS-PAGE, the fractions containing E7 were collected and the protein was subjected to 2 step-dialysis at 4°C in native buffers. The first step was performed in 2 L of buffer containing 25 mM Tris, 50 mM NaCl pH 7.5 (TN) in presence of 1 mM DTT and the second step was performed in 2 L of TN buffer only. E7 was concentrated in a centrifugal filter device up to a final concentration of 2 mg/ml. All the reagents were ultrapure grade. The E7 protein yield was 20 mg/l of medium culture. The protein was quantified by standard methods (Protein BC assay, BIORAD); its purity and identity were monitored by SDS-PAGE followed by Coomassie brilliant blue staining and western blotting (30). The endotoxin contamination was as low as 0.5 EU/mg protein as monitored by LAL assay (QCL-1000, Lonza). The presence of E7 particles was monitored by negative stain EM.

Protein samples were separated in 12.5% polyacrylamide gels in Leammli Tris-Glycine buffer and blotted into an Immobilon-P membrane. In a non-reducing gel, the protein samples were denatured in SDS-loading buffer [30] without β-mercapto-ethanol. The protein was identified by Western blot using both commercial monoclonal and in-house prepared polyclonal anti-E7 antibodies [30]. A peroxidase-conjugate rabbit anti-mouse IgG (H+L) (Sigma-Aldrich) was used as secondary antibody. The immune complexes were revealed with a chemiluminescence substrate (PIERCE).

10 μl samples of the E7 preparation (2 mg/ml) were adsorbed for 1 min onto Formvar-coated copper grids, then rinsed briefly with water and negatively stained with 2% filtered aqueous sodium phosphotungstate adjusted to pH 7.0. Negatively stained preparations were observed with a Philips 208S transmission electron microscope at 80 kV.

Three samples of different E7 preparations and, as a control, three samples of Glutathione-S-transferase (GST) were analysed for their content of ⁶⁶Zn and ⁶⁸Zn analytical masses. The GST protein was produced in pGEX-2T transformed E. coli and purified by glutathione affinity chromatography (PIERCE). Measurements were performed by means of High Resolution Inductively Coupled Plasma-Mass Spectrometry (HR-ICP-MS), using an Element2 apparatus (Thermo-Finnigan, Bremen, Germany). HR-ICP-MS is a well established and powerful analytical technique for the determination of trace and ultra-trace elements in biological samples. The calibration of the method was performed by the adoption of the standard addition mode: diluted single-element standards were added to the analytical solutions. To compensate for instrumental drifts and matrix effects, indium was added to each sample as an internal standard.

6-8 week-old female C57BL/6 mice were purchased from Charles River Laboratories and maintained under pathogen-free conditions for one week before the experiment. The animal care and the experiments followed the European Directive 86/609 EEC. The protocol of animal use was evaluated by the Service for Biotechnology Animal Welfare of the Istituto Superiore di Sanità, and approved by the Italian Ministry of Health. Three groups of mice (14 per group) were inoculated subcutaneously with 1, 2 or 3 doses of 10 μg E7 respectively, at 1 week intervals. A fourth mouse group was inoculated with a saline solution and used as a control (naïve). Two weeks after the last immunization, 4 mice of each group were sacrificed to analyse the immune response and 10 mice were inoculated subcutaneously with 1 × 10⁵ TC-1 cells/mouse, as described [35]. The TC-1 cells were grown in complete medium with 0.4 mg/ml G418. Cells at 50% confluence were harvested, counted and rinsed in Hank's medium at 1 × 10⁶cells/ml for the injection in mice. Tumour growth was monitored by visual inspection and palpation once a week for 2 months. The experiment was performed twice.

