A Genetically Modified attenuated Listeria Vaccine Expressing HPV16 E7 Kill Tumor Cells in Direct and Antigen-Specific Manner

Attenuated Listeria monocytogenes (L. monocytogenes, LM) induces specific CD8+ and CD4+ T cell responses, and has been identified as a promising cancer vaccine vector. Cervical cancer is the third most common cancer in women worldwide, with human papillomavirus (HPV), particularly type 16, being the main etiological factor. The therapeutic HPV vaccines are urgently needed. The E7 protein of HPV is necessary for maintaining malignancy in tumor cells. Here, a genetically modified attenuated LM expressing HPV16 E7 protein was constructed. Intraperitoneal vaccination of LM4Δhly::E7 significantly reduced tumor size and even resulted in complete regression of established tumors in a murine model of cervical cancer. We provided evidence that recombinant LM strains could enter the tumor tissue and induce non-specific tumor cell death, probably via activation of reactive oxygen species and increased intracellular Ca2+ levels. LM4Δhly::E7 effectively triggered a strong antigen-specific cellular immunity in tumor-bearing mice, and elicited significant infiltration of T cells in the intratumoral milieu. In summary, these data showed LM4Δhly::E7 to be effective in a cervical cancer model and LM4Δhly::E7 induced an antitumor effect by antigen-specific cellular immune responses and direct killing of tumor cells, indicating a potential application against cervical cancer.


INTRODUCTION
Cervical cancer is the third most common female cancer worldwide. About 530,000 new cases and 275,000 deaths are reported annually, with 85% and 88% of cases and deaths, respectively, occurring in developing countries (Ferlay et al., 2015;Torres-Poveda et al., 2016). Conventional therapeutic approaches for cervical cancer are based on chemotherapy and radiation therapy, but the 5-year survival of patients is dismal (Lee et al., 2013;Killock, 2015). Over 99% of cervical cancer is caused by persistent infection with human papillomavirus (HPV), particularly type 16. Currently, although two prophylactic virus-like particle-based vaccines are available, they have no effect on existing HPV-associated diseases. Furthermore, HPV infections remain extremely common.
Thus, there remains an urgent need to develop therapeutic HPV vaccines (Depuydt et al., 2016;Stanley, 2016). Cancer immunotherapy has emerged as an alternative, innovative approach that can induce the persistent cell-mediated immunity necessary to retard tumor growth. So far, HPV16 E6 and E7, which are constitutively expressed in cervical cancer cells, seem to be the most promising target genes for therapeutic vaccines against cervical cancer (Grabowska et al., 2015). Several types of therapeutic vaccines against HPV-associated cancer have been developed and tested in preclinical trials, including protein, live vector, nucleic acid, and cell-based vaccines. Live bacterial vectorbased therapeutic HPV vaccines are one of the most promising vector vaccines because of highly immunogenic and convenient. Nowadays, Listeria monocytogenes (L. monocytogenes, LM), Lactobacillus lactis, and salmonella are the most common bacteria-based vectors for tumor immunotherapy . Among the intracellular bacteria, Listeria has been recognized as a promising vector because of its unique immunologic properties (Rothman and Paterson, 2013).
L. monocytogenes is a gram-positive, facultative intracellular bacterium. Once inside cells, the bacterium produces specific virulence factors which lyse the vaculolar membrane and allow escape into the cytoplasm. L. monocytogenes in the cytosol is capable of inducing CD8 + and CD4 + T cell responses via MHC class I and MHC class II pathways (Pillich et al., 2015). LM also induces a decrease in regulatory T cells (Tregs) in tumor tissue, thus favoring immune responses that can kill tumor cells . These characteristics make attenuated LM a promising candidate cancer vaccine vector (Wood and Paterson, 2014;Wan et al., 2015).
Previous studies have reported that attenuated LM was applied to deliver the HPV E7 gene via a plasmid-based complementation system, but there existed a high risk of plasmid loss and persistence of antibiotic resistance genes in clinical trials (Peng et al., 2004;Chen et al., 2014). Additionally, Wallecha demonstrated that detoxified listeriolysin (LLO) as a fusion partner, or mixed with E7 antigen which were expressed by E. coli BL21(DE3) could augment specific immune responses and significant reduction of tumor growth (Wallecha et al., 2013), but the protein-based vaccine was expensive and low immunogenicity. In the present study, a genetically modified attenuated recombinant LM4 hly::E7 strain was constructed to express and secrete the LLO-E7 fusion protein. The study presented here demonstrates that this recombinant (r) LM vaccine is safe and effectively retards tumor growth in a mouse model of cervical cancer. Furthermore, we provide evidence that the antitumor mechanism is due to induction of specific cytotoxic T lymphocyte (CTL) responses and direct killing of tumor cells.

