Development and evaluation of a novel E7 multi-epitopic vaccine for human papillomavirus type 16: design, expression, purification, and immunological characterization

Abstract

Background

Persistent infection with high-risk Human papillomavirus (HPV), specifically HPV-16, is the leading cause of cervical cancer. Although preventative vaccines have shown significant efficacy in preventing HPV infection, cervical cancer is a significant public health issue that affects millions of women worldwide. Modern therapeutic approaches, such as peptide vaccines, could be promising and have potential for the treatment of the HPV-infected population.

Methods

A HPV16-E7 multi-epitopic vaccine (MEVE7) was designed to comprise potent CD4 + and CD8 + T cell epitopes and optimally expressed in a prokaryotic expression system. Polyclonal antibodies were generated, and their reactivity with immunizing antigen and native protein in E7 expressing cells (TC-1) was assessed by ELISA and immunofluorescent staining, respectively. The efficacy of the vaccine was assessed in a therapeutic animal model of HPV-induced cancer.

Results

Our study revealed that the final construct was successfully expressed in E. coli BL21 (DE3)-gold within 4 h of induction as inclusion bodies. Among the tested solubilization buffers, the buffer with a pH of 12 and containing 2 M urea showed the highest solubilization effect. Polyclonal antibodies directed against the E7 multi-epitope vaccine were able to react strongly with the immunizing antigen and E7-bearing cells (TC-1). Immunization of TC-1 tumor-bearing mice with HPV16-E7, markedly delayed tumor growth and propagation.

Conclusion

The poly-epitope vaccine for HPV16-E7, as expressed and purified in this research, is highly immunogenic and capable of triggering E7-specific antibodies, making it a potential therapeutic HPV vaccine. Further research is needed to optimize the vaccination schedule and assess the E7-specific immune cell profile.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-024-10343-x.

Introduction

Cervical cancer is a significant public health issue that affects millions of women worldwide. Persistent infection with high-risk Human papillomavirus (HPV), specifically HPV-16, is the leading cause of cervical cancer, and it is crucial to have effective treatment options in combating this disease [1, 2].

To prevent HPV viral infection, countries worldwide have introduced prophylactic vaccines such as Gardasil and Cervarix. While these vaccines effectively prevent new HPV infections and reduce the morbidity rate associated with cervical cancers, these vaccines lack therapeutic efficacy against pre-existing HPV infections. They are not recommended for women already diagnosed with HPV or HPV-mediated cervical cancer. Consequently, cervical cancer remains a high-fatality cancer among women [3].

In response to this challenge, contemporary research has delved into innovative therapeutic methods, notably vaccines, to address the HPV-infected population [4]. Numerous HPV vaccine formulations, including peptide and protein, live vector (bacterial and viral), DNA/RNA, and dendritic cell, have been generated to combat cervical cancer [5]. Among these, peptide-based vaccines are a promising approach for developing therapeutic HPV vaccines. Peptide vaccines are safe, easily prepared with high purity, and transportable. They are free from infectious components, minimize secondary adverse reactions, and exhibit chemical stability [6]. By utilizing short peptides that bind to major histocompatibility complexes (MHC), these peptides can be identified by T-cells as antigens specific to the tumor. Using peptides derived from HPV antigens as a vaccine represents a promising strategy for immunotherapy against HPV infections [7, 8]. The development of peptide vaccines depends on identifying and selecting tumor-reactive T-cell peptide epitopes, eliciting responses from both antitumor CD8 + T cells (cytotoxic) and CD4 + T cells (helper). Therefore, a comprehensive understanding of the HPV genome and the progression from persistent HPV infection to cervical lesions is imperative for the advancement of these therapeutic strategies.

Therapeutic peptide vaccines, while advantageous, suffer from poor immunogenicity, necessitating additional methods to enhance their efficacy [9]. One approach involves utilizing a poly-epitope construct, wherein specific peptides are designed to enhance both innate and adaptive immunity, resulting in increased CD8 + T cell responses. This strategy involves safeguard the peptides from protease degradation, ensuring their sustained effectiveness in targeting the immune system. To address these limitations, researchers are exploring innovative techniques, such as incorporating lipids and other adjuvants, to augment the potency of therapeutic peptide vaccines and improve their overall immunogenicity [3, 4].

The HPV genome codes nine proteins, two structural proteins (specifically L1 and L2), and seven early proteins (E1, E4, E2, E6, E7, E5, and E8), which play crucial roles in viral replication and infectivity. Among them, E6 and E7 proteins possess the unique ability to transform infected cells [10]. Their expression leads to the inhibition of essential tumor suppressor genes, namely p53 and pRb, promoting increased cell proliferation and contributing to the development of cervical tumors and precancerous lesions [5, 6].

Given the frequent expression of these viral proteins in HPV-infected cells, they represent promising targets for developing immunotherapeutic strategies against HPV infection and associated cancers [7]. Previous research has shown that invasive cervical malignancies and high-grade dysplasia are primarily induced by E7, with E6 playing a supporting role in sustaining and enhancing malignancy. Moreover, E7 is more abundantly expressed and highly conserved compared to E6. Consequently, the E7 antigen emerges as a potent candidate for designing and developing therapeutic and protective vaccines [1, 8, 11].

In this study, we were about to design an E7 multi-epitope vaccine, optimize its expression in a prokaryotic expression system, characterize the generated recombinant protein, and assess its efficacy as a potential therapeutic vaccine in a mouse model of HPV-induced cancer.

Methods and materials

Epitope prediction

The prediction of MHC class I (H-2D^b and H-2Kb) and II (H2-AIb) epitopes was conducted using the Immune Epitope Database and Analysis Resource (IEDB) tools, available at http://tools.iedb.org. The consensus amino acid sequence derived from our human study [12] was employed to predict peptides within the mentioned alleles. Non-overlapping peptides with the highest ranks were selected, with each epitope being nine amino acids in length for MHC class I binding and 16 amino acids in length for MHC-II binding analysis.

