Immunogenicity of a Trivalent Human Papillomavirus L1 DNA-Encapsidated, Non-Replicable Baculovirus Nanovaccine

Previously, we developed a non-replicating recombinant baculovirus coated with human endogenous retrovirus envelope protein (AcHERV) for enhanced cellular delivery of human papillomavirus (HPV) 16L1 DNA. Here, we report the immunogenicity of an AcHERV-based multivalent HPV nanovaccine in which the L1 segments of HPV 16, 18, and 58 genes were inserted into a single baculovirus genome of AcHERV. To test whether gene expression levels were affected by the order of HPV L1 gene insertion, we compared the efficacy of bivalent AcHERV vaccines with the HPV 16L1 gene inserted ahead of the 18L1 gene (AcHERV-HP16/18L1) with that of AcHERV with the HPV 18L1 gene inserted ahead of the 16L1 gene (AcHERV-HP18/16L1). Regardless of the order, the bivalent AcHERV DNA vaccines retained the immunogenicity of monovalent AcHERV-HP16L1 and AcHERV-HP18L1 DNA vaccines. Moreover, the immunogenicity of bivalent AcHERV-HP16/18L1 was not significantly different from that of AcHERV-HP18/16L1. In challenge tests, both bivalent vaccines provided complete protection against HPV 16 and 18 pseudotype viruses. Extending these results, we found that a trivalent AcHERV nanovaccine encoding HPV 16L1, 18L1, and 58L1 genes (AcHERV-HP16/18/58L1) provided high levels of humoral and cellular immunogenicity against all three subtypes. Moreover, mice immunized with the trivalent AcHERV-based nanovaccine were protected from challenge with HPV 16, 18, and 58 pseudotype viruses. These results suggest that trivalent AcHERV-HPV16/18/58L1 could serve as a potential prophylactic baculoviral nanovaccine against concurrent infection with HPV 16, 18, and 58.


Introduction
Human papillomaviruses (HPV) are a heterogeneous group of double-stranded DNA viruses that cause malignant tumors of the anogenital tract, leading to cervical cancer, a common cancer accounting for approximately 12% of all cancers in women [1]. Over 100 different types of HPV, divided into low-risk and highrisk, have been identified [2]. The high-risk HPV types include HPV 16,18,31,45, and 58. Among high-risk HPV types, HPV 16 and 18 predominate, accounting for more than 70% of cervical cancers. Currently available prophylactic HPV vaccines target high-risk types such as HPV 16 and 18 [3,4]. HPV 58 is also clinically significant, especially in Asia, where it is the third-most prevalent HPV type among cervical cancers reported in Korea, Japan, and southern and eastern parts of China [5]. The larger share of disease burden of HPV 58 in Asia may reflect differences in host genetics as well as the oncogenicity of circulating variants. This unique pattern of epidemic HPV58 prevalence should be considered in the development of next-generation HPV vaccines [6].
Several experimental vaccines have been studied for their potential to generate neutralizing antibodies against HPV. Current vaccination approaches include virus-like particles, recombinant fusion proteins, recombinant fusion peptides, live recombinant bacteria and recombinant viruses [7]. DNA vaccines have received particular research attention as next-generation vaccines that may replace current subunit or live-attenuated vaccines. DNA vaccines offer several advantages compared to conventional vaccines, including relative stability and safety, capacity to induce cellmediated immune responses and ease of manipulation. Moreover, they can be created using less complex production processes and are thus less expensive to produce on a large scale. Despite these advantages and initial high hopes, research progress in this area since the first report about two decades ago has been slow, with only a few DNA vaccines reaching clinical trials to date [8,9]. One major limitation that has hampered the successful development of DNA vaccines is the intracellular delivery issue: because of their highly negative charge and large size, naked plasmid DNA cannot effectively penetrate the cell membrane [10,11].
To improve the efficacy of DNA vaccine cellular delivery, researchers have investigated various nonviral and viral vectors. Nonviral cationic liposomes [12] and polymers [13] have been studied as delivery systems for plasmid DNA vaccines, and physical methods have been applied for introducing DNA into cells [14,15]. Recombinant adenovirus [16] and vaccinia virus [17] have been investigated as delivery systems for antigenencoding DNA. Although viral vectors have advantages over nonviral vector systems in terms of intracellular delivery efficacy, they suffer from at least two major drawbacks from the standpoint of clinical development. First, most viral vectors can be converted to pathogenic forms after replication. Second, viral vectors are immunogenic, limiting repeated dosing with DNA vaccines.
Overcoming the limitations of currently studied viral vectors requires the development of new viral vectors that do not replicate in human cells, which would eliminate the potential conversion to pathogenic forms and immunogenicity, thereby allowing repeated dosing with DNA vaccines [18]. We previously developed a viral DNA vaccine against HPV 16 using recombinant baculovirus [19]. The baculovirus system is advantageous because it does not support replication in mammalian cells while increasing the efficiency of gene delivery [20][21][22][23]. To enhance delivery of the HPV L1 gene into human hosts, we previously constructed a recombinant baculovirus containing the envelope glycoprotein of human endogenous retrovirus (AcHERV). These AcHERV-based vaccines induced strong humoral and cellular immune responses in mice as well as pig models [24,25].