Mice and Cell Lines
C57BL/6 mice (7-8 weeks old) were purchased from the Comparative Medical Center of Yangzhou University. Animals were housed and used in accordance with the protocols approved by the Experimental Animal Center Institutional Committee of Yangzhou University. The TC-1 cell line, C57BL/6 lung tumor epithelial cells immortalized by HPV16 E6/E7 and transformed with the c-Ha-ras oncogene, was purchased from the Tumor Cell Center of the Chinese Academy of Medical Sciences (Beijing, China). RAW264.7 murine macrophage-like cells were maintained in our laboratory. Cells were cultured in RPMI 1640, supplemented with 10% fetal calf serum, 1 mmol/L sodium pyruvate, 100 U/mL penicillin, and 100 µg/mL streptomycin in a 37 • C incubator with 5% CO 2 .

Construction of Recombinant L. monocytogenes and Antigen Expression
The upstream fragment hlya and downstream fragment of hlyb were amplified from the LM4 genome using the following primers: hlya (F) 5 ′ -CCGGTACCAGGTTTGTTGTGTCAGG TAGAGC-3 ′ , hlya (R) 5 ′ -GTAGGTGTATCTCCATGTGAAATT GAATTTTCTTTAT-3 ′ ; hlyb (F) 5 ′ -TGTTCTCAGAAACCACC AATCGAAAAGAAACACG-3 ′ , hlyb (R) 5 ′ -GGGTCTAGAAC TTGAGATATATGCAGGAGG-3 ′ ; the BamHI and XhoI sequences are underlined. HPV16 E7 gene was amplified from pcDNA3.1-E7 using the following primers: E7 (F) 5 ′ -ATAAAG AAAATTCAATTTCACATGGAGATACACCTAC-3 ′ , E7 (R) 5 ′ -CGTGTTTCTTTTCGATTGGTGGTTTCTGAGAACA-3 ′ . hlya, hlyb and E7 were spliced by overlap extension PCR with the primers hlya (F) and hlyb (R). The fusion fragment aEb was cloned into the shuttle vector pKSV7 with restriction enzyme sites BamHI and XhoI. The resulting plasmid pKSV7-aEb was electroporated into electrocompetent LM4 cells. The E7 gene was cloned in-frame with the hly open reading frame and inserted downstream of the hly signal sequence by allelic exchange, and a 42 bp region was deleted from the N-terminal Pro-Glu-Ser-Thr (PEST)-like sequence of the hly gene. The recombinant plasmidand antibiotic-resistance gene-free strain LM4 hly::E7 was constructed to express and secrete LLO-E7 fusion protein, retaining a significant degree of attenuation. An irrelevant Listeria construct, LM4 hly::esat-6 (LM control), which was based on the same method of construction, was included to account for the antigen-independent effects of Listeria on the immune system. Supernatants from the recombinant strain cultures were concentrated by trichloroacetic acid (TCA), and detected by Western blot using a monoclonal antibody against HPV16 E7 protein.

Hemolysis Test
LM4 and LM4 hly::E7 were cultured overnight in brain-heart infusion (BHI) medium. The culture supernatants were collected by centrifugation, and the same OD 600 value was adjusted using PBS buffer. Seventy microliters of each supernatant were added to a 96-well plate, and then two-fold serial dilution was performed. Subsequently, 30 µL of 1% sheep erythrocytes were added, and hemolysis was observed after 1 h at 37 • C. Hemolysis titers were expressed as the dilution of the supernatant that lysed 50% of total sheep erythrocytes.