To evaluate the capacity of the selected epitopes for interferon-gamma induction, the IFN epitope server (https://crdd.osdd.net/raghava/ifnepitope/predict.php) was utilized. The prediction relied on the Support Vector Machine (SVM) algorithm and the IFN gamma versus non-IFN gamma model employed in our study.

Additionally, an assessment of the potential allergenicity of the selected peptides, intended for incorporation into the vaccine, was conducted. The PA3P server (http://lpa.saogabriel.unipampa.edu.br:8080/pa3p/pa3p/pa3p.jsp) was employed to calculate allergenicity based on Allergen online and AFDS-motif.

The peptides were further scrutinized as antibody epitopes, a pivotal aspect of vaccine design aiming to activate B lymphocytes and provoke a humoral immune response. To predict linear B-cell epitopes, we employed the Kolaskar & Tongaonkar Antigenicity model (Kolaskar & Tongaonkar, 1990), accessible at http://tools.iedb.org/bcell/.

Epitope selection

Experimentally confirmed CD8 + and CD4 + epitopes were systematically compiled from the Immune Epitope Database (IEDB) available at http://www.iedb.org. For CD8 + epitopes, the search criteria included human papillomavirus 16 (ID: 333760) as the organism, E7 as the antigen, in linear form, positive in T-cell assays conducted in mice, and involvement in any disease. Similarly, for CD4 + epitopes, the search parameters encompassed human papillomavirus 16 (ID: 333760) as the organism, E7 as the antigen, in linear form, positive results in T-cell assays in mice, and association with any disease. This meticulous curation process ensured the collection of relevant, pertinent epitopes for the study.

Vaccine design

In constructing a recombinant poly-epitopic vaccine comprising multiple epitopes, those epitopes that have shown significant impact on cervical cancer based on previous studies [7, 13–16], along with epitopes predicted to have high scores, were carefully selected. A novel multi-epitope construct was then designed, incorporating these chosen epitopes and specific spacers. The Jcat tool (http://www.prodoric.de/JCat) was applied to translate the protein sequence into a DNA sequence. Subsequently, Vector NTI software was used for optimizing DNA sequence according to codon usage in E. coli. In the subsequent step, XhoI and NcoI restriction sites were introduced at the3’ and 5’ends, respectively. The desired DNA encoding the multi-epitope construct was synthesized in a pGH vector, having a length of 1043 bp (Generay, China). This precise and systematic approach ensured the development of an optimized and functional multi-epitope vaccine construct.

Construction of expression vector

To construct a recombinant vector expressing HPV16-E7 multi-epitopic vaccine (MEVE7), the designed sequence and the pET28a vector (Takara, Japan) were treated with XhoI and NcoI restriction enzymes. The resulting digested fragments were purified from a 1% (w/v) agarose gel using a gel extraction kit (Yektatajhiz, Iran). The gel-purified MEVE7 fragment was then ligated into the pET28a vector in the presence of T4 DNA ligase (Thermo Fisher Scientific, US) to create the pET-28a-MEVE7 construct. The design of this construct was aimed at allowing the expression of C-terminal His-tagged recombinant protein, facilitating the processes of purification and identification.

To confirm the integrity of the pET28a-MEVE7 plasmid, colony PCR and DNA sequencing were performed. For the PCR reaction, a single colony was mixed in a PCR premix (Takara, Japan) to amplify the MEVE7 fragment using the T7 promoter and T7 terminator universal primers. A thermal cycler was applied to perform PCR reaction Followed by 30 cycles, each comprising 30 s at 94 °C, 40 s at 50 °C, and 1 min at 72 °C. Following these cycles, a final extension step of 10 min at 72 °C was performed. Subsequently, the PCR product was subjected to electrophoresis on a 1% agarose gel, and the size of the polymerized fragment was determined using a 1 kb DNA ladder, ensuring the successful construction of the pET-28a-MEVE7 plasmid.

The sequencing of the pET28a vector with HPV16-MEVE7 was carried out using T7 promoter and T7 terminator universal primers, employing an automatic sequencer from Bioneer, Korea.

Recombinant protein expression

The chemically competent E. coli strains (BL21 (DE3), Rosetta, and BL21-gold) were transformed with the recombinant plasmid using the heat shock transformation procedure. The transformed cells were then plated on LB plates containing kanamycin (50 µg/ml) and incubated overnight at 37 °C. Plates with approximately 20 colonies were selected and stored at 4 °C for further analysis.

The protein expression of HPV16-MEVE7 was evaluated using different expression strains and conditions, including various temperatures (20 °C, 25 °C and 37 °C), different IPTG concentrations (0.10 mM and 1.00 mM), and other post-induction times (0, 2, 4, 6, 8, and 16 h). Pre-cultures were grown in Terrific Broth (TB) medium with 50 µg/ml kanamycin at 37 °C with shaking overnight. Subsequently, cultures were inoculated (1/60) with the pre-culture and incubated at appropriate temperature with shaking at 200 rpm. Induction of HPV16-MEVE7 expression was initiated by adding optimal concentration of IPTG when the selected OD600 reached. A time-course analysis of cell growth was performed by measuring OD600, and the expression analysis was conducted using 12.5% SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) using the Bio-Rad Mini-protein system (USA) and staining with Coomassie Brilliant Blue G-250. Through these analyses, optimal conditions for efficient HPV16-MEVE7 expression were achieved. Selecting the optimal strain and conditions was crucial for subsequent large-scale production. For large-scale production, a pre-culture of BL21-gold was prepared in 5 ml TB medium at 37 °C for 4 h. This culture was then diluted into 1 L of TB supplemented with kanamycin. IPTG was added (0.1 mM), when the OD600 reached 0.5 and after 4 h of induction, centrifugation at 5000 g for 15 min was done to harvest the cells. Western blot was performed to confirm MEVE7 expression. The cells lysate was run on an SDS-PAGE gel and then transferred onto nitrocellulose membranes. The membranes were blocked in 5% skim milk overnight at 4 °C. Then, the membranes were incubated for 2 h at room temperature with horseradish peroxidase (HRP)-conjugated monoclonal anti-His antibody (Sigma, USA) at a 1:2000 dilution. After incubation, the membranes were thoroughly washed and developed using a DAB solution (Sigma, USA). This process enabled the detection and confirmation of the expression of 6x histidine-tagged MEVE7 through specific antibody binding and subsequent visualization.

