A replication-incompetent adenoviral vector encoding for HSV-2 gD2 is immunogenic and protective against HSV-2 intravaginal challenge in mice

Abstract

Herpes Simplex virus (HSV) is the cause of genital herpes and no prophylactic treatment is currently available. Replication-incompetent adenoviral vectors are potent inducers of humoral and cellular immune responses in humans. We have designed an adenoviral vector type 35 (Ad35)-based vaccine encoding the HSV-2 major surface antigen gD2 (Ad35.HSV.gD2). Immunization of mice with Ad35.HSV.gD2 elicited virus neutralizing antibody titers (VNT) and cellular responses against HSV-2 and HSV-1. While immunity was lower than for CJ2-gD2, both vaccines showed 100% survival against intravaginal challenge with HSV-2 G strain and a strong inverse correlation was observed between HSV-2 infection (as measured by viral shedding) and VNT. A combination of Ad35.HSV.gD2 with Ad35 encoding for gB2 (Ad35.HSV.gB2) resulted in increased VNT and lower infection, compared with Ad35.HSV.gD2 alone. Transfer of immune serum into naïve BALB/c mice before intravaginal challenge confirmed the role of antibodies in the protection of mice against infection although other immune factors may play a role as well.

Fig 1. Immunization with Ad35.HSV.gD2 induces HSV-2 specific humoral immune responses in BALB/c mice.

Female BALB/c mice were immunized intramuscularly once (T = 0d) or twice (T = 0d and T = 28d) with 10⁸ to 10¹⁰ vp/mouse Ad35.HSV.gD2 (N = 5/group), twice with 7x10⁵ pfu/mouse CJ2-gD2 (N = 5) (T = 0d and T = 28d) or mock immunized with formulation buffer (N = 5) and sacrificed 6 weeks after the first immunization. Serum samples were collected (T = 42d) and analyzed for (a) HSV-2 Virus Neutralizing antibody titers (shown as log₂ IC₅₀) and (b) gD2 IgG ELISA titers (shown as log10 EU/ml). (c) Cross-reactive neutralizing antibody titers against HSV-1 KOS virus (shown as log₂ IC₅₀) and (d) gD1 IgG ELISA titers (shown as log₁₀ endpoint titers) were only measured in sera from mice receiving 2-dose regimens. Dotted lines indicate the lower limit of quantification (LLoQ, gD2 ELISA) or limit of detection (LOD, other assays) and horizontal lines indicate the mean value per group. (a-b) (Across dose) statistical comparison and (c-d) highest dose comparison was performed by ANOVA with Bonferroni correction. Results of statistical analysis are depicted by asterisks: *p<0.05, ** p<0.01, *** p<0.001; ns: not significant.

Fig 2. Immunization with Ad35.HSV.gD2 induces gD2 IgG1 and IgG2a subclass ELISA titers.

Female BALB/c mice were immunized as described in Fig 1. (a) IgG2a and (b) IgG1 subclass antibodies against the gD2 protein were assessed in serum by ELISA (T = 42d) (shown as log₁₀ endpoints titers). (c) Log₁₀ IgG2a/IgG1 ELISA titer ratio is shown. Dotted lines indicate the limit of detection (LOD) and horizontal lines indicate the mean response per group. Across dose level comparison between 1- and 2-dose regimens was performed by ANOVA and comparison between highest vaccine dose levels were performed by ANOVA with Bonferroni correction. Results of statistical analysis are depicted by asterisks: *p<0.05, *** p<0.001; ns: not significant.

Fig 3. Prime and prime boost immunization with Ad35.HSV.gD2 induces HSV-2 specific IFNγ responses.

Female BALB/c mice were immunized as described in Fig 1. Cellular immune responses were assessed by IFN-γ ELISpot on splenocytes harvested from BALB/c mice (T = 42d). (a) Ad35.HSV.gD2 splenocytes were stimulated with a gD2 peptide pool overnight in R10 medium while (b) CJ2-gD2 splenocytes were stimulated in R0 medium with a gD2 peptide pool, gB2 peptide pools (gB-1, gB-2 and gB-3), ICP27 (13-mer: AFLTGADRSVRLA) and VP11/12 (9-mer: HGPSLYRTF). Results are depicted as spot forming units (SFU)/10⁶ splenocytes. Dotted lines indicate the background of the assay (95th percentile of medium stimulated samples) and the horizontal lines depict the geometric mean response per group. Across dose statistical comparison between the one- and 2-dose regimens was performed by ANOVA. Results of statistical analysis are depicted by asterisks: ** p<0.01.

Fig 4. Prime and prime-boost immunization with Ad35.HSV.gD2 induces complete protection and reduced viral shedding in BALB/c mice after HSV-2 intravaginal challenge.