Macrophage Invasion Study In vitro
Firstly, pNF8 was electroporated into electrocompetent LM4 and LM4 hly::E7 cells, resulting in construction of LM4(pNF8) and LM4 hly::E7(pNF8). Subsequently, RAW264.7 cells cultured in 24-well plates were infected with LM4(pNF8) or LM4 hly::E7(pNF8) at a multiplicity of infection of 1:20 for 1 h. The cells were incubated in medium containing 50 µg/mL of gentamicin for 1 h at 37 • C to eliminate extracellular bacteria. The cells were fixed and stained with phalloidin (Invitrogen), which stains filamentous actin, by taking samples at 1 and 5 h. Intracellular bacteria were monitored by confocal-transmission electron microscopy.

Tumor Model and Immunization
To establish the tumor model, 2 × 10 5 TC-1 cells were subcutaneously injected into the left flank of C57BL/6 mice. After 1 week, tumor-bearing mice (tumor diameter reached 5 mm) were intraperitoneally administered with 0.1 50% lethal dose (LD 50 ) of LM4 hly::E7, LM4 hly::esat-6 or PBS on days 7 and 14 after tumor cell injection. Tumors were monitored, and measured at the longest (L) and widest (W) points with calipers. Tumor volume was calculated as L × W 2 × π/6. Experiments were initiated in the tumor-bearing mice to analyze the antitumor immune response.

Kinetics and Distribution of Lm4δhly::E7 in In vivo Infection Model
The tumor-bearing mice were intraperitoneally injected with LM4 hly::E7 on days 0, 7, and 14. Spleens, livers and tumors were homogenized on days 1 and 3 after each immunization. Bacterial numbers were determined by plating the cell suspensions on BHI agar.

In vivo Cytotoxicity Assay
A standard in vivo killing assay was performed as previously described (Coles et al., 2002). In brief, splenocytes were pooled from naive C57BL/6 mice and divided into two groups. One was labeled with 2.5 µM carboxyfluorescein succinimidyl ester (CFSE high ; Molecular Probes, Invitrogen, Carlsbad, CA) buffer, another was labeled with 0.25 µM CFSE (CFSE low ). CFSE high cells were pulsed with the E7 peptide. CFSE high and CFSE low cells were mixed at a 1:1 ratio and injected intravenously into C57BL/6 mice at 7 days post-immunization. Fifteen hours later, spleens were harvested and processed into single cell suspensions. The splenocytes were analyzed by flow cytometry according to fluorescence density, and the specific lysis ratio was calculated with the following formula:100 -(100 × (% CFSE high immunized/%CFSE low immunized)/(% CFSE high control/% CFSE low control)).
Flow Cytometric Analysis of CD4 + and CD8 + T Cells On day 7 after the second immunization, tumors were excised, and cut into 1-2 mm pieces using a sterile razor blade and digested with PBS buffer containing 2 mg/mL collagenase type I and 12 U/mL DNase (Invitrogen, Carlsbad, CA, USA) for 2 h at 37 • C with agitation. Single cell suspensions were pooled through a nylon mesh.

Immunohistochemical Analysis of the Tumor Tissue
Tumor specimens on day 7 after the second immunization were fixed in 13% neutral buffered formalin and embedded in paraffin. Sections were cut at 5 µm thick, and incubated with anti-mCD8 (clone 53-6.7; R&D Systems, Minneapolis, MN) or anti-mCD4 (clone GK1.5; R&D Systems) antibody overnight at 4 • C. The anti-mCD8 or anti-mCD4 antibody was visualized with a HRP-DAB Cell and Tissue Staining Kit (R&D Systems).

Annexin V-FITC/PI Staining
Tumor cell suspensions were obtained as described above. The cells were stained with FITC-labeled Annexin V (BD Pharmingen) and propidium iodide (PI; BD Pharmingen) for 15 min in the dark at room temperature. Subsequently, the rate of cell apoptosis and cell death were evaluated quantitatively by flow cytometry.

