Intranasal vaccination with recombinant adeno-associated virus type 5 against human papillomavirus type 16 L1

Abstract

Adeno-associated viruses (AAV) have been developed and evaluated as recombinant vectors for gene therapy in many preclinical studies, as well as in clinical trials. However, only a few approaches have used recombinant AAV (rAAV) to deliver vaccine antigens. We generated an rAAV encoding the major capsid protein L1 (L1h) from the human papillomavirus type 16 (HPV16), aiming to develop a prophylactic vaccine against HPV16 infections, which are the major cause of cervical cancer in women worldwide. A single dose of rAAV5 L1h administered intranasally was sufficient to induce high titers of L1-specific serum antibodies, as well as mucosal antibodies in vaginal washes. Seroconversion was maintained for at least 1 year. In addition, a cellular immune response was still detectable 60 weeks after immunization. Furthermore, lyophilized rAAV5 L1h successfully evoked a systemic and mucosal immune response in mice. These data clearly show the efficacy of a single-dose intranasal immunization against HPV16 based on the recombinant rAAV5L1h vector without the need of an adjuvant.

FIG. 1.

Vector constructs and characterization of viral stocks. (a) AAV vector constructs. The humanized L1 gene (L1h) of HPV16 or the Gfp gene, respectively, was inserted between the two inverted terminal repeats (ITRs) of AAV2. (b) Western blot analysis of AAV5 capsid proteins. A sample of the purified vector stock (5 × 10¹¹ gp) was analyzed by Western blot analysis with the polyclonal rabbit serum VP51. The three capsid proteins VP1, VP2, and VP3 are shown. (c) Western blot analysis of L1 protein in the vector stock. A total of 5 × 10¹¹ gp of the final vector stock were analyzed by using the HPV16 L1-specific antibody Camvir-1. Two concentrations of L1-VLP were loaded as positive controls. (d) Control of transgene expression. HeLa cells were infected with AAV5 L1h (MOI = 100) and superinfected with Ad5 (MOI = 10). Cells were harvested 48 h postinfection, and lysates were analyzed for HPV16 L1 expression by using Camvir-1 antibody.

FIG. 2.

HPV16L1-specific antibody titers after 3 months. Five mice per group were immunized intranasally with AAV5 L1h (5 × 10¹⁰ gp/dose) or AAV5 Gfp (5 × 10¹⁰ gp/dose). Positive control mice were vaccinated with L1-VLP (5 μg/dose) or L1-capsomeres (10 μg/dose), respectively. The negative control group was treated with AAV5 Gfp or PBS alone. Mice received after the initial dose (AAV5 L1h single) a second immunization after 7 days (AAV5 L1h double) and a booster dose after 42 days (AAV5 L1h triple). The AAV5 L1h lyo- and AAV5 Gfp-treated groups received three doses, and AAV5 L1h high-vaccinated mice received three doses with 5 × 10¹¹ gp/dose. After 3 months, sera (a) and vaginal washes (see Table 1) were analyzed for the presence of HPV16 L1-specific IgGs by using a VLP-based ELISA. The specific IgG titers were expressed as the reciprocal of the highest dilution that yielded an optical density at 450 nm four times that of control mice (PBS group). Presented are the geometric means of five mice per group and the standard deviation indicated as error bars. (b) Individual sera were also tested for neutralization of 293T cell infection with pseudovirions. Mean percent neutralizations obtained by incubation with a 1:50 dilution of the sera of the different groups are presented. Neutralization activity was compared to a high-titer rabbit anti-L1 antiserum (r#4543), which gave a neutralization average of 78%.

FIG. 3.

Development of antibody titers over 60 weeks. Mice were immunized intranasally with single, double, or triple doses (5 × 10¹⁰ gp/dose) of AAV5 L1h. Control mice were vaccinated with L1-VLP (5 μg/dose) or with PBS alone. Time points of immunizations are indicated by arrowheads. At different time points, sera were collected and stored at −20°C. After 1 year, HPV16 L1-specific IgG in all samples was analyzed by using a VLP-based ELISA. The specific IgG titers were expressed as the reciprocal of the highest dilution that yielded an optical density at 450 nm four times that of control mice (PBS group). Presented are the geometric means of five mice per group and the standard deviations indicated as error bars.

FIG. 4.

Cellular immune response against HPV16 L1 after 60 weeks. Mice were immunized intranasally with single, double, or triple doses (5 × 10¹⁰ gp/dose) of AAV5 L1h or a triple dose of AAV5 Gfp or with L1-VLP (5 μg/dose). The initial dose was given at day 1 (single), the second dose was given at day 7 (double), and the third dose was given at day 42 (triple). Splenocytes were isolated and used in the IFN-γ-ELISPOT assay. The spot number of HPV16 L1-specific IFN-γ producing splenocytes/total splenocytes was determined. The geometric means of five mice per group (the AAV5 L1h triple group contained four animals) and the standard deviations indicated as error bars are presented.

FIG. 5.

AAV-directed antibody titers after 3 months. Mice were immunized intranasally with AAV5 L1h (5 × 10¹⁰ for single, double, or triple immunizations; 5 × 10¹¹ gp/dose for high); or AAV5 Gfp (5 × 10¹⁰ gp/dose), L1-VLP (5 μg/dose), L1-capsomeres (10 μg/dose), or PBS alone as described in Fig. 2. After 3 months, sera (a) and vaginal washes (b) were analyzed for the presence of AAV5 specific antibodies by using an AAV5 capsid-based ELISA. The specific IgG titers were expressed as the reciprocal of the highest dilution that yielded an optical density at 450 nm four times that of control mice (PBS group). Presented are the geometric means of five mice per group and the standard deviations indicated as error bars. (c) Readministration of AAV5 L1h. Groups of five mice were preimmunized intranasal with empty capsids of AAV2, AAV4, or AAV5. AAV-specific seroconversion was confirmed by ELISA after 6 weeks, and mice were also given a single dose of AAV5 L1h intranasally (5 × 10¹⁰ gp/dose). Six weeks later, L1-specific seroconversion was determined by ELISA. Readministration was successful in the presence of AAV2- or AAV4-specific humoral immune responses but not if AAV5-specific antibodies were present.

FIG. 6.

In vivo vector distribution of AAV vectors after intranasal application. At 3 months after the first immunization, mice were sacrificed and heart, liver, kidney, and lung tissues were excised and immediately frozen in liquid nitrogen. Genomic DNA was isolated, and L1-specific primers were used for PCR amplification of a 339-bp fragment. Presented are the results of mice immunized three times with AAV5 L1h (5 × 10¹¹ gp/dose). Mice 1 and 3 revealed L1h vector DNA only in their lung tissue.