Humoral immune responses following administration of AcHERV-based multivalent nanovaccines
The humoral immune responses induced by bivalent and trivalent AcHERV-based nanovaccines were similar to those induced by the monovalent AcHERV-based AcHERV-HP16L1, AcHERV-HP18L1, and AcHERV-HP58L1 nanovaccines. The immunization schedules for seven groups are summarized in Table 1. Six weeks after the first administration, the levels of serum IgG antibodies specific for HPV 16 (Fig. 4A) were not significantly different among groups treated with AcHERV-HP16L1, AcHERV-HP16/18L1, AcHERV-HP18/16L1, or AcHERV-HP16/18/58L1. Similarly, the levels of serum IgG antibodies specific for HPV 18 were not significantly different among groups treated with AcHERV-HP18L1, AcHERV-HP16/18L1, AcHERV-HP18/16L1, or AcHERV-HP16/18/58L1 (Fig. 4B). Figure 4C shows that the levels of serum IgG antibodies specific for HPV58 were not significantly different between groups receiving AcHERV-HP58L1 or AcHERV-HP16/18/58L1. Because mucosal antibodies are crucial in protecting against sexually transmitted infections and form the first line of defense against such infectious agents [26], the secretory IgA response is an import marker of mucosal immunity. Following administration of AcHERV-based nanovaccines, vaginal IgAs specific for HPV 16 ( To test whether gene expression levels were affected by the order of gene insertion, we compared the efficacy of the bivalent AcHERV-based nanovaccines, AcHERV-HPV16/18L1 and AcHERV-HPV18/16L1. Regardless of the position of the two genes, bivalent AcHERV DNA vaccines induced levels of HPV16L1-specific IgG (Fig. 4A) and IgA (Fig. 5A) antibodies similar to those induced by AcHERV-HP16L1. Moreover, bivalent AcHERV DNA vaccines induced levels of HPV18L1specific IgG (Fig. 4B) and IgA (Fig. 5B) antibodies similar to those induced by AcHERV-HP18L1.

Cell-mediated immune responses induced by AcHERVbased trivalent nanovaccines
In addition to humoral immune responses, cell-mediated immune responses were induced by immunization with AcHERV-based nanovaccines (Fig. 7). Cellular immunity was determined by measuring IFN-c (Th1) and IL-4 (Th2) in splenocytes from mice stimulated with trivalent, bivalent, or monovalent AcHERV-based nanovaccines. Similar to the humoral immune response data, the levels of IFN-c (Fig. 7A) and IL-4 ( Fig. 7B) produced by stimulated splenocytes were not significantly different among groups treated with trivalent, bivalent, or monovalent AcHERV-based nanovaccines. Intramuscular injection of mice with PBS did not induce the production of IFN-c or IL-4.