Recombinant protein purification

The cell pellets were carefully resuspended in a solution containing 50mM Tris-HCl (pH = 8.5), 5mM EDTA (Ethylenediaminetetraacetic acid), and 1mM PMSF (Phenylmethylsufonyl fluoride) before undergoing sonication for lysis. Sonication was performed through 20 cycles of 10 s each (1 s on, 1 s off) with a 20-second gap between cycles. Following sonication, the lysed pellets were subjected to centrifuge at 20,000 g for 15 min to separate the inclusion bodies. The supernatant was carefully removed, and the inclusion bodies were washed with 10 mL of the previously mentioned Tris-HCl buffer and then centrifuged using the same protocol. The resulting inclusion bodies were collected and stored at -20 °C until further solubilization or purification steps.

In the solubilization process of MEVE7 inclusion bodies, six different solubilization buffers (A-F) were evaluated. Equal amounts of inclusion body pellets were suspended in each buffer and incubated for 16 h at room temperature with mild shaking. After incubation, the solubilized protein samples were centrifuged at 20,000 g for 15 min at 4 °C to separate the insoluble (precipitated) and soluble (supernatant) fractions. The composition of the different solubilization buffers (A-F) is provided in Table 1. Subsequently, both fractions were analyzed using 12% SDS-PAGE gels to identify the right conditions for solubilizing the inclusion bodies. The protein bands were detected through staining with Coomassie Brilliant Blue, and the solubilization percentage was quantified using ImageJ software.

Table 1.

Different solvents A–F composition for the solubilization of MEVE7 inclusion bodies

Solubilization Buffer	Composition
A	50 mM Tris–HCl, 5 mM EDTA, 1 mM PMSF, and 8 M urea (pH 8.5)
B	50 mM Tris–HCl, 5 mM EDTA, 1 mM PMSF, and 6 M GdnHCl (pH 8.5)
C	50 mM Tris–HCl, 5 mM EDTA, 1 mM PMSF, and 2 M urea (pH 12)
D	50 mM Tris–HCl, 5% glycerol, 0.1 mM EDTA, 50 mM NaCl, and 0.4% sarkosyl (pH 7.9)
E	50 mM Tris–HCl, 5 mM EDTA, 1 mM PMSF, 6 M n-propanol, and 2 M urea (pH 8.5)
F	50 mM Tris–HCl, 5 mM EDTA, 1 mM PMSF, 6 M β-mercaptoethanol, and 2 M urea (pH 8.5)

Animal and cell lines

C57BL/6 female mice (8–12 weeks) were obtained from Pasteur Institute of Iran and were housed under 12-hour controlled light in pathogen-free condition.

HPV-16 E6 and E7-expressing TC-1 cells were obtained from ATCC (ATCC no. CRL-2785). The TC-1 cell line established by co-transformation of C57BL/6 primary lung with HPV-16 E6, E7 and c-Ha-ras oncogenes. These cells were grown at 37˚C in a humidified CO2 incubator in RPMI-1640 medium with 10% fetal bovine serum (FBS), 1% penicillin-streptomycin, 1% glutamine, 1% MEM-non essential, and 1% sodium pyruvate (all from Gibco).

Assessment of E7-specific immune responses

To evaluate the immunogenicity of the purified MEVE7, C57BL/6 mice were injected subcutaneously with the three doses of MEVE7 vaccine with two weeks interval. The initial and booster doses consisted of 20 and 10 µg of MEVE7 respectively. Blood samples were obtained one week following the last dose via cardiac puncture, and the sera were separated. The HPV16 E7-specific antibody response was assessed using an enzyme-linked immunosorbent assay (ELISA). Flat-bottom 96-well Maxisorp microtiter plates (Nunc, Denmark) were coated with MEVE7 at a concentration of 5 µg/mL in phosphate-buffered saline (PBS) (pH 7.4) and incubated overnight at 4 °C. The plates were then blocked with 3% (w/v) skimmed milk (Sigma Aldrich) in PBST (PBS containing 0.05% v/v Tween-20, Sigma-Aldrich). Serially diluted mouse sera were added to the plates in the blocking buffer and incubated at room temperature for 90 min. Following the incubation, the plates were washed three times with PBST, and plates were then incubated with horse-radish peroxidase (HRP)-conjugated sheep monoclonal anti-mouse IgG (produced in-house) at a dilution of 1:1000 in the blocking buffer at room temperature for 45 min. After another round of washing with PBST, the reactions were developed by adding tetramethylbenzidine (TMB) substrate solution (Pishtaz Teb, Iran) for 15 min. The enzymatic reaction was stopped by adding 1 M H2SO4. The optical density (OD) was measured at 450 nm and 630 nm as reference wavelength using a microplate reader (Biotek, USA).

Immunofluorescent staining

The TC-1 cells were seeded on Diagnostic (IFA) microscope slide (Tekdon, USA) and fixed with ice cold acetone for 2 min. Cells were then maintained at 4 °C and room temperature (RT) respectively for 15 min and washed twice for 2 min. The cells were incubated with hyper immune mouse sera at different dilution (1:100, 1:250, 1:500) for 90 min. Control slides for the negative reagent received non-immune mouse serum at the same dilution. Following a PBS + 2%FBS wash, the cells were incubated in the dark for 45 min with FITC-conjugated sheep anti-mouse Ig (diluted 1:100) and washed as mentioned above. DAPI (0.5 mg/ml) was used to stain the nuclei. Ultimately, the cells were observed under a fluorescence microscope (Olympus BX51, Japan) equipped with a DP71 CCD camera.