Measurement of Reactive Oxygen Species (ROS)
The generation of intracellular ROS was measured using a dichloro-dihydro-fluorescein diacetate (DCFH-DA) dye assay. DCFH-DA penetrates the cells and is hydrolyzed to DCFH by non-specific esterases. In the presence of cellular oxidizing agents, DCFH is rapidly oxidized to the highly fluorescent compound dichlorofluorescein (DCF), which is trapped inside the cells. Therefore, the fluorescence intensity is proportional to the amount of ROS generated in the cells. In brief, tumor cell suspensions were obtained as described above. The cells were stained with DCFH-DA (10 µM, Beyotime Institute of Biotechnology, Beijing) at 37 • C for 20 min, and then washed to remove extracellular DCFH-DA. Changes in DCF fluorescence were assayed by flow cytometry.

Measurement of Intracellular Calcium Ion (Ca 2+ ) Concentration
Changes in intracellular Ca 2+ levels were detected by Fluo-3 acetoxymethyl ester (Fluo-3 AM, Molecular probe Inc. USA). Intracellular Fluo-3 AM is deacetylated by esterases to yield Fluo-3, which is trapped inside the cells. Upon Ca 2+ binding, Fluo-3 exhibits strong fluorescence. Briefly, the cells isolated from tumor tissue were incubated with 5 µM Fluo-3 AM at 37 • C for 20 min, and then washed with PBS buffer to remove extracellular Fluo-3 AM. The fluorescence intensity of Fluo-3 was analyzed by flow cytometry.

Statistical Analysis
Statistical analysis for in vitro and in vivo experiments was carried out using GraphPad Prism Software (GraphPad Software Inc., La Jolla, CA). Student's t-test and one-way ANOVA were used for analysis of the comparisons between the groups. Statistical significance was established as P < 0.05 ( * P < 0.05, * * P < 0.01, * * * P < 0.001).
LM4 hly::E7 Can Infect and Replicate in Macrophages LM4(pNF8) and LM4 hly::E7(pNF8) expressing green fluorescent protein were constructed and used to infect RAW264.7 cells, following which the multiplicity of these LM strains in RAW264.7 cells was determined. As shown in Figures 2A,B and S1, there were only a few bacteria infecting any given cell after 1 h incubation. However, after 5 h, the amounts of LM4(pNF8), and LM4 hly::E7(pNF8) were increased (Figures 2C,D and S1).

Infection Kinetics in Organs from
Tumor-Bearing Mice Immunized with LM4 hly::E7 To determine the infection kinetics in different organs of tumorbearing mice immunized with LM4 hly::E7, spleens, livers and brains were harvested, along with tumors from the immunized mice at different time-points. Tissues were homogenized and CFU/organ was calculated. The results showed that the bacterial numbers in spleen and liver peaked on the 3rd day after the first immunization, and gradually decreased if further immunization doses were administered. As expected, LM was FIGURE 3 | Infection kinetics in organs of mice immunized with LM4 hly::E7. Groups of five tumor-bearing mice were intraperitoneally immunized with LM4 hly::E7 on days 0, 7, and 14. Spleens, livers, brains, and tumors were extracted and homogenized on 1 and 3 days after each immunization. Homogenates were diluted with PBS buffer and plated on BHI agar to determine bacterial numbers. (A) CFUs/organ was calculated on 1 and 3 days after each immunization, and infection kinetics in the organs are shown. (B) LM4 hly::E7 bacilli in the tumors were observed by electron microscopy. Scale bar 0.5 µm, white arrows, LM4 hly::E7 bacilli. Data are expressed as the mean and SD values from three independent experiments.
Frontiers in Cellular and Infection Microbiology | www.frontiersin.org not found in brain homogenates. LM4 hly::E7 was quantified in tumor tissues on the 1st day after the first immunization, and bacterial numbers started to increase, peaking at day 3 after the second immunization ( Figure 3A). Using electron microscopy, numerous LM4 hly::E7 bacilli were observed in the tumors in Figure 3B.

Vaccination with LM4 hly::E7 Leads to Stasis in Tumor Growth
To assess the antitumor efficacy of LM4 hly::E7, tumor-bearing mice were intraperitoneally immunized twice with LM4 hly::E7, LM4 hly::esat-6 or PBS buffer at weekly intervals. The tumor volume was measured twice weekly. As shown in Figures 4A,B, LM41hly::E7 significantly reduced tumor sizes and slowed the growth of tumors compared to the LM4 hly::esat-6 and PBS groups. Moreover, three of eight mice in the LM4 hly::E7immunized group demonstrated complete tumor regression.