Discussion
In this study, we exploited the high gene capacity of baculoviral vectors, which can deliver DNAs up to ,30 kbp in length, to construct the multivalent recombinant baculoviruses, AcHERV-HP16/18L1, AcHERV-HP18/16L1, and AcHERV-HP16/18/ 58L1, and compared their immunogenicity with those of monovalent AcHERV-HPL1s. The results obtained suggested no significant differences in transgene expression depending on the location of the gene in the baculoviral system. Importantly, bivalent (AcHERV-HP16/18L1 and AcHERV-HP18/16L1) and trivalent (AcHERV-HP16/18/58L1) AcHERV nanovaccines retained a degree of immunogenicity comparable to that of the three individual monovalent AcHERV-based nanovaccines.
Mice immunized with monovalent, bivalent, or trivalent AcHERV-HPL1 DNA vaccines exhibited similar high levels of IgG, IgA, and neutralizing antibody titers compared with monovalent AcHERV-HPL1 DNA vaccines. Among mice immunized with different AcHERV-based nanovaccines, those administered AcHERV-HP16/18/58L1 showed a greater induction of T cell immune responses, measured as IFN-c and IL-4 levels, than mice treated with monovalent or bivalent AcHERV DNA vaccines. This observation suggests that our trivalent AcHERVbased nanovaccines elicited both Th2 and Th1 immune responses.
Notably, the trivalent AcHERV-based nanovaccine provided immune responses against HPV16, 18, and 58. Although the currently available prophylactic HPV vaccines Cervarix and Gardasil induce antibodies against HPV16 and 18, they do not provoke immune responses against HPV58. A recent, clinical study has shown that Cervarix provides no cross-protective immunity for HPV 58 [27]. It has been reported that HPV 58 is detected not only in cervical cancers, but also in high-grade squamous intraepithelial lesions (HISL) in women throughout Asia, Central America, and South America [5,[28][29][30]. Indeed, HPV 58 is known to be the second-and third-most common HPV type in Asian cases of cervical cancer and HSIL, respectively [31,32]. A previous study reported that the genetic relationship between HPV 16 and 58 was closer than that between HPV 18 and 58 [27].
Understanding the important factors associated with vaccine efficacy is essential in developing and improving the next generation of vaccines. Vaccine efficacy may reflect a combination of a strong humoral response, cell-mediated response, and crossreactivity characteristics. Cross-reactivity, in particular, is an important factor in vaccine development. Our cervicovaginal HPV pseudovirus challenge study revealed that AcHERV-based nanovaccines provided complete protection from each homologous pseudovirus (Fig. 8), but virtually no cross-reactivity with heterologous PVs (Table 2). Notably, the absence of neutralizing activity against HPV 58 psedovirus challenge by bivalent AcHERV-HP16/18L1 and AcHERV-HP18/16L1 emphasizes the lack of cross-reactive immune responses among HPV 16, 18, and 58 (Table 2). Cross-reactivity of an HPV 16/18 vaccine against HPV 31, 33, and/or 45 has been previously demonstrated [33,34]; however, consistent with our observations, HPV 58 was found to show low cross-reactivity against an HPV 16/18 vaccine [35,36].
In conclusion, the strong humoral and cell-mediated responses and virtual absence of cross-reactivity that follow immunization with multivalent AcHERV-HPL1s DNA vaccines suggest that these nanovaccines represent an efficient prophylactic treatment option that could contribute to global HPV prevention. Moreover, our AcHERV system provides a platform for future DNA antigenencoding nanovaccines. Given the epidemiology of high-risk HPV types, the AcHERV-based HPV 16/18/58 trivalent nanovaccine could be clinically useful for the prevention of HPV transmission in Asia.

Ethics
Six-week-old female BALB/c mice were purchased from Orient-Bio (Seungnam, Kyonggi-do, Korea) and housed in filtertop cages, with water and food provided ad libitum. Mice were maintained in accordance with the Guide for the Care and Use of Laboratory Animals of Konkuk University (Seoul, Republic of Korea), and were housed in a Bio-safety Level 2 facility. The use of animals in these experiments was approved by the Institutional Animal Care and Use Committee of Konkuk University (Approval No. KU12078). Throughout the study, the condition of the animals was monitored twice a day. In this study, no mice exhibited symptoms of illness or appeared to be close to death. Moreover, no mice died during the monitoring phase. After final monitoring, mice were humanely euthanized using cervical dislocation according to the AVMA guidelines for the euthanasia of animals.