Therapeutic experiment

Ten C57BL/6 mice were randomly split into two groups of experiment and control (n = 5 per group). On day 0, all mice were challenged subcutaneously in their left flanks with 1 × 10⁶ TC-1 cells. On day 5, mice were either subcutaneously injected with vaccine (experimental group) or PBS (control group) into the contralateral side from the tumor cell inoculation side (100 µl per mouse). The vaccine consisted of a mixture (1:1) of purified recombinant MEVE7 (20µgr), CpG ODN 1825 (5µgr) from Invivogen, and RAS-01 adjuvant (oil in water). The mixture was gently shaken for 2 h before injection to form a stable water-in-oil emulsion. Tumor dimensions, recorded twice a week with a digital caliper, encompassed the shortest (as width) and longest (as length) surface diameters. The tumor size was determined by applying the following formula based on these two measurements:

Tumor volume (mm3) = ½ × (length×width2). The mice were kept until they died, or the tumor reached a volume of 2.5 cm³ (The mice were anesthetized by intraperitoneal injection of ketamine (2 mg/mice)/Xylazine (0.2 mg/mice)).

Results

MHC I and II-restricted HPV16-E7 epitope selection

E7 cytotoxic T lymphocyte (CTL) epitopes were predicted based on MHC class I alleles (H-2Db and H-2Kb) using the Immune Epitope Database (IEDB) with a percentile rank threshold of < 20. The IEDB analysis resource Consensus tool, integrating predictions from Artificial Neural Network (ANN) aka NetMHC, Stabilized Matrix Method (SMM), and Comblib, identified nineteen epitopes with lower percentile ranks and higher binding affinity for H-2Db and H-2Kb alleles [17]. Table 2 summarizes percentile rank, MHC I-restricted alleles, and methods used for different peptide predictions. Epitopes with weak binding affinity to the MHC allele (percentile rank > 20) were excluded from consideration.

Table 2.

Binding profile of the conserved MHC-I epitopes

Peptide	Start	End	Allele	Method used	Percentile rank
RAHYNIVTF	51	59	H-2-Db	Consensus (ann/comblib_sidney2008/smm)	0.2
STHVDIRTL	73	81	H-2-Db	”	3.2
DTPTLHEYM	4	12	H-2-Db	”	3.8
TDLYCYEQL	20	28	H-2-Kb	Consensus (ann/smm)	4.25
TLHEYMLDL	7	15	H-2-Kb	“	6.2
GTLGIVCPI	87	95	H-2-Db	Consensus (ann/comblib_sidney2008/smm)	8.5
RAHYNIVTF	51	59	H-2-Kb	Consensus (ann/smm)	9.7
VTFCCKCDS	57	65	H-2-Kb	“	10.45
STHVDIRTL	73	81	H-2-Kb	“	12.2
YMLDLQPET	11	19	H-2-Db	Consensus (ann/comblib_sidney2008/smm)	14.0
TPTLHEYML	5	13	H-2-Kb	Consensus (ann/smm)	15.5
FCCKCDSTL	59	67	H-2-Db	Consensus (ann/comblib_sidney2008/smm)	16.0
STLRLCVQS	65	73	H-2-Kb	Consensus (ann/smm)	16.5
YMLDLQPET	11	19	H-2-Kb	”	18.0
STLRLCVQS	65	73	H-2-Db	Consensus (ann/comblib_sidney2008/smm)	18.0
VDIRTLEDL	76	84	H-2-Kb	Consensus (ann/smm)	18.5
TDLYCYEQL	20	28	H-2-Db	Consensus (ann/comblib_sidney2008/smm)	19.0
YEQLNDSSE	25	33	H-2-Db	“	19.0
CVQSTHVDI	70	78	H-2-Kb	Consensus (ann/smm)	19.75

Prediction of T-helper epitopes from E7 was carried out by utilizing IEDB reference set for the MHC class II allele (H2-IAb), employing a percentile rank threshold of less than 55. Among these epitopes, ten were selected based on lower percentile rank and higher binding affinity for H2-IAb alleles using the IEDB analysis resource Consensus tool. Table 3 provides the binding profile of the MHC II epitopes, including the percentile rank, MHC II-restricted alleles, and methods used for various peptide predictions.

Table 3.

Binding profile of the conserved MHC-II epitopes

Peptide	Start	End	Allele	Method used	Percentile rank
IDGPAGQAEPDRAHYN	40	55	H2-IAb	Consensus (smm/nn)	43.00
DEIDGPAGQAEPDRAH	38	53	H2-IAb	“	45.50
DGPAGQAEPDRAHYNI	41	56	H2-IAb	“	47.50
TDEIDGPAGQAEPDRA	37	52	H2-IAb	”	47.50
EIDGPAGQAEPDRAHY	39	54	H2-IAb	”	48.50
GPAGQAEPDRAHYNIV	42	57	H2-IAb	”	49.00
DTDEIDGPAGQAEPDR	36	51	H2-IAb	”	53.00
AGQAEPDRAHYNIVTF	44	59	H2-IAb	”	55.00
PAGQAEPDRAHYNIVT	43	58	H2-IAb	”	55.00
EDTDEIDGPAGQAEPD	35	50	H2-IAb	”	55.00

Considering the experimentally confirmed peptides, six MHC class I and four MHC class II epitopes were finally selected to design the poly-epitopic E7 construct for further analysis.

Design and production of E7 poly-epitope protein

Based on published data [7, 13–16] indicating positive cellular immune responses in humans and, or C57BL/6 mice, and incorporating our new epitope predictions, the poly-epitopic E7 construct was designed to include HPV16-E7 immunogenic epitopes. Seven MHC class I-restricted epitopes and four MHC class II-restricted epitopes, selected for their optimal results and lowest scores, were included in the construct. These epitopes were chosen to induce CD8 + and CD4 + T lymphocytes, respectively, as they play crucial roles in tumor clearance and long-lasting cytotoxic responses. (Table 4).