LM4 hly::E7 Elicits Strong E7-Specific Cellular Immunity in Tumor-Bearing Mice
In order to investigate the cellular immune responses induced by LM4 hly::E7 in tumor-bearing mice, the numbers of IFN-γand IL-4-secreting cells in splenocytes harvested 7 days after the second immunization were determined using an ELISPOT assay. As shown in Figure 5A, immunization with LM4 hly::E7 significantly increased the amount of E7-specific IFN-γ-secreting cells as compared with the control LM4 hly::esat-6 and PBS groups ( * * P < 0.01). Also, the number of IFN-γ-secreting cells were significantly expanded compared with IL-4-secreting cells in the LM4 hly::E7-immunized group ( * * P < 0.01), which demonstrated that the immune response induced by LM4 hly::E7 is skewed toward a Th1 type response in tumorbearing mice. CTL activity in vivo was determined using a CFSE-based assay. As shown in Figures 5B,C, tumor-bearing mice immunized with LM4 hly::E7 generated a robust cytotoxic response with an E7specific lysis ratio of 61.62 ± 6.2%, which was significantly higher than the control groups ( * * P < 0.01).

LM4 hly::E7 Induces the Highest Ratio of T Cell Subpopulations in Tumors
By comparing the percentages of T cell subpopulations isolated from tumor-bearing mice in the different experimental groups, we observed by flow cytometry that more CD4 + and CD8 + T cells infiltrated into the tumor site in the LM4 hly::E7 and LM4 hly::esat-6 groups compared with the FIGURE 4 | Tumor regression in tumor-bearing mice treated with rLMs. Groups of eight mice were subcutaneously injected with 2 × 10 5 TC-1 cells. When the diameter of the tumor was 5 mm, mice were intraperitoneally immunized with LM4 hly::E7, LM4 hly::esat-6 or PBS buffer. One week later, mice were given a booster immunization, and the tumors were monitored.  PBS group (Figures 6A,B). A similar result was found in the immunohistochemistry assay: high positive staining levels of CD4 + and CD8 + T cells were observed in the rLM-immunized groups ( Figure 6C).

Recombinant Listeria Strains Induce Tumor Cell Death and Cell Apoptosis
Next, we investigated the changes of cell death in the tumor tissues. Tumor tissues from tumor-bearing mice treated with rLM were extracted and processed into single cell suspensions. By Annexin-V-FITC and PI staining, our results showed that LM4 hly::E7 and LM4 hly::esat-6 could induce tumor cell death, which were significantly different from the PBS group ( * * P < 0.01). However, there were no significant differences between the two rLM-immunized groups (Figure 7).

Recombinant Listeria Strains Induce Higher Intracellular ROS and Ca 2+ Levels in Tumor Cells
Intracellular ROS assays in tumor cells indicated that LM4 hly::E7 and LM4 hly::esat-6 could induce the generation of intracellular ROS compared with the PBS group ( Figure 8A). Furthermore, the results revealed that rLM strains also induced elevations in intracellular Ca 2+ concentration ( Figure 8B).