Construction of recombinant baculoviruses
Recombinant baculoviruses were produced using the Bac-to-Bac baculovirus expression system (Invitrogen) according to the manufacturer's instructions. The recombinant baculovirus encoding the HPV 16L1 sequence ahead of 18L1 was designated AcHERV-HP16/18L1; the recombinant baculovirus encoding the HPV 18L1 sequence ahead of 16L1 was designated AcHERV-HP18/16L1; and the recombinant baculovirus encoding the HPV 16L1 sequence followed by 18 L1 and 58 L1 was designated AcHERV-HP16/18/58L1. Recombinant AcHERV-HP16/18L1, AcHERV-HP18/16L1, and AcHERV-HP16/18/58L1 baculoviruses were further amplified by propagation in Sf9 cells. Baculoviruses were purified by first centrifuging at 20006 g at 4uC for 10 minutes to remove virus-infected cell debris, after which supernatants were overlaid on a 30% sucrose cushion and centrifuged at 35,000 rpm at 4uC for 1.5 hour in a 50.2Ti rotor (Beckman Coulter Inc., Brea, CA, USA). The pellet was resuspended in phosphate-buffered saline (PBS) and used for immunization.

Western blotting
Western blotting was used to test the expression of HPV 16L1, 18L1, and 58L1 proteins in cells after delivery of the corresponding AcHERV-HPL1s. 293TT cells were treated with each recombinant baculovirus at a MOI of 100. Three days after infection, cells were lysed with lysis buffer, and proteins in lysates were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes. HPV L1 protein was detected by first incubating the membranes with primary anti-HPV16 L1 antibody (Camvir-1; Santa Cruz Biotechnology, Santa Cruz, CA, USA), and polyclonal   . Induction of HPV type-specific neutralizing antibodies after immunization with AcHERV-based nanovaccines. Serum was sampled at 2, 4, 6 weeks after intramuscular administration with PBS or AcHERV-based nanovaccines. Neutralization assays were performed with HPV 16 (A) or HPV 18 (B) or HPV 58 (C) pseudoviruses. Data are expressed as the genometric means (log) of reciprocal serum dilutions that yielded a 50% reduction in SEAP. * p,0.05 (ANOVA and Student-Newman-Keuls test): Bivalent and trivalent groups were compared with each monovalent group. doi:10.1371/journal.pone.0095961.g006 mouse sera specific for HPV 18 or 58 prepared in our laboratory. An anti-b-actin antibody (1:2,000 dilution; Santa Cruz Biotechnology) was used as a protein loading control.

Immunization of mice with AcHERV-HPL1s
Six-week-old female BALB/c mice were immunized by intramuscular injection into the hind legs with 1610 7 plaqueforming units (PFU) of AcHERV-HPL1s. Sixty BALB/c mice were divided into seven groups (Table 1). Mice were immunized three times with AcHERV-HPL1s at 2-week intervals. Sera and vaginal washes from each group were obtained 0, 2, 4, and 6 weeks after the first immunization. Blood samples and vaginal washings were collected after anesthetizing the mice by intramuscular injection with 40mg/kg of Zoletil50 (Virbac Laboratories, Carros, France), and 51 mg/kg of Rompun (Bayer Korea, Seoul, Republic of Korea). Blood was sampled from right external jugular vein, and vaginal samples were collected by rinsing out the vaginal cavity with 50 ul of PBS 5 times. Vaginal washing samples were then centrifuged at 13,000 rpm for 10 min, and the supernatants were used for further assays.

Enzyme-linked immunosorbent assay
Production of HPV 16-, 18-, and 58L1-specific antibodies was tested by enzyme-linked immunosorbent assay (ELISA) using pseudoviruses as coating antigens. Sixty microliters of pseudovirus (0.001 mg/ml) was added to each well of a 96-well plate and incubated for 16 hours at 4uC. After washing, plates were blocked with 2% (w/v) bovine serum albumin in PBS containing 0.1% Nonidet P-40 (Sigma). Serially diluted mouse sera or vaginal washings (60 ul/well) were added and incubated at room temperature for 2 hours. After washing, peroxidase-conjugated goat anti-mouse IgG antibody (1:2000; Santa Cruz Biotechnology) or goat anti-mouse IgA antibody (1:1000; Santa Cruz Biotechnology) was added. For color development, 1-Step Turbo TMB (3,39,5,59-tetramethyl benzidine substrate solution; Pierce, Rockford, IL, USA) was added. Endpoint titers were defined as the highest serum dilutions that resulted in an absorbance value twice that of non-immunized serum (cutoff value, 0.1) and were expressed as group means 6 SDs.

