Optimization of mucosal responses after intramuscular immunization with integrase defective lentiviral vector

Abstract

Many infectious agents infiltrate the host at the mucosal surfaces and then spread systemically. This implies that an ideal vaccine should induce protective immune responses both at systemic and mucosal sites to counteract invasive mucosal pathogens. We evaluated the in vivo systemic and mucosal antigen-specific immune response induced in mice by intramuscular administration of an integrase defective lentiviral vector (IDLV) carrying the ovalbumin (OVA) transgene as a model antigen (IDLV-OVA), either alone or in combination with sublingual adjuvanted OVA protein. Mice immunized intramuscularly with OVA and adjuvant were compared with IDLV-OVA immunization. Mice sublingually immunized only with OVA and adjuvant were used as a positive control of mucosal responses. A single intramuscular dose of IDLV-OVA induced functional antigen-specific CD8+ T cell responses in spleen, draining and distal lymph nodes and, importantly, in the lamina propria of the large intestine. These results were similar to those obtained in a prime-boost regimen including one IDLV immunization and two mucosal boosts with adjuvanted OVA or vice versa. Remarkably, only in groups vaccinated with IDLV-OVA, either alone or in prime-boost regimens, the mucosal CD8+ T cell response persisted up to several months from immunization. Importantly, following IDLV-OVA immunization, the mucosal boost with protein greatly increased the plasma IgG response and induced mucosal antigen-specific IgA in saliva and vaginal washes. Overall, intramuscular administration of IDLV followed by protein boosts using the sublingual route induced strong, persistent and complementary systemic and mucosal immune responses, and represents an appealing prime-boost strategy for immunization including IDLV as a delivery system.

Figure 1. Expression of ovalbumin from IDLV.

Western blot analysis of cell lysates and supernatants from 293T cells transduced with IDLV-OVA or IDLV-GFP as negative control, using an anti-OVA polyclonal antibody. Ovalbumin protein (OVAp, predicted size 47 kD) was used as positive control. Samples included supernatants showing secreted OVA (lane 1 and lane 2, 14 µl and 7 µl from 293T cells cultured at 5×10⁵/ml, respectively), and cell lysates (lane 3 and lane 4, corresponding to 3×10⁵ and 0.75×10⁵ cells equivalent, respectively). Supernatants (lane 5, 14 µl) and cell lysates (lane 6, 3×10⁵ cells equivalent) from IDLV-GFP transduced cells did not produce OVA. OVA protein (OVAp) at indicated amounts was used as positive control.

Figure 2. Analysis of systemic antigen-specific CD8+ T cell response at the peak of immune response.

Two weeks after the final immunization, animals from all groups (vaccination regimens are described in Table 1) were sacrificed and cells from different sites were used to perform INFγ ELISPOT. Splenocytes (A) and lymphocytes from draining and distal lymph nodes (B) were stimulated overnight with medium alone (blank bars) or with H-2K^b restricted OVA-specific 8mer peptide (SIINKFEL) (filled bars). Results are expressed as mean IFNγ secreting cells (measured as spot forming cells)/10⁶ cells presented as group means ± standard deviations. ING, SM and MES LN: inguinal, submandibular and mesenteric lymph nodes, respectively. The asterisks specify statistically significant differences (p<0.05) between groups indicated within the graph.

Figure 3. Frequency of mucosal antigen-specific CD8 T cells.

Two weeks after the final immunization, mice from all groups (vaccination regimens are described in Table 1) were sacrificed. Lymphocytes derived from large intestine lamina propria were stained with fluorescent H-2K^b-SIINKFEL dextramers, anti-mouse CD3 and anti-mouse CD8 and analyzed by FACScalibur. The analysis was performed on gated CD3+CD8+ cells from immunized or naïve mice. Results are expressed as percentage of CD3+CD8+ dextramers+ cells presented as group means ± standard deviations.

Figure 4. Analysis of antigen-specific antibodies in plasma and mucosal secretions.

(A) Kinetics of plasma anti-OVA IgG titers in mice belonging to different groups (vaccination regimens are described in Table 1) at 2 weeks after each immunization. Results are expressed as mean titer presented as group means ± standard deviations. The statistical analysis is described and discussed in the text. (B) Analysis of anti-OVA IgA titer in saliva and vaginal washes (VWs) collected 2 weeks after the final immunization. Results are expressed as mean titer presented as group means ± standard deviations.

Figure 5. Persistence of systemic antigen-specific CD8+ T cell response.

Six months after the last immunization mice from all groups (vaccination regimens are described in Table 1) were sacrificed and splenocytes used for the analysis of OVA-specific T cells. (A) IFNγ ELISPOT. Splenocytes were stimulated overnight with medium alone (blank bars) or with H-2K^b restricted OVA-specific 8mer peptide (SIINFEKL) (filled bars). IFNγ-producing T cells are expressed as the number of spot forming cells per 10⁶ cells. Results are presented as group means ± standard deviations. The asterisks indicate statistically significant differences (p<0.05) between indicated groups. (B) Analysis of multifunctional antigen-specific CD8+ T lymphocytes by intracellular assay for IFNγ and TNFα production. A representative experiment is shown. The analysis was performed on gated CD3+CD8+ cells from immunized or naïve mice. The percentages of single or double-cytokine producing cells were calculated and are indicated within the dot plots.

Figure 6. Persistence of mucosal antigen-specific CD8+ T cell response.

Six months after the last immunization mice from all groups (vaccination regimens are described in Table 1) were sacrificed. (A) Mesenteric lymph node-derived lymphocytes were stained with H-2K^b-SIINKFEL dextramers, anti-mouse CD3 and anti-mouse CD8. Results are expressed as percentage of CD3+CD8+ dextramers+ cells presented as group means ± standard deviations. The asterisks indicate statistically significant differences (p<0.05) between indicated groups. (B) Lymphocytes derived from large intestine lamina propria (LP) of mice from indicated group were stained with fluorescent H-2K^b-SIINKFEL dextramers, anti-mouse CD3 and anti-mouse CD8. The analysis was performed on gated CD3+CD8+ cells from immunized or naïve mice. Results are expressed as percentage of CD3+CD8+ dextramers+ cells presented as group means ± standard deviations. The asterisks indicate statistically significant differences (p<0.05) between indicated groups.

Figure 7. Persistence of IDLV in the immunized mice.

(A) PCR analysis for evaluation of vector presence in DNA extracted from indicated tissue samples at 6 months from IDLV-OVA injection. The 293 cell line stably transduced with the TY2-GFP-IRES-Neo vector (293/LV-Neo) was used as standard for evaluating vector presence, as already described . DNA quality and integrity of all samples was evaluated by PCR amplification of GAPDH on 200 ng of DNA. PCR samples were run on a 2% agarose gel. G3PDH, glyceraldehyde 3-phosphate dehydrogenase; NT, muscle from naïve mice. Persistence of anti-OVA IgG in plasma. (B) Anti-OVA IgG titer in plasma samples collected from immunized mice at 6 months after the final immunization (vaccination regimens are described in Table 1). Results are expressed as mean titer presented as group means ± standard deviations.