Therapeutic immunization with a mixture of herpes simplex virus 1 glycoprotein D-derived “asymptomatic” human CD8+ T-cell epitopes decreases spontaneous ocular shedding in latently infected HLA transgenic rabbits: association with low frequency of local PD-1+ TIM-3+ CD8+ exhausted T cells

Abstract

Most blinding ocular herpetic disease is due to reactivation of herpes simplex virus 1 (HSV-1) from latency rather than to primary acute infection. No herpes simplex vaccine is currently available for use in humans. In this study, we used the HLA-A*02:01 transgenic (HLA Tg) rabbit model of ocular herpes to assess the efficacy of a therapeutic vaccine based on HSV-1 gD epitopes that are recognized mainly by CD8(+) T cells from "naturally" protected HLA-A*02:01-positive, HSV-1-seropositive healthy asymptomatic (ASYMP) individuals (who have never had clinical herpes disease). Three ASYMP CD8(+) T-cell epitopes (gD(53-61), gD(70-78), and gD(278-286)) were linked with a promiscuous CD4(+) T-cell epitope (gD(287-317)) to create 3 separate pairs of CD4-CD8 peptides, which were then each covalently coupled to an Nε-palmitoyl-lysine moiety, a Toll-like receptor 2 (TLR-2) ligand. This resulted in the construction of 3 CD4-CD8 lipopeptide vaccines. Latently infected HLA Tg rabbits were immunized with a mixture of these 3 ASYMP lipopeptide vaccines, delivered as eye drops in sterile phosphate-buffered saline (PBS). The ASYMP therapeutic vaccination (i) induced HSV-specific CD8(+) T cells that prevent HSV-1 reactivation ex vivo from latently infected explanted trigeminal ganglia (TG), (ii) significantly reduced HSV-1 shedding detected in tears, (iii) boosted the number and function of HSV-1 gD epitope-specific CD8(+) T cells in draining lymph nodes (DLN), conjunctiva, and TG, and (iv) was associated with fewer exhausted HSV-1 gD-specific PD-1(+) TIM-3+ CD8(+) T cells. The results underscore the potential of an ASYMP CD8(+) T-cell epitope-based therapeutic vaccine strategy against recurrent ocular herpes.

Importance: Seventy percent to 90% of adults harbor herpes simplex virus 1 (HSV-1), which establishes lifelong latency in sensory neurons of the trigeminal ganglia. This latent state sporadically switches to spontaneous reactivation, resulting in viral shedding in tears. Most blinding herpetic disease in humans is due to reactivation of HSV-1 from latency rather than to primary acute infection. To date, there is no licensed therapeutic vaccine that can effectively stop or reduce HSV-1 reactivation from latently infected sensory ganglia and the subsequent shedding in tears. In the present study, we demonstrated that topical ocular therapeutic vaccination of latently infected HLA transgenic rabbits with a lipopeptide vaccine that contains exclusively human “asymptomatic” CD8(+) T-cell epitopes successfully decreased spontaneous HSV-1 reactivation, as judged by a significant reduction in spontaneous shedding in tears. The findings should guide the clinical development of a safe and effective T-cell-based therapeutic herpes vaccine.

FIG 1

Detection of HLA-A*02:01 molecules in PBMC, corneas, and trigeminal ganglia of HLA transgenic rabbits. (A and B) Peripheral blood mononuclear cells (PBMC) from either HLA transgenic (HLA Tg) rabbits or wild-type nontransgenic rabbits were stained with PE-conjugated anti-HLA-A2 MAb, (clone BB7.2) and analyzed by flow cytometry. (C) Coexpression of HLA-A*02:01 (red) and rabbit MHC class I (green) on the surface of a PBMC derived from an HLA Tg rabbit and detected by fluorescence microscopy (magnification, ×20). PBMC from the HLA Tg rabbit were double stained with two different primary antibodies, HLA-A2.1 MAb (BB7.2) and a rabbit MHCI, followed by staining with secondary antibodies conjugated with either Alexa Fluor 594 (red) or Alexa Fluor 488 (green), respectively. Merged yellow signals indicate the colocalization of these two molecules on the surface of PBMC. (D) Corneal and TG sections from either HLA transgenic rabbits or wild-type nontransgenic rabbits were immunostained with FITC (green)-conjugated BB7.2 MAb (anti-human HLA-A*02:01) and analyzed by fluorescence microscopy (see Materials and Methods). Cell nuclei are shown in red.

FIG 2

ASYMP HSV-1 gD lipopeptide vaccine targeting ocular surface-derived APCs highly expressing TLR-2. (A) Schematic representation of prototypes of human ASYMP CD4-CD8 lipopeptide vaccines. The C-terminal end of a promiscuous CD4⁺ T-cell peptide epitope (gD_287–317) was covalently joined with the N-terminal end of one of three different HSV-1 ASYMP gD CD8 T-cell epitopes: gD_53–61, gD_70–78, or gD_278–286. The N-terminal end of each resulting CD4-CD8 peptide was extended by a lysine covalently linked to a TLR-2 ligand (one molecule of palmitic acid [PAM]). This results in 3 separate pairs of ASYMP CD4-CD8 lipopeptides. (B) Amino acid sequence of each CD4⁺ and CD8⁺ T-cell epitope. (C) Representative FACS data showing that high levels of Toll-like receptor 2 (TLR-2) were expressed by rabbit cornea- and conjunctiva-derived CD11b/c⁺ antigen-presenting cells (APCs) isolated from naive conjunctiva (10⁶ per assay) first surface stained with 1 μg/ml of anti-rabbit CD11b/c-FITC MAb and then intracellularly stained with 3 μg/ml of PE-labeled MAbs specific to either TLR-2 or TLR-4 (thick black lines) or with a PE-labeled isotype control MAb (dotted black lines). The numbers above each histogram plot represent the intensity of expression of each TLR, represented by mean fluorescence intensity (MFI). (D) Average expression of TLR-2 (black circles) and TLR-4 (white circles) in spleens, corneas, and conjunctivas of five rabbits. *, P < 0.05 compared to the MFI of TLR-2 expression to the isotype control. The results are representative of two independent experiments.

FIG 3

Immunotherapeutic regimen of HLA transgenic rabbits latently infected with McKrae.

FIG 4

Therapeutic immunization of latently infected HLA Tg rabbits with HSV-1 ASYMP lipopeptide vaccine decreases virus spontaneous virus shedding in tears. Ocularly infected HLA Tg rabbits were immunized with a mixture of 3 SYMP lipopeptide vaccines or a mixture of 3 ASYMP lipopeptide vaccines delivered as eye drops in sterile PBS or with PBS alone (mock), as illustrated in Fig. 3. HSV-1 shedding detected in tears and reduced recurrent ocular herpetic disease were monitored for 100 days postimmunization, as described in Materials and Methods. (A) Maximal viral loads detected during the initial 30-day monitoring period in eye swabs. (B) Percentage of positive tears during the initial 30-day monitoring period. (C) Percentage of positive tears during the last 30-day monitoring period.

FIG 5

Local HSV-1 gD epitope-specific CD8⁺ T cells boosted by the ASYMP therapeutic lipopeptide vaccine in latently infected HLA Tg rabbits. Forty latently infected HLA Tg rabbits were immunized with the ASYMP vaccine on days 28, 49, and 70 postinfection, as illustrated in Fig. 3. Tears were collected daily for 30 days beginning 10 days after the final immunization, and the rabbits were divided into two groups. Protected rabbits (n = 10) had no detected virus shedding in either eye. Unprotected rabbits shed virus at least once in one or both eyes (n = 10). The frequency and absolute numbers of CD8⁺ T cells specific to each HSV-1 gD epitopes in protected versus unprotected HLA Tg rabbits were determined following therapeutic immunization with the ASYMP CD4-CD8 lipopeptide vaccine. CD8⁺ T cells were counted from DLN from the protected versus unprotected rabbits used for Fig. 4. (A) The cell suspensions were immunostained with an FITC-labeled MAb specific to rabbit CD8 and with a PE-labeled human HLA-A*0201/tetramer specific to each of the three human ASYMP CD8⁺ T-cell epitopes. The numbers show the percentage of tetramer-positive/CD8⁺ T cells specific to the gD_53–61, gD_70–78, or gD_278–286 epitope from each protected and unprotected rabbit. (B) Conjunctiva sections from either protected or unprotected HLA Tg rabbits were immunostained with MAb specific to rabbit CD8⁺ and analyzed by fluorescence microscopy (see Materials and Methods). Panel b shows representative data for CD8⁺ T cells aggregate in “protected” conjunctiva compared to dispersed CD8⁺ T cells in unprotected conjunctiva in panel a. Yellow represents stained CD8⁺ T cells. (C) Correlation of the average number of tetramer-positive/CD8⁺ T cells in TG specific to three human CD8⁺ T-cell gD epitopes (gD_53–61, gD_70–78, and gD_278–286) with virus titers detected in tears of latently infected and ASYMP lipopeptide-vaccinated HLA Tg rabbits. The results are representative of two independent experiments.

FIG 6

HSV-1 ASYMP CD4-CD8 lipopeptide vaccine-induced CD8⁺ T-cell responses in protected and unprotected HLA Tg rabbits. (A and D) Representative data for GzmB (A) and IFN-γ (D) and CD8⁺ T-cell responses in TG and DLN of ASYMP CD4-CD8 lipopeptide-immunized protected and unprotected HLA Tg rabbits. (B, C, E, and F) Individual responses for week 4 in TG and DLN of ASYMP CD4-CD8 lipopeptide-immunized protected and unprotected HLA Tg rabbits. A one-way ANOVA was used to calculate the P values for the comparison of the magnitude of the CD8⁺ T-cell responses. Horizontal lines indicate medians and vertical lines interquartile ranges. (G) Proportions, at week 4, of HSV-1 gD epitope-specific CD8⁺ T cells (i.e., cells having at least one function) that make any given combination of the two functions (GzmB cytotoxicity and IFN-γ). The proportions of HSV-1 gD epitope-specific CD8⁺ T cells capable of simultaneously secreting IFN-γ (IFN⁺) and expressing cytotoxic activity (GzmB⁺) in protected versus unprotected HLA Tg rabbits are shown.

FIG 7

HSV-specific CD8⁺ T cells from unprotected symptomatic HLA Tg rabbits are phenotypically and functionally exhausted. (A and B) Representative dot plot (A) and mean (B) of percentages of HSV-1 gD_53–61-specific PD-1⁺ CD8⁺ T cells in protected (ASYMP) and unprotected (SYMP) HLA Tg rabbits. (C and D) representative dot plot (C) and mean (D) of percentages of HSV-1 gD_53–61-specific TIM3⁺ CD8⁺ T cells in protected (ASYMP) and unprotected (SYMP) HLA Tg rabbits. (E) Proportions of exhausted HSV-1 gD epitope-specific CD8⁺ T cells expressing PD-1, TIM3, and both PD-1 and TIM3 in unprotected SYMP HLA Tg rabbits. (F) Percentages of reactivated TG in protected versus unprotected HLA Tg rabbits. (G) Correlation between the percentages of exhausted PD-1⁺ TIM3⁺ CD8⁺ T cells and percentages of reactivated TG. (H) Percentages of reactivated TG in protected HLA Tg rabbits in the presence and absence of anti-CD8 and anti-CD4 MAbs and IgG controls. The results are representative of 2 independent experiments. The indicated P values, calculated using one-way ANOVA, show statistical significance of differences between protected and nonprotected HLA Tg rabbits.