Integrase Defective Lentiviral Vector as a Vaccine Platform for Delivering Influenza Antigens

Abstract

Viral vectors represent an attractive technology for vaccine delivery. We exploited the integrase defective lentiviral vector (IDLV) as a platform for delivering relevant antigens within the context of the ADITEC collaborative research program. In particular, Influenza virus hemagglutinin (HA) and nucleoprotein (NP) were delivered by IDLVs while H1N1 A/California/7/2009 subunit vaccine (HAp) with or without adjuvant was used to compare the immune response in a murine model of immunization. In order to maximize the antibody response against HA, both IDLVs were also pseudotyped with HA (IDLV-HA/HA and IDLV-NP/HA, respectively). Groups of CB6F1 mice were immunized intramuscularly with a single dose of IDLV-NP/HA, IDLV-HA/HA, HAp alone, or with HAp together with the systemic adjuvant MF59. Six months after the vaccine prime all groups were boosted with HAp alone. Cellular and antibody responses to influenza antigens were measured at different time points after the immunizations. Mice immunized with HA-pseudotyped IDLVs showed similar levels of anti-H1N1 IgG over time, evaluated by ELISA, which were comparable to those induced by HAp + MF59 vaccination, but significantly higher than those induced by HAp alone. The boost with HAp alone induced an increase of antibodies in all groups, and the responses were maintained at higher levels up to 18 weeks post-boost. The antibody response was functional and persistent overtime, capable of neutralizing virus infectivity, as evaluated by hemagglutination inhibition and microneutralization assays. Moreover, since neuraminidase (NA)-expressing plasmid was included during IDLV preparation, immunization with IDLV-NP/HA and IDLV-HA/HA also induced functional anti-NA antibodies, evaluated by enzyme-linked lectin assay. IFNγ-ELISPOT showed evidence of HA-specific response in IDLV-HA/HA immunized animals and persistent NP-specific CD8+ T cell response in IDLV-NP/HA immunized mice. Taken together our results indicate that IDLV can be harnessed for producing a vaccine able to induce a comprehensive immune response, including functional antibodies directed toward HA and NA proteins present on the vector particles in addition to a functional T cell response directed to the protein transcribed from the vector.

Keywords: T cell response; antibody; influenza; lentiviral vector; vaccine.

Figure 1

Analysis of hemagglutinin (HA) expression. 293 T Lenti-X cells were transfected with HA expressing plasmid and stained at 24 h post-transfection for detection of HA on the plasma membrane as described in Section “Materials and Methods.” Cells were fixed and expression of HA was quantitatively measured by flow cytometry (A) or observed by confocal laser scanning microscopy (B). (A) The percentage of HA-expressing cells is indicated. The overlay line (green) represents the fluorescence distribution of cells stained only with the secondary antibody. (B) Images represent single central optical sections (a–c) and a 3D reconstruction (d). Nuclei are colored in blue by DAPI staining and green color represents membrane associate HA protein. Scale bars, 8 µm. Untransfected cells (a) were used as negative control. Results from one representative experiment are shown for each analysis.

Figure 2

Production and validation of integrase defective lentiviral vector (IDLV) pseudotyped with hemagglutinin (HA). (A) Reverse transcriptase (RT) activity of IDLV expressing HA. Vectors were produced as described in Section “Materials and Methods” and in the presence of plasmids expressing transmembrane protease serine 2 (TMPRSS2), human airway trypsin (HAT), or NACal09, as indicated in the graph. Data are expressed as the mean result from three independent experiments. The error bars represent the standard errors of the mean. (B) Western blot (WB) of lysates from concentrated stocks of IDLV pseudotyped with HA showing incorporation of HA into IDLV. Note that IDLVs were produced in the presence of TMPRSS2 protease, resulting in the cleavage of HA0 to produce HA1 (not visualized here) and HA2. HA protein (HAp*, purified hemagglutinin vaccine subunit from influenza virus H1N1 A/California/7/2009) and IDLV-GFP were used as positive and negative control, respectively.

Figure 3

Immunization schedule and kinetics of anti-H1N1 binding antibodies (Abs). (A) Immunization schedule. CB6F1 mice (four mice/group) were primed once intramuscularly with integrase defective lentiviral vector (IDLV) expressing hemagglutinin (HA) and pseudotyped with HA (IDLV-HA/HA), IDLV expressing NP and pseudotyped with HA (IDLV-NP/HA), HA protein (*purified hemagglutinin vaccine subunit from influenza strain H1N1 A/California/7/2009) in combination with MF59 as an adjuvant (HAp + MF59), HA protein alone (HAp) or left untreated (Naive). All groups except for naive were boosted with HAp at 24 weeks after the prime. Blood was collected at several time points in order to perform ELISA and IFNγ ELISPOT assays. (B) Kinetics of serum anti-H1N1 IgG Abs. Serum samples from all groups were collected at the indicated time points after the prime and were assayed for the presence of anti-H1N1 IgG by ELISA. Results are expressed as predicted mean endpoint titers. The predicted mean values at each time point were estimated by a system of piecewise linear regressions including all antibody measurements within a structural equation modeling (SEM) frame work. Error bars indicate the 95% confidence interval. Asterisks indicate significant differences between groups at all the analyzed time points; **p-value <0.01.

Figure 4

Kinetics of functional antihemagglutinin antibodies (Abs). Serum samples from all groups were collected at the indicated time points after the prime and were assayed for the presence of neutralizing Abs against A/California/7/09 virus, by hemagglutination inhibition assay (A) and microneutralization assay (B). Results are expressed as predicted mean endpoint titers. The predicted mean values at each time point were estimated by a system of piecewise linear regressions including all antibody measurements within a structural equation modeling frame work. Error bars indicate the 95% confidence interval. Asterisks indicate significant differences compared to the HAp group, at the indicated time points; **p < 0.01; *p < 0.05.

Figure 5

Kinetics of functional anti-neuraminidase (anti-NA) antibodies (Abs). Serum samples from all groups were collected at the indicated time points after the prime and were assayed for the presence of anti-NA Abs by ELLA assay (see Materials and Methods). Results are expressed as predicted mean endpoint titers. The predicted mean values at each time point were estimated by a system of piecewise linear regressions including all antibody measurements within a structural equation modeling frame work. Error bars indicate the 95% confidence interval. Asterisks indicate significant differences compared to the HAp group at all the analyzed time points; **p–value <0.01.

Figure 6

Analysis of antigen-specific T cells response. (A) Kinetics of nucleoprotein (NP)-specific T cell response in mice immunized with IDLV-NP/HA. IFNγ ELISPOT was performed using blood cells collected at 4, 12, and 24 weeks post-prime and splenocytes collected at 42 weeks post-prime. Results are expressed as mean spot forming cells (SFC) per 10⁶ cells. Cells were stimulated overnight with the MHC I-restricted epitope derived from NP protein sequence (black bars) or left untreated (white bars). Error bars indicate the SD among mice of the same group. (B) Analysis of HA-specific T cell response in immunized mice. IFNγ ELISPOT was performed using splenocytes collected at 42 weeks post-prime, using 10 pools of 15mers spanning the full length of HA protein, as described in Section “Materials and Methods.” Results are expressed as cumulative mean SFC per 10⁶ cells for each pool among mice of the same indicated group.