Distinguishing Features of High and Low Dose Vaccine Against Ocular HSV-1 Infection Correlates with Recognition of Specific HSV-1-Encoded Proteins

Abstract

The protective efficacy of a live attenuated herpes simplex virus type 1 (HSV-1) vaccine, HSV-1 0ΔNLS, was evaluated in mice prophylactically in response to ocular HSV-1 challenge. Mice vaccinated with the HSV-1 0ΔNLS were found to be more resistant to subsequent ocular virus challenge in terms of viral shedding, spread, the inflammatory response, and ocular pathology in a dose-dependent fashion. Specifically, a strong neutralizing antibody profile associated with low virus titers recovered from the cornea and trigeminal ganglia (TG) was observed in vaccinated mice in a dose-dependent fashion with doses ranging from 10³-10⁵ PFU HSV-1 0ΔNLS. This correlation also existed in terms of viral latency in the TG, corneal neovascularization, and leukocyte infiltration and expression of inflammatory cytokines and chemokines in infected tissue with the higher doses (10⁴-10⁵ PFU) of the HSV-1 0ΔNLS vaccinated mice displaying reduced viral latency, ocular pathology, or inflammation in comparison to the lowest dose (10³ PFU) or vehicle vaccine employed. Fifteen HSV-1-encoded proteins were uniquely recognized by antisera from high dose (10⁵ PFU) vaccinated mice in comparison to low dose (10³ PFU) or vehicle vaccinated animals. Passive immunization using high- but not low-dose vaccinated mouse sera showed significant efficacy against ocular pathology in HSV-1-challenged animals. In summary, we have identified the minimal protective dose of HSV-1 0ΔNLS vaccine in mice to prevent HSV-mediated disease and identified candidate proteins that may be useful in the development of a non-infectious prophylactic vaccine against the insidious HSV-1 pathogen.

INTRODUCTION

The normal corneal stroma of the eye is composed predominantly of collagen lamellae highly organized into interwoven fibrils in the anterior stroma that run parallel to the cornea surface in the posterior stroma (1). This architectural arrangement along with the organization of the epithelial and endothelial layers provides a durable cover to protect the remainder of the eye from environmental insult and allow passage of light to the lens and retina. Although the eye is considered an immunologically privileged organ (2), resident leukocytes including macrophages, dendritic cells and mast cells populate primarily the peripheral cornea or limbal arcade proximal to the vasculature (3–5). In response to environmental stimuli including trauma or infection, the immune privilege dynamics of the cornea dramatically change.

Herpes simplex virus (HSV)-1 is a highly successful human pathogen that has a seroprevalence rate above 50% worldwide but is declining in prevalence in the United States and elsewhere (6). It continues to be a significant ocular pathogen that can elicit immune-driven, irreversible damage to the cornea in patients that experience episodic reactivation of latent virus. Experimentally, in response to ocular infection resident and infiltrating myeloid-derived cells are activated and/or initially recruited to the cornea followed by NK cells and T lymphocytes that collectively facilitate clearance of the pathogen insult but also lead to severe inflammation and irreversible tissue pathology including vascularization (blood and lymphatic vessel genesis) of the normally avascular central cornea (7–9). In addition, the innervation of the cornea which normally maintains the homeostasis of the ocular surface (10), is dramatically altered and can lead to dry eye disease (11–14). A compromised visual axis as a result of HSV-1-mediated corneal pathology that cannot be managed successfully often leads to corneal transplant, a high risk surgical procedure with frequent graft failures (15). As there is currently no intervention to permanently alleviate patient suffering or prevent the acquisition of HSV-1 or HSV-2, strategies are sought to enhance patient resistance to this infection that in 2013 had an estimated economic burden of over $90 million dollars in the United States alone (16).

The potential to protect individuals from pathogens or products encoded by pathogens has been realized and demonstrated experimentally for well over 100 years (17). Early vaccine work against HSV-1 suggested antigens associated with the envelope or the specific subunit glycoprotein (g)D were protective in preventing mortality or the establishment of latency following acute infection in mice (18, 19). Follow-up studies targeting gD or other HSV-1 subunits as prophylactic or therapeutic vaccines have demonstrated various degrees of efficacy in the generation of sterile immunity, reducing the establishment of latency, or preventing reactivation of latent virus (20–25). Most subunit vaccine approaches likely generate an antibody response with modest T cell input. As T cells and specifically CD8⁺ T cells have been shown to control HSV-1 reactivation in mice (26–28), recent studies by one group have focused on prophylactic vaccines that elicit a protective CD8⁺ T cell response using HLA-restricted transgenic mice and rabbits (29–31). Specifically, peptide epitopes of gB and the tegument proteins VP11/12 and VP13/14 identified for polyfunctional CD8⁺ T cell responses from seropositive, asymptomatic HLA-A*201–01 individuals used in CpG-adjuvant vaccines prevented HLA-A*2:01 transgenic mouse and rabbit mortality associated with a significant drop in ocular viral replication following acute HSV-1 challenge. A follow-up study using a different set of HLA-A*02:−01-restricted epitopes from UL9, UL25, and UL44 gene products in a prime/pull therapeutic strategy found this approach significantly increased the tissue resident effector memory CD8⁺ T cell population in the ganglion of HSV-1 latently infected mice and prevented virus reactivation and reduced ocular disease scores (32). While these results hold promise in the development of candidate prophylactic or therapeutic vaccines against ocular HSV-1 infection, with few exceptions, none of the studies referenced above included an evaluation of the visual axis in terms of quantifiable pathological changes of the cornea including function as well as analysis of visual performance.