Female BALB/c mice were immunized intramuscularly once at T = 0d or twice (T = 0d and T = 28d) with 10¹⁰ vp/mouse Ad35.HSV.gD2 (N = 10/group), twice with 7x10⁵ pfu/mouse CJ2-gD2 (N = 10) or mock immunized with formulation buffer (N = 10). Mice were intravaginally challenged 6 weeks after the first immunization (T = 42d) with 200LD50 (3.2x10⁴ pfu/mouse) WT HSV-2 G strain and monitored daily for (a) survival and (b) clinical scores over 21 days after infection. (c) Viral shedding was assessed by plaque assay on vaginal swabs collected on days 1, 2 and 4 after challenge (viral titers expressed as log₁₀ plaque forming units (pfu)/mL). HSV-2 VNT (titer expressed as log₂ IC₅₀), gD2 ELISA and gB2 ELISA (titer expressed as log₁₀ EU/mL) were performed on sera collected 1 day before challenge and correlation analyses were performed between (d) HSV-2 VNT vs viral shedding (Area Under the Curve, AUC) per individual animal on Day 1, 2, 4 (Day 1–4), (e) HSV-2 VNT vs gD2 ELISA titers and (f) HSV-2 VNT vs the sum of gD2 and gB2 ELISA titers. Dotted lines indicate the limit of detection (LOD) and horizontal lines represent mean value per group. Statistical comparisons of viral shedding between immunized groups (c) were performed by a pairwise Wilcoxon test with a 6-fold Bonferroni adjustment. Results of statistical analysis are depicted by asterisks: **p<0.01, *** p<0.001, ns: not significant. Correlation analyses were performed using the Spearman rank correlation method with the correlation coefficient and two-tailed p value for each analysis depicted in the figure.

Fig 5. Mice receiving Ad35.HSV.gD2 and CJ2-gD2 immune serum show partial protection and viral shedding after challenge.

Mice (N = 10/group) received three dilutions (High, Mid, Low) of serum pools from Ad35.HSV.gD2 or CJ2-gD2 immunized mice 3 and 2 days before intravaginal challenge (0.5 ml serum/mouse/day). Serum pools were diluted to equal gD2 ELISA titers before transfer. Control mice received naïve serum pool or active immunization with Ad35.HSV.gD2 (10¹⁰ vp/mouse) 42 and 14 days before challenge. Mice were challenged intravaginally with 200LD₅₀ HSV-2 G strain on day 0. After challenge mice were monitored daily for (a) survival for 21 days. One day before challenge, serum was collected and HSV-2 VNT and gD2 ELISA titers were measured. Logistic regression analysis was performed using (b) HSV-2 VNT (log₂ IC₅₀) versus survival status (0 = dead at day 21; 1 = alive at day 21) and (c) gD2 ELISA titers (log₁₀ EU/mL) versus survival status. Thick black and grey lines depict the regression curves for Ad35.HSV.gD2 and CJ2-gD2, respectively. Thin black lines and thin dotted lines depict the 95% confidence intervals for Ad35.HSV.gD2 and CJ2-gD2, respectively.

Fig 6. Inverse correlation between HSV-2 VNT and viral shedding in mice receiving immune serum before intravaginal challenge.

Mice were passively immunized and challenged intravaginally with HSV-2 as described in Fig 5. Correlation analysis between (a) HSV-2 VNT vs the AUC (Area Under the Curve) viral titer (log₁₀) after challenge (Day1-4) are shown. Dotted line depicts the lower limit of quantification (LLOQ). Correlation plot was generated by Spearman rank correlation method resulting in r = -0.3411 (**p<0.01).

Fig 7. Immunization with a combination of Ad35.HSV.gD2 and Ad35.HSV.gB2 shows enhanced protection compared with individual components after HSV-2 intravaginal challenge.

Female BALB/c mice were immunized intramuscularly once at T = 0d with Ad35.HSV.gD2, Ad35.HSV.gB2 (10¹⁰ vp/mouse, N = 10/group), Ad35.HSV.gD2/ Ad35.HSV.gB2 (10¹⁰ vp/vector, N = 10/group) or mock immunized with formulation buffer (N = 10). Mice were intravaginally challenged 6 weeks after the first immunization (T = 42d) with 200LD50 (3.2x10⁴ pfu/mouse) WT HSV-2 G strain and monitored daily for (a) survival and (b) clinical scores over 21 days after infection. (c) Viral shedding was assessed by plaque assay on vaginal swabs collected on days 1, 2 and 4 after challenge (viral titers expressed as log₁₀ plaque forming units (pfu)/mL). HSV-2 VNT (titer expressed as log₂ IC₅₀), gD2 ELISA and gB2 ELISA (titer expressed as log₁₀ EU/mL) were performed on sera collected 1 day before challenge and correlation analyses were performed between (d) HSV-2 VNT vs viral shedding (Area Under the Curve, AUC) per individual animal on Day 1, 2, 4 (Day 1–4), (e) HSV-2 VNT vs gD2 ELISA titers and (f) HSV-2 VNT vs the sum of gD2 and gB2 ELISA titers. Dotted lines indicate the limit of detection (LOD) and horizontal lines represent mean value per group. Statistical comparisons of viral shedding between immunized groups (c) were performed by a pairwise Wilcoxon test with a 6-fold Bonferroni adjustment. Results of statistical analysis are depicted by asterisks: **p<0.01, *** p<0.001, ns: not significant. Correlation analyses were performed using the Spearman rank correlation method with the correlation coefficient and two-tailed p value for each analysis depicted in the figure.