Table 4.

CD + 4 and CD + 8 E7 epitopes were applied to assemble in the poly-epitopic E7

Epitope type	Amino acid sequence*	Position	MHC dependency
CD8+ T-lymphocyte	DTPTLHEYM	4–12	H-2Db
CD8+ T-lymphocyte	TLHEYMLDL	7–15	H-2Kb
CD4+ T-lymphocyte	LHEYMLDLQPETTDL	8–22	H-2Kb, H2-IAb
CD8+ T-lymphocyte	TDLYCYEQL	20–28	H-2Kb
CD4+ T-lymphocyte	DEIDGPAGQAEPDRA	36–50	H2-IAb
CD8+ T-lymphocyte	RAHYNIVTF	49–57	H-2Db, H-2Kb
CD4+ T-lymphocyte	RAHYNIVTFCCKCD	49–63	H-2Db, H-2Kb, H2-IAb
CD8+ T-lymphocyte	VTFCCKCDS	55–63	H-2Db, H-2Kb
CD8+ T-lymphocyte	STHVDIRTL	71–79	H-2Db, H-2Kb
CD4+ T-lymphocyte	MGTLGIVCPICSQKP	84–98	H2-IAb
CD8+ T-lymphocyte	TLGIVCPIC	86–94	H-2Kb

* The length of the epitopes included in the poly-epitopic E7 construct is 9 amino acids for MHC class I and 16 amino acids for MHC class II

The arrangement of epitopes in the poly-epitopic E7 construct was carefully designed to enhance its efficacy.

The current study employed the GPGPG linker to connect the HLA-II epitopes; for the joining of HLA-I epitopes, the AAY linker (Ala-Ala-Tyr) was used. The construction incorporated AYY and GPGPG linkers to avoid the formation of junctional epitopes. This inclusion aimed to enhance the immune processing of the antigens and enable the practical separation of each epitope within the human body [18].

The final sequence of the poly-epitopic E7 construct (Gene Bank accession number: MT510551) was as follows:

“E7_{36 − 50}GPGPG E7_8 − 22GPGPG E7_{84 − 98}GPGPG E7_{49 − 63}GGGAAA E7_{49 − 57}AAY E7_{20 − 28}AYY E7_{55 − 63}AAA E7_7 − 15AAY E7_{71 − 79}AYY E7_4 − 12AAA E7_{86 − 94}GGGAAA E7_{36 − 50}GPGPG E7_8 − 22GPGPG E7_{84 − 98}GPGPG E7_{49 − 63}GGGAAA E7_{49 − 57}AAY E7_{20 − 28}AYY E7_{55 − 63}AAA E7_7 − 15AAY E7_{71 − 79}AYY E7_4 − 12AAA E7_{86 − 94}AAY E7_{49 − 57}AAYRRR”.

Prediction of IFN-γ-inducing and allergen epitopes

In this study, the IFN epitope server was employed to analyze the selected E7 peptides and identify IFN-inducing epitopes, which play a crucial role in boosting the host immune response against HPV infection and associated cancers. The analysis, detailed in Table 5, identified four specific peptides (DTPTLHEYM, LHEYMLDLQPETTDL, RAHYNIVTFCCKCD, and STHVDIRTL) capable of stimulating the production of IFN-γ. Additionally, these selected peptides were tested for allergenicity, with the results confirmed their non-allergenic properties.

Table 5.

IFN-γ Inducing Scores of the Engineered E7 Epitopes

EPITOPE SEQUENCE	SCORE	ALLERGENICITY
DTPTLHEYM	+ 0.84 (POSITIVE)	None
TLHEYMLDL	− 0.36	None
LHEYMLDLQPETTDL	+ 0.04 (POSITIVE)	None
TDLYCYEQL	− 0.82	None
DEIDGPAGQAEPDRA	− 1.066	None
RAHYNIVTF	+ 0.126	None
RAHYNIVTFCCKCD	+ 0.618 (POSITIVE)	None
VTFCCKCDS	− 0.145	None
STHVDIRTL	+ 0.47 (POSITIVE)	None
MGTLGIVCPICSQKP	− 0.75	None
TLGIVCPIC	− 0.45	None

Linear B-Cell epitope prediction

The prediction of Linear B cell epitopes in HPV16 E7 was accomplished by analyzing protein structures, focusing on factors like hydrophobicity and surface charge. Utilizing the Kolaskar and Tongaonkar technique, seven specific peptides were identified as B cell linear epitopes within HPV16 E7, as outlined in Table 6.

Table 6.

Linear B cell Epitopes of HPV16-E7

EPITOPE SEQUENCE	POSITION	LENGTH
GIVCPICSQ	45–53	9
YNIVTFCCKC	64–73	10
YNIVTFA	84–90	17
YTDLYCYEQLAYYVTFCCKCDSA	92–114	23
LGIVCPIC	154–161	8
GIVCPICSQ	212–220	9
YNIVTFCCKC	231–240	10
YNIVTFA	251–257	7
YTDLYCYEQLAYYVTFCCKCDSA	259–281	23
LGIVCPICAAYR	321–332	12

Expression of the poly-epitopic E7 antigen

To increase yield of non-prokaryotic recombinant proteins in a prokaryotic system, various E. coli strains were utilized under different conditions to express MEVE7, considering codon usage and regulatory factors. The synthetic gene of HPV16-MEVE7 was subcloned into the pET-28a (+) vector. The accuracy of ligation was confirmed through colony PCR (Fig. 1A) and sequencing (Fig. 1B). Upon receipt of the constructs, the recombinant plasmid was transformed into three E. coli strains: BL21 (DE3), Rosetta, and BL21-gold.

Fig. 1.