DISCUSSION
Cervical cancer is a serious health concern, and is due to persistent infection with high-risk subtypes of HPV, including types 16, 18, 31, and 45 (Yuji and Nakada, 2016). Preventive vaccines for cervical cancer do not exert therapeutic effects on pre-existing HPV infections and HPV-associated lesions, and thus recent advances in immunotherapy treatment of cervical cancer have gained much attention lately (Yang M. C. et al., 2016). These therapeutic vaccines aim to generate cellular immune responses to tumor cells. L. monocytogenes is capable of escaping phagocytosis and inducing a strong cellular immune response (Cory and Chu, 2014). Studies have indicated that attenuated LM was able to induce antitumor immunity against a number of tumor associated antigens such as hepatic cancer stem cell biomarker CD24 and malignant melanoma MIA gene (Qian et al., 2012;Yang et al., 2014). Furthermore, fusion of the FIGURE 6 | Characterization of CD4 + and CD8 + T cells in tumor tissues after immunotherapy. (A,B) Tumor cells were isolated 7 days after the final vaccination. Cells were then stained with anti-CD3-FITC, anti-CD4-PE, and anti-CD8-APC monoclonal antibodies, and analyzed by flow cytometry. (C) CD4 + and CD8 + T cells in tumor tissues were analyzed by immunohistochemistry (Orignal magnification ×200), black arrows, CD4 + T cells (brown granules), white arrows, CD8 + T cells (brown granules). Images are representative in the immunized groups and the PBS group. Data are expressed as the mean and SD values from three independent experiments. *P < 0.05; **P < 0.01, Student's t-test.
antigen (for example, the epidermal growth factor receptor family of tyrosine kinases, HER-2/neu; prostate-specific antigen, PSA and so on) to dtLLO or ActA (encoded by the actA gene) and secretion by L. monocytogenes was found to significantly enhance anti-tumor immune responses in comparison to an LM vector that secreted only the tumor antigen (Singh et al., 2005;Shahabi et al., 2008). In the present study, a new strategy involving genetic modification of an attenuated LM strain to express and secrete LLO-E7 fusion protein has been developed. While established tumors (tumor diameter>5 mm) are known to reside in a highly immunosuppressive microenvironment, LM4 hly::E7 could lead to the eradiction of the tumors by the end of the experiment. Plasmid based complementation system was always applied as a method of constructing recombinant LM. However, plasmid retention often requires antibiotic resistance markers, the presence of which has been discouraged in clinical applications by the Food and Drug Administration (FDA) (Azizoglu et al., 2014). Perviously, studies reported that LLO-E7 fusion protein resulted in reductions in tumor volume (Wallecha et al., 2013). Subunit vaccines, however, suffer from low immunogenicity and most are presented through the MHC class II pathway which activates the production of antibodies instead of generating a CTL response. So HPV16 E7 was integrated into the downstream of hly signal sequence within LM genome by homologous exchange in this study, and LLO-E7 fusion protein stably secreted by Listeria. Although expression quantity of foreign antigen in this way is lower than that of plasmid retention system, the recombinant strain LM4 hly::E7 here led to the long-term regression of established tumors in a transplantable tumor model. These results indicated that the recombinant strain with this construction method was a more efficient delivery vector.
The role of therapeutic cancer vaccines is mainly to induce activation of antigen-specific CD8 + T cells, which can release proinflammatory cytokines and cytolytic agents to attack and kill tumor cells (Shahabi et al., 2008). Our results showed that T cells from immunized mice were able to recognize and lyse peptide-pulsed target cells in vitro, and that LM4 hly::E7 induced robust and durable E7-specific CTL responses in tumor-bearing mice. IFN-γ is a crucial cytokine in Th1 cellmediated immunity against tumor cell proliferation (Mandai et al., 2016). The number of IFN-γ-producing T lymphocytes and the level of IFN-γ secretion in the LM4 hly::E7 immunized FIGURE 7 | Cell death and apoptosis in the tumors of tumor-bearing mice treated with rLM. Tumors were excised 7 days after the final vaccination, and processed into single cell suspensions. Cells were then stained with Annexin-V-FITC and PI, and cell apoptosis and cell death were analyzed by flow cytometry. The rates of cell death were calculated. Data are expressed as the mean and SD values from three independent experiments. **P < 0.01, Student's t-test. group were significantly higher than control groups. The strong cell-mediated response triggered by LM4 hly::E7 in a transplantable tumor model was effective for antitumor efficacy.
Interestingly, our results showed that intraperitoneally injected rLM strains entered the tumor tissues on day 1 after the first immunization and replicated. This phenomenon could be explained by the fact that the tumor microenvironment is an immunoprivileged site, which provides protection against the host immune system (Yu et al., 2004;Loeffler et al., 2009). In studies by Yang et al. found that an LM-based vaccine could stimulate CD8 + T cells from tumor infiltrating lymphocytes (TILs) and suppress the recruitment and differentiation of Tregs in the tumor microenvironment (Yang et al., 2014). We also observed by immunohistochemistry and flow cytometry that our LM-based vaccine induced the accumulation of T cells within the tumor microenvironment. Based on the above results, we speculated that LM-based vaccines may deliver foreign antigens to tumor tissues in a targeted manner, and induce antigenspecific CTLs from TILs to participate in the antitumor immune responses at local tumor sites, thus contributing to tumor regression.
Ca 2+ ion and ROS play a key role in both the initiation and effectuation of cell death (Rodriguez-Serrano et al., 2012;Nickel et al., 2016). Previously, Kim et al. reported that infection of tumor cells by an attenuated LM construct resulted in tumor cell death by activation of NAPDH oxidase and elevated levels of cytosolic ROS in vitro (Kim et al., 2009). We found that tumor cells from tumor-bearing mice immunized with rLM strains (LM4 hly::E7 and LM4 hly::esat-6) led to non-apoptotic cell death, the generation of intracellular ROS and elevated Ca 2+ concentrations in vivo. Increases in intracellular ROS and Ca 2+ levels were strongly associated with tumor cell death induced by LM-based vaccines. However, such tumor cell death induced by rLM strains was not influenced by delivery of specific antigen.
FIGURE 8 | The level of intracellular ROS and Ca 2+ in tumor cells from rLM-treated mice. Tumors were excised 7 days after the final vaccination, and processed into single cell suspensions. The cells were then stained with DCFH-DA dye assay or Fluo-3 AM assay, and the intracellular ROS (A) and Ca 2+ levels (B) were determined by flow cytometry. Data are expressed as the mean and SD values from three independent experiments. *P < 0.05; **P < 0.01, Student's t-test.
Hence, these results indicated that LM4 hly::E7 can induce antitumor efficacy via non-specific direct killing of tumor cells.
LLO is necessary for intracellular proliferation of Listeria, which is a novel adjuvant with PAMP-like activity (Birmingham et al., 2008;Neeson et al., 2008). A 19-amino-acid PEST-like sequence (KENSISSMAPPASPPASPK) of the LLO signal peptide correlates with the virulence of L. monocytogenes (Decatur and Portnoy, 2000). Previous studies in our laboratory demonstrated that 14 amino acids (SMAPPASPPASP) of the PEST-like sequence were essential for the pathogenicity of L. monocytogenes, and showed similar activity. In this study, we fused the E7 gene with hly ORF in a 42 bp deletion region. LM4 hly::E7 was highly attenuated, as indicated by the reduced LD 50 of this strain, and replicated only to a limited extent without causing severe pathology ( Figure S2). However, our macrophage infection results showed that LM4 hly::E7 could infect and grow intracellularly in RAW264.7 macrophages despite the deletion of the LLO partial sequence, a modification which is necessary for successful delivery and presentation of foreign antigen. These observations showed LM4 hly::E7 to be less virulent while retaining its ability to present delivered antigen to the immune system. Also, the recombinant strain was stable and retained the expression of the LLO-E7 fusion protein after 30 passages in vitro, confirmed by RT-PCR (data not shown). Therefore, these properties suggest potential future clinical applications for this vector.
In summary, although TC-1-induced tumors gradually grew along with the expansion of Treg suppressed cells, immunization with LM4 hly::E7 successfully induced a robust cellular immunity. LM4 hly::E7 effectively retards tumor growth in a murine model of cervical cancer. Furthermore, we observed that the antitumor mechanisms were dependent on antigenspecific CTL activity and direct killing of tumor cells. Future studies include evaluation of LM4 hly::E7 as a candidate vaccine in a transgenic mouse model, in which immunological tolerance can be addressed. Furthermore, immunomodulatory agents such as Toll-like receptor 9 (TLR9) ligands and/or radiation therapy can be combined to further enhance their antitumor efficacy for application of LM-based vaccines in clinic. Taken in its entirety, our study provides strong evidence for future clinical translation of LM-based vaccines, as well as a basis for investigation into other HPV-related cancer models.

ETHICS STATEMENT
The mice were housed, handled and immunized at the animal biosafety facilities and all procedures were approved by the institutional animal experimental committee of Yangzhou University.

AUTHOR CONTRIBUTIONS
YJ, YY, and XJ designed the experiments. WT and FD performed the experiments and analyzed the data. YY, ZP, XC, and XJ contributed reagents/materials/analysis tools. YJ, YY, and XJ wrote and revised the paper.