Splenocytes from mice of the same immunization group were pooled and enriched in CD4⁺ and CD8⁺ cells using the Dynal Mouse T cell Negative isolation kit (Invitrogen). Cells were cultured in RPMI 1640 (Lonza) supplemented with 10% FCS, 1% penicillin/streptomycin, 2 mM glutamine, 1 mM pyruvate and 1% non-essential amino acids (Lonza) (complete RPMI). To assess cell proliferation, the splenocyte pools (2 × 10⁵ cells/well, in triplicate) were stimulated for five days in the presence of 5 μg/ml of two 8- and 9-mer E7 peptides, DLYCYEQL (aa 21-28) and RAHYNIVTF (aa 49-57), already known to efficiently bind the H-2 K^b complex of C57 Black/6 mice [36]. On day 6, the cells were pulsed with 0.5 μCi [³H] thymidine per well and incubated for 18 h. The cells were then harvested onto filters using an automatic harvester and counted in a Beta Counter (Wallac). The results were expressed in stimulation index (SI), calculated by dividing the mean counts per minute (cpm) of cells exposed to the E7 peptides by the mean cpm of cells incubated only with medium. The IFN-γ ELISPOT assay was performed using commercially available reagents (Mabtech AB). T-cell enriched splenocytes were seeded in triplicate (5 × 10⁵ cells per well) in 200 μl complete medium with the E7 stimulator peptides. After 18 h at 37°C in a humidified 5% CO₂ incubator, the plates were analysed for the presence of IFN-γ as described in [35].

The sera from each group of immunized mice were pooled and analysed. To determine the anti-E7 specific IgG titre the sera pools were serially diluted (two-fold) and assayed by ELISA [30]. The end-point dilution corresponded to an OD absorbance < 0.1 at 450 nm. Sera pools diluted 1:100 were used to analyse the anti-E7 IgM, IgA and the IgG isotypes (IgG1, IgG2b, IgG2c and IgG3). Antigen-antibody complexes were detected using the following HRP-secondary antibodies (Sigma-Aldrich): rabbit anti-mouse IgG (H+L), goat anti-mouse IgM (μ-chain), goat anti-mouse IgA (α-specific), goat anti-mouse IgG1, IgG2b, IgG3, IgG2c. HRP activity was revealed using tetramethyl benzidine substrate (TMB) in the presence of H₂O₂. After 30 min at RT, the enzymatic reaction was stopped by adding 50 μl of 1 M sulphuric acid/well. Washing steps were done with 400 μl/well of PBS containing 0.05% Tween-20 in an automatic washer.

Significance analysis was performed using the Student t test for unpaired data. Differences were considered significant if P < 0.05.

The E7 protein was expressed in E. coli with a [His]₆ tag and purified via a one-step denaturing protocol until a high level of homogeneity and low endotoxin content were achieved [30, 34]. The protein was prepared in a soluble suspension state by dialysis in Tris buffer and then analysed by western blotting in reducing and non-reducing SDS-PAGE. In the reducing gel, the E7 protein appears in monomeric form (Figure 1, lane 1). In non-reducing gels, based on the analysis of the molecular mass marker, E7 appears in forms consistent with the mass of monomer, dimer, trimer, tetramer, octamer and higher oligomers, suggesting that, in these conditions, the E7 monomer is the oligomerization unit (lane 2).

Preparations of purified E7 were analysed by negative staining EM. Figure 2 shows representative EM micrographs of the E7 preparation samples. The protein appears dispersed on the grid as aggregates of different shape and size (panel A). The protein appears clustered in compact-looking spheroidal microaggregates, the majority ranging between 100 and 200 nm in size (panel B). In these same samples, E7 also appears assembled in structured particles that seem to derive from the aggregation of smaller particles (panel C). These particles resemble the previously described E7 oligomers [27]. A semi-quantitative analysis by EM counts of micro and nano-sized particles, ranging between 45-200 nm, indicates that E7-aggregates are in the order of 10⁵ particles/ml (not shown). The E7 preparations were also subjected to EM immunolabelling but neither commercial anti-E7 monoclonal nor in-house prepared polyclonal antibodies [30] revealed any significant reaction, suggesting that these antibodies were unsuitable for the EM observation of E7-particles (data not shown).

HPV16 E7 contains a Zinc finger-like domain that binds the metal even when the protein is expressed in E. coli[10, 11]. Since the E7 purification protocol used does not employ either chelating agents or Zinc salts, several E7 samples were analysed for Zn⁺⁺ content by high resolution mass spectrometry (HR-ICP-MS). The recombinant GST protein was used as a control. The Zn⁺⁺ ion concentration in the E7 preparations was 1.13 μg/g ± 0.10 while GST showed only trace level of Zn⁺⁺ (0.19 μg/g ± 0.02). The metal/protein ratio was calculated to be 0.19, therefore only about 19% of the E7 molecules were bound to Zn⁺⁺.

To investigate if this E. coli-derived HPV16-E7 preparation, administrated without adjuvant, was able to induce a tumour-protective immunity, groups of mice were inoculated with 10 μg of protein per mouse, 1, 2 or 3 times, at one week intervals. As a control group, mice were inoculated with a saline solution (naïve group). Two weeks after the last immunization, some of the animals were bled and killed to analyse the immune response, in vitro. The remaining animals were challenged with the TC-1 tumour cells and the inhibition of tumour growth in these mice was monitored for 2 months.

To quantify the humoral immune response and to compare the results obtained from several animal groups, the sera from the animals of each group were pooled, then sequentially diluted to determine the antibody titres by end-point dilution in an E7-based ELISA [30]. The anti-E7 IgG titre was 1:200 after a single immunization and progressively increased in mice immunized 2 and 3 times, reaching 1:8000 and 1:16000 respectively. The presence of IgM and IgA was analysed in comparison with the IgG and the results are shown in Figure 3. In panel A, the anti-E7 specific antibodies of mice immunised 1, 2 or 3 times either with E7 or a saline solution are shown. The sera show an increase of anti-E7 specific IgG already after the second protein dose; IgMs were detected only after the third E7-dose, while IgAs were never detectable. Animals inoculated with 3 doses of saline solution did not show any E7 specific antibody response (naïve, panel A).

The therapeutic effector functions of antibodies depend on their class and subclasses [37]. In order to better evaluate the E7-specific humoral immune response, the anti-E7 specific IgG1, IgG2b, IgG2c (IgG2a) and IgG3 antibody subclasses were also determined. The results, shown in Figure 3, panel B, show that the IgG2b level was significant after the second immunization while the level of IgG2c was significant only after the third immunization. The level of IgG1 was not significant and IgG3s were undetectable. This anti-E7 IgG isotype profile indicates that the immune response induced in vaccinated mice is a mixed Th1/Th2 type.

To analyse the induction of the cell-mediated immune response in mice after 1, 2 or 3 doses of the E7 preparation, T-enriched splenocytes from mice of the same immunization group were stimulated in vitro, with the E7-specific CTL peptides and processed for T cell proliferation and γ-IFN ELISPOT assays. Splenocytes from naïve mice were pulsed with an unrelated mixture of peptides used as control. The results are shown in Figure 4. Splenocytes from mice immunised with 2 and 3 doses of E7 showed a high Stimulation Index (SI) suggesting that specific T clone selection occurred after E7 peptide stimulation. Splenocytes from naïve mice and from mice immunised once showed non-significant SI. Conversely, in the γ-IFN ELISPOT assay (panel B) only the splenocytes of mice that received 3 E7 doses, stimulated with the E7-specific CTL peptides, showed a significant level of E7-specific γ-IFN producing cells (panel B, 3).

To evaluate the efficacy of the E7 preparation as inductor of anti-tumour immunity, the mice immunised with 1, 2 or 3 doses of E7 were challenged with the TC-1 tumour cells, and the tumour growth was monitored for two months after the challenge. The results are shown in Figure 5. Mice vaccinated with three doses of E7 particles were fully protected from tumour growth. Only 40% of the mice immunised with 2 doses of E7 were tumour free, whereas the mice immunised with 1 dose and the naïve mice developed a palpable tumour within 4 weeks of tumour-monitoring, after the challenge with TC-1 cells.