Neutralization assay
Neutralizations assays were performed using SEAP-expressing pseudoviruses (PV16, PV18, and PV58) according to a previously described method [24]. Briefly, OptiPrep-purified, SEAP-expressing HPV 16, 18, and 58 pseudoviruses were diluted 3,000-fold and incubated on ice for 1 hour with 3-fold serial dilutions of mouse sera. 293TT cells were infected by incubating with each pseudovirus-antibody mixture for 3 days. The SEAP content in 10 ul of clarified cell supernatant was determined using a Great EscAPe SEAP Chemiluminescence Kit (Clontech, Mountain View, CA, USA). Neutralization titers were defined as the reciprocal of the highest serum dilution that caused at least a 50% reduction in SEAP activity.

IFN-c/IL-4 enzyme-linked immunospot assay
The production of interferon (IFN-c) and interleukin 4 (IL-4) from splenocytes of immunized mice was detected by enzymelinked immunospot (ELISPOT) assay. A 96-well plate was coated with 0.2 ug of anti-mouse IFN-c and anti-mouse IL-4 capture antibodies, and then blocked by incubating with 10% FBS at 37uC. Splenocytes were seeded at 1610 6 cells per well in 100 ul of medium, and stimulated by adding 1610 6 HPV pseudoviruses and incubating for an additional 24 hours at 37uC. Plates were then washed with PBS containing 0.05% Tween-20 and treated with 20 ng of biotinylated anti-mouse IFN-c and anti-mouse IL-4 detection antibodies. After 2 hours, streptavidin-alkaline phosphatase was added to the wells, and color was developed with an AEC substrate reagent (BD Biosciences, Franklin Lakes, NJ, USA). The number of spots was counted using an ELISPOT reader (AID Elispot Reader ver. 4; Strassberg, Germany).

Challenge test with HPV pseudoviruses
Six weeks after the final immunization with AcHERV-HPL1s (AcHERV-HP16L1, AcHERV-HP18L1, AcHERV-HP58L1, AcHERV-HP16/18L1, ACHERV-HP18/16L1, or AcHERV-HP16/18/58L1), mice were challenged with HPV pseudoviruses, as described previously [24] Seven days before in vivo genital challenge with pseudoviruses, mice were synchronized in a diestrus-like status by subcutaneous injection of 3 mg DepoProvera (Pfizer AG, Zurich, Switzerland). Six hours prior to pseudovirus challenge, mice were deeply anesthetized by intra-muscular injection with 40 mg/kg of Zoletil50 (Virbac Laboratories), and 5 mg/kg of Rompun (Bayer Korea), then pretreated intravaginally with 20 ul of 4% nonoxynol-9 (Sigma). Mice were genitally challenged with 1610 7 IFU of HPV 16, 18, and 58 pseudoviruses, each in a 20 ul solution containing 2% carboxymethylcellulose (Sigma). Three days later, all mice were anesthetized by intramuscular injection with 40 mg/kg of Zoletil50 (Virbac Laboratories), and 5 mg/kg of Rompun (Bayer Korea). Anesthetized mice were injected intraperitoneally with luciferin (30 ul at 7 mg/ml) to detect luciferase expressed upon pseudoinfection by pseudoviruses encapsidating pLucf, a plasmid carrying the luciferase gene [http://home.ccr.cancer.gov/lco/]. The expression of luciferase was detected by measuring light emission over 10 minutes with an IVIS 200 bioluminescence imaging system (Xenogen, Cranbury, NJ, USA). Equal-sized areas encompassing the site of virus inoculation were analyzed using Living Image 2.20 software (Xenogen) [37]. Due to the lack of pathogenicity of pseudotype HPV virus, all the pseudoviruschallenged mice survived until the end of the study, and did not show any signs of illness such as ruffled fur, decreased activity, and weight loss regardless of vaccination. After final monitoring, all the challenged mice were humanely euthanized using cervical  Table 2. Reactivity of AcHERV-based nanovaccines against HPV16, 18, and 58.

Statistical analysis
All data were analyzed by analysis of variance (ANOVA) with Student-Newman-Keuls post hoc tests using SigmaStat software (Systat Software, Richmond, CA, USA). P-values less than 0.05 were considered significant.