Confirmation of ligation accuracy. (A) Gel electrophoresis of PCR products. Lanes 1: expected PCR product of ∼ 1043 bp; M: Marker (Qiagen, M10kpz). (B) Result of MEVE7 DNA sequencing

Protein analysis was performed using 12.5% SDS-PAGE to examine the profiles of control and induced cells. The optimal expression level was achieved in 0.1 mM IPTG (OD600 = 0.5) after 4 h of induction at 37 °C in BL21-gold (Fig. 2A, B, C). Under these conditions, BL21 (DE3) and Rosetta strains also expressed the MEVE7 protein, albeit with lower yields. The results revealed that the desired protein was predominantly expressed as inclusion bodies, and soluble protein was detected in limited amounts across all strains and conditions (Fig. 2D).

Fig. 2.

Impact of different conditions and various host strains on MEVE7 expression level. (A) Impact of different temperatures on MEVE7 expression level in BL21-gold. Lane 1: BL21-gold/pET28- MEVE7 without IPTG induction; lane 2: at 37 °C; Lane 3: at 25 °C; M: molecular weight marker. The arrow indicates the expressed MEVE7. SDS-PAGE analysis demonstrated that protein expression was increased at 37 °C compared to 25 °C, and no protein expression was observed at 20 °C (data not shown). (B) Impact of different concentrations of IPTG on MEVE7 expression level at 37 °C. Lane 1: positive control protein (45 kD); Lane 2: BL21/pET28- MEVE7 without IPTG induction; Lane 3: 0.1 mM IPTG; Lane 4: 1.0 mM IPTG; M: molecular weight marker. The arrow indicates the expressed MEVE7. 0.1 mM IPTG was found optimal for protein expression (at 37 °C). (C) Impact of various incubation times after induction on MEVE7 expression level in BL21-gold. Lane1: MEVE7 was expressed at different time points after IPTG induction 1 h (Lane 1); 2 h (Lane 2); 4 h (Lane 3); 8 h (Lane 4), 16 h (Lane 5); M: molecular weight marker. The arrow indicates the expressed MEVE7. The optimal induction time was 4 h. (D) Expression of MEVE7 in different strains of (E) coli in the presence of 0.1 mM IPTG (OD600 = 0.5) after 4 h induction at 37 °C. B.I: Before induction, A.I: After induction, IB: insoluble fraction, S: soluble fraction. The optimal expression level was achieved in in BL21-gold. Under these conditions, BL21 (DE3) and Rosetta strains also expressed the MEVE7 protein, albeit with lower yields. E. SDS-PAGE analysis of pET28-MEVE7 expression in E. coli BL21-gold. Lane 1: BL21/pET28- MEVE7 without IPTG induction; lanes 2: induced with 0.1 M IPTG for 16 h. The arrow indicates the expressed MEVE7. (F) Western blot analysis (using anti-His tag antibody) of MEVE7. M: molecular weight marker, Lane 2: MEVE7 (Additional unrelated lanes which were run in parallel were cropped and lane of MEVE7 was placed near molecular weight marker lane)

The HPV16 poly-epitopic E7 protein exhibited an apparent molecular mass of 36 kDa in induced cell pellets, as observed in Fig. 2E, and was absent in the control samples. Also, the HPV16 poly-epitopic E7 protein was characterized using western blot analysis with an anti-his-tag monoclonal antibody (mAb) (Fig. 2F).

Purification of the poly-epitopic E7 antigen

As expected, the poly-epitopic vaccines were primarily expressed as inclusion bodies. These inclusion bodies containing MEVE7 from E. coli were isolated through a washing process using a buffer containing Tris and EDTA. This specific buffer was instrumental in eliminating E. coli cell wall and membrane contaminants, as illustrated in Fig. 3A. The percentage of inclusion bodies in the pellets was estimated using ImageJ software. During this stage, the level of contaminating proteins significantly decreased, with approximately 57.1% of the remaining proteins in the washed pellets being identified as MEVE7 inclusion bodies. The purified inclusion bodies, primarily composed of MEVE7, were then subjected to solubilization steps before undergoing further purification.

Fig. 3.

(A) Isolation of the MEVE7-containing inclusion bodies with Tris buffer (pH: 8.5). Lane 1: Positive control (45 kDa); Lane 2: inclusion bodies without washing; Lane 3: washed IB. The arrow indicates the expressed MEVE7. (B) IB was solubilized with different buffers A-F (Table 1) and the precipitates (P) and soluble (Sol) fractions were subjected to SDS-PAGE analysis. Lane 1: Washed IB; Lane 2: Sol after solubilization with buffer A; Lane 3: P remained after solubilization with buffer A; Lane 4: Sol of buffer C; Lane 5: P of buffer C; Lane 6: Sol of buffer D; Lane 7: P of buffer D; Lane 8: Sol of buffer E; Lane 9: P of buffer E; Lane 10: Sol of buffer F, Lane 11: P of buffer F; Lane 12: Sol of buffer B; Lane 13: P of buffer B; M: molecular weight marker. The highest levels of soluble form were obtained in buffer C (Lane 4) and buffer A (Lane 2)

The efficacy of various denaturants and ionic detergents in solubilizing the inclusion bodies was evaluated using different buffers with varying pH values, and the percentage of solubilization of MEVE7 was quantified using ImageJ software. Figure 3B illustrates that buffer C, comprising 2 M urea at pH 12, resulted in the highest concentration of soluble MEVE7. Furthermore, buffer A, containing 8 M urea at pH 8.5, showed significant solubility. Specifically, 99% and 0.96% of MEVE7 were solubilized in buffers C and A, respectively. In contrast, other solubilization buffers performed less effectively in compared to buffers C and A. Consequently, an alkaline pH of 12.0, combined with a moderate urea concentration (8 M), facilitated the enhanced solubilization of MEVE7 from inclusion bodies at pH 8.5.