Previous studies by our group have identified the application of a live, attenuated HSV-1 mutant (HSV-1 0ΔNLS) as a highly efficacious prophylactic vaccine against HSV-1 (33). HSV-1 0ΔNLS was demonstrated to be a safe vaccine using IFNAR1 deficient mice and provided superior protection compared to a subunit vaccine used in clinical trials (33, 34). Notably, the efficacy of the vaccine in the control of virus replication in the cornea was linked to early expression of complement and the neonatal Fc receptor which supported the correlate of protection to be antibody (35). Finally, HSV-1 0ΔNLS vaccinated mice retained corneal function with preservation of the visual axis, the first study to report such findings (36). In the current investigation, a dose-response study was conducted using the HSV-1 0ΔNLS vaccine to determine the lowest dose required to maximize the protective efficacy. While all doses of vaccine protected mice from HSV-1-mediated mortality, there were significant differences in terms of level of inflammation and corneal pathology with mice vaccinated with the high dose inoculum showing the least pathology in comparison to the low dose vaccinated mice. We identified 15 viral-encoded proteins uniquely recognized by antiserum from high dose vaccinated mice that may serve as candidates for further testing as surrogate vaccines for the live attenuated HSV-1 0ΔNLS used as a vaccine in the current study.

MATERIALS AND METHODS

Mice, vaccination procedure, and ocular infection

Female and male outbred CD1 mice were obtained from Charles River Laboratories (Wilmington, MA). Ai14/Rosa26-tdTomato-Cre-reporter (Rosa) male and female mice on a C57BL/6 background were originally purchased from the Jackson Laboratory (Bar Harbor, ME) and bred in house. All animals were housed in a specific pathogen free facility at the Dean McGee Eye Institute on the University of Oklahoma Health Sciences Center (OUHSC) campus. Investigators adhered to procedures approved by the OUHSC Institutional Animal Care and Use Committee (Protocol # 16–087-SSIC and #19–060-ACHIX), and animals were handled in accordance with the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals. Mice were anesthetized for all procedures by i.p. injection of ketamine (100 mg/kg) plus xylazine (5.0 mg/kg) and were euthanized by cardiac perfusion of 10 ml PBS for tissue collection. Animals (6–10 weeks old) were vaccinated using a prime/boost approach via ipsilateral footpad (subcutaneous) and quadriceps (intramuscular) injection 3 weeks later as described (33). The immunization dosage ranged from 1 × 10³ – 1 × 10⁵ plaque forming units (PFU) HSV-1 0ΔNLS (KOS strain) in 10 μl PBS (primer and boost). PBS alone served as the vehicle control.

CD-1 mice were infected 30 days following the secondary boost by applying 1 × 10³ PFU HSV-1 McKrae to each cornea following partial epithelial debridement with a 25-gauge needle. Rosa26 mice were infected 30 days post boost by applying 1 × 10⁴ PFU HSV-1 SC16 expressing Cre recombinase (37) to each cornea following debridement as described above.

Serological and virological assays

Peripheral blood was obtained from the facial vein of anesthetized mice at 30 days post boost and fractionated using Microtainer serum separation tubes (Becton Dickinson, Franklin Lakes, NJ). Ab-containing serum was evaluated for virus-neutralizing titers in the presence of guinea pig complement (Rockland, Limerick, PA) on Vero cell monolayers as described (33). To quantify virus found in the tear film or tissue during acute infection of mice, corneas were swabbed with cotton-tipped applicators and tissue were excised, homogenized, and the clarified supernatant was assayed for infectious virus by standard plaque assay (33). HSV-1 genome copy number was conducted by PCR on total DNA isolated from the trigeminal ganglia (TG) of surviving mice 30 days post infection (DPI) using a proprietary primer-probe mix, according to the manufacturer’s instructions (Virusys Corp., Taneytown, MD).

Analysis of corneal pathology

Gross corneal pathology was conducted at 7 DPI by a masked observer examining eyes through a Kowa SL14 portable slit lamp biomicroscope (Kowa Optimed Inc., Torrance, CA.) using the following scoring scheme: 0, no pathology; 1, injected eye, no opacity; 2, focal opacity; 3, hazy opacity over entire cornea; 4, dense opacity in central cornea with remainder haze; 5, same as 4 but with ulcer; 6, corneal perforation as previously described (38).

Visualization of blood and lymphatic vessel genesis was performed in which corneas from enucleated eyes of euthanized mice were fixed in a 4% solution of paraformaldehyde (Sigma-Aldrich, St. Louis, MO) for 30 min, followed by two 5 min washes in PBS. The tissue was then incubated in PBS containing 1% Triton X-100 overnight. Labeling, imaging, and analysis of corneal vessels were performed using an Olympus FV1200 confocal microscope and MetaMorph Imaging Suite software (Sunnyvale, CA) as previously described (39).

Visualization and analysis of Cre-inducible tomato red-staining cells was conducted using an Olympus FV1200 confocal microscope and MetaMorph Imaging suite software. Specifically, the trigeminal ganglia (TG) from vaccinated mice were harvested 30 DPI and placed in PBS for 5 min. PBS was removed and 4% paraformaldehyde was added to each TG. The samples were processed in 5 μm sections, and placed onto slides by Excalibur Pathology Inc. (Norman, OK). Slides were then imaged and the threshold area calculated for each section visualized using MetaMorph Imaging suite software.

Flow cytometry

Corneas, TG, and submandibular lymph nodes (MLN) were harvested from vaccinated mice at 3 and 7 DPI following exsanguination. Briefly, TG or MLN pairs were macerated into single-cell suspensions in RPMI 1640 medium containing 10% heat-inactivated FBS, 1x antibiotic/antimycotic solution, and 10