Reactivity of mouse sera against MEVE7

To evaluate the immunogenicity of the purified MEVE7, C57BL/6 mice were injected subcutaneously with three doses of MEVE7 vaccine, with a two weeks interval between each dose. ELISA was done to measure E7-specific antibody levels in sera of immunized mice. Our results clearly showed that anti-E7 antibodies raised significantly after three immunization doses. Antibody titer was high as 1:10000 dilution of hyperimmune sera yielded an optical density above 1. Sera from control mice did not react with E7, indicating specificity of the reaction (Fig. 4A).

Fig. 4.

Assessment of humoral immune response in the MEVE7 vaccinated mice. (A) After vaccination, mice were bled and serum titer of anti-MEVE7 was assessed by an indirect ELISA. The results clearly demonstrated induction of vaccine-specific immune responses in the vaccinated mice with optical density over 1.0 at dilution of 1:10000. Sera from control mice did not react with E7, indicating specificity of the reaction. (B) Reactivity of immunized mice sera with E7 expressing mice cell line TC-1 was evaluated with immunofluorescent staining. Antibodies generated by MEVE7 immunization exhibited strong reactivity with TC-1 cells, predominantly localized near the cell membrane (For immunoreactivity, FITC-labeled anti-mouse Ig antibody was used.)

Immunofluorescent staining for cross-reactivity analysis

To investigate the reactivity of antibody responses against TC-1 cells in mice receiving the vaccine, a blood sample was taken one week after the last immunization. The serum was then separated for further analysis. Our finding indicated that sera from the control group did not react with TC-1 as negative control. Indeed, we used B16F10 cells as negative control cells. As expected, sera from immunized mice exhibited no reaction with these cells. However, sera from vaccine group mice highly reacted with the cognate antigen in TC-1 cells mainly situated at subsurface and cell surface (Fig. 4B).

Therapeutic immunization of tumor-bearing mice

The anti-tumor effect induced by vaccination was assessed by the tumor size and survival rate of vaccinated mice. In the therapeutic strategy, tumor-bearing mice received either E7 (single dose) or PBS immunization five days after injection of TC-1 cells, and tumor growth was then measured twice weekly. The size of tumors was monitored until they gained maximum size. The results indicated that tumors in the vaccine group begin to emerge around day 54 and persist until approximately day 85 after the tumor challenge. In contrast, the control group exhibits a faster onset of tumor growth, commencing on the 7th day after the tumor challenge and reaching its maximum size around the 21st day. So, vaccination delayed tumor development in vaccinated C57BL/6 mice, meanwhile tumors grew and developed in size continual in control mice receiving PBS (Fig. 5A). As shown in Fig. 5B, E7 (single dose) vaccination remarkably increased survival rate compared to control group.

Fig. 5.

Assessment of the MEVE7 vaccine capacity to control tumor development and progression: Mice (n = 5 in each group) were subcutaneously inoculated with 1 × 10⁶ TC-1 and five days later received either PBS or a single dose of MEVE7 vaccine at the contralateral side. Mice were regularly monitored for tumor volume 2–3 times weekly until tumor sizes exceeded 2500 mm³. Significant inhibition of TC-1 tumor growth was observed following immunization with MEVE7 emulsified in CpG and o/w adjuvant (A). Survival of mice was recorded over the experimental period (21 Days for control group and 85 days for experimental group). **: p < 0.01

Discussion

Cervical cancer is a significant public health issue that affects millions of women worldwide. Persistent infection with HPV, specifically HPV-16, is the leading cause of cervical cancer. Prophylactic vaccines like Gardasil and Cervarix have been developed against HPV viral infections. While these vaccines effectively prevent new HPV infections, their widespread adoption remains limited, particularly in low-income and densely populated countries with the highest incidence of cervical cancer. Consequently, cervical cancer remains a significant cause of mortality among women. In light of this, the design of an effective therapeutic vaccine to address the HPV-infected population and reduce the impact of this deadly disease is necessary [19, 20].

While some studies suggest that HPV late proteins, particularly L2 protein as VLP vaccine, may eliminate persistent HPV infections in mouse models [21, 22], the WHO’s 2024 update highlights that antibodies alone are inadequate for clearing persistent HPV infections or reducing precancerous lesions. Consequently, while current prophylactic HPV vaccines depend on antibody-mediated protection, post-exposure therapeutic vaccines will likely need to elicit robust cell-mediated immunity, including effective T cell responses targeting early viral proteins across genetically diverse populations [23].

Various forms of therapeutic HPV vaccines have been designed for the treatment of cervical cancers, including peptide and protein [24], live vector (both bacterial and viral) [25], DNA/RNA [26], and dendritic cell [20]. Among these, peptide-based vaccines offer distinct advantages. They are characterized by simple and cost-effective synthesis methods, enhanced stability, and high safety profiles. Also, the peptide vaccine can be designed with a wide range of epitopic peptides to properly trigger the immune responses against many cancer and infectious diseases.

Recently, multi-epitope vaccines have demonstrated the capability to elicit responses restricted by a diverse range of HLA molecules, fostering a balanced immunity involving CD4 + and CD8 + T cells [27]. The focus of this investigation was to develop a novel multi-epitope vaccine to improve immune responses for prevention of tumor growth. Therefore, we outlined the process for producing a novel recombinant protein E7, from design to characterization as a therapeutic vaccine for HPV-mediated cervical cancer. This protein holds potential for the treatment of precancerous lesions of the cervix and cervical cancer in women infected with HPV-16. In previous investigation [12], we identified immunogenic and conserved HPV16-E7 epitopes designed for interaction with human HLAs. In the present study, we verified their interaction with mice MHC alleles, specifically those of C57BL/6 mice. Our findings revealed that administering the designated vaccine can elicit antibody responses capable of recognizing the native E7 protein located near the cell membrane of TC-1 cells. Notably, in our therapeutic study, the vaccine administered as a single dose after tumor challenge with 1 × 10 ⁶ TC-1 cells, significantly delayed tumor onset for 54 days, after which tumors began to grow in a slow progression rate compared to the control group. A follow up for 15 days after tumor development revealed that, tumor size increased from 300 to only 600 mm³ in the vaccine group, while control mice experienced a very fast kinetic of tumor from 100 to 2500 mm³ during the same period. Accordingly, the slope of the tumor growth in the vaccinated was almost 8 times lower that the control group (21 versus 171). These findings indicate that our vaccine not only kept tumor in a latent phase and delayed tumor onset but also significantly reduced tumor growth.

Several experimental studies have shown the immunogenicity of various identified antigenic peptides derived from HPV16/18 E6 and E7 proteins. For example, a study illustrated that administering the three doses of the entire sequence of HPV16, 18, 6, 11 E7-Tat (20 µg) along with adjuvant resulted in a high anti-E7 antibody titer. Additionally, they demonstrated that vaccination with E7-Tat elicited elevated frequencies of E7-specific CD8 + T cells, and extended long-term survival by approximately 45 days in therapeutic animal models [28].

In another study, exosomes containing crocin and curcumin compounds, were combined with the L1-E7 polypeptide to assess their anti-tumor effects in a C57BL/6 mouse model. After tumor challenge with 0.1 × 10⁶ TC-1 cells, the mice were injected three times with the L1-E7 polypeptide (5 µg) and ExoCrocin (2 mM crocin loaded in 200 µg exosome) emulsified in Montanide 720. Tumor growth was then monitored for 65 days. In vivo investigations demonstrated that combining the L1-E7 polypeptide with ExoCrocin or ExoCurcumin resulted in a substantial immune response characterized by Th1 response and cytotoxic T lymphocyte (CTL) activity. These combinations exhibited therapeutic effects against tumor cells [27]. Oliveira et al. described the development and evaluation of a recombinant multi-epitope protein that includes immunogenic epitopes from HPV-16 E6 and E7. C57BL/6 mice were initially subcutaneously injected with 7.5 × 10⁴ TC-1 cells. After Three days, the immunization regimen commenced, involving three subcutaneous doses of E6E7 (18 µg/dose). The vaccine demonstrated the induction of robust activation in E6/E7-specific T cells. In a therapeutic context, the E6E7 protein provided comprehensive protection against mice tumor development for 70 days [7]. The assessment of therapeutic vaccines relies on various factors. (i)Tumor inoculation Dose: The greater the number of injected tumor cells, the greater the vaccine’s challenge. (ii) Vaccine dosage and frequency: The quantity and frequency of vaccine doses impact the activation of the immune system. Higher doses and increased frequency contribute to a more robust immune response. (iii) Time interval between tumor challenge and vaccine injection: A longer interval between tumor challenge and vaccine injection provides tumor cells more time to increase, presenting a more significant challenge for the vaccine. (iv) Duration of protection: evaluating the duration of protection after administering the last vaccine dose is crucial in understanding the vaccine’s long-term efficacy. (v) Proximity of injection site to tumor challenge: An effective vaccine should be able to induce an appropriate response even if it is not injected at the same site as the tumor site.

Considering the five factors mentioned above, i.e.: inoculation with a higher number of tumor cells, using a single dose of vaccine, the longer interval between tumor challenge and vaccine injection, extended protection time, and injection of the vaccine at a site far from the tumor challenge site, our vaccine demonstrated superior and more prolonged effects compared to the vaccine reported so far. These remarkable results highlight the potential of our vaccine to be further explored and refined by defining the number of vaccinations, optimal dosing and vaccination intervals required to maximize efficacy and to induce a robust, long-term immune response.

Indeed, the effectiveness of therapeutic vaccines, including our HPV vaccine, is largely dependent on the capacity of the host’s immune system. It is well-documented that individuals with compromised immune systems, such as those living with HIV, are at an increased risk for cervical cancer and often face worse outcomes due to impaired immune responses. Given the higher incidence of cervical cancer in HIV-positive patients, they represent a key population that could greatly benefit from therapeutic HPV vaccines. Although HIV-related immunosuppression presents challenges, vaccine strategies that can elicit strong immune responses hold potential for improving outcomes in these individuals. Thus, enhancing the vaccine’s efficacy for immunocompromised populations, particularly those with HIV, should be a priority for future research and clinical development [29].

Conclusion

In the present study, we illustrated that vaccinating mice with the MEVE7, encompassing CTL and T helper epitopes, can evoke antibody responses capable of identifying the native E7 protein and effectively impede tumor growth in a therapeutic context. These results hold significant promise for the development of an effective vaccine targeting HPV-induced cervical tumor cells. Our team is now focusing to improve this vaccine efficacy by optimizing the vaccination schedule, dosing and formulation.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Abbreviations

HPV

Human papillomavirus

MEVE7

HPV16-E7 multi-epitopic vaccine

TC-1

Tumor cell 1

IEDB

Immune Epitope Database and Analysis Resource

SVM

Support Vector Machine

EDTA

Ethylenediaminetetraacetic acid

PMSF

Phenylmethylsufonyl fluoride

PBS

Phosphate-buffered saline

ANN

Artificial Neural Network

SMM

Stabilized Matrix Method

Author contributions

Conceptualization: ZA, AZ; Original draft preparation: BB; methodology: BB, MH; in silico study: MR, BB; writing review and editing: ZA, AZ. All authors have read and agreed to the published version of the manuscript.

Funding

This research work (Project No: CCF-0007) received financial support from the Cancer Control Research Center, Cancer Control Foundation, Iran University of Medical Sciences, Tehran, Iran, and in part from the Iranian Research Organization for Science and Technology (IROST).

Data availability

The gene sequence has already been reported in NCBI with Gene Bank accession number: MT510551 (https://www.ncbi.nlm.nih.gov/gene/?term=+MT510551), other data analyses in our study are available through the corresponding author on reasonable request.

Declarations

Ethical approval

For this study, ethical approval was obtained from Research Ethics Committee of Tehran University of Medical Sciences (IR.TUMS.SPH.REC.1403.123).

Conflict of interest

The authors declare no conflict of interest.

Consent for publication

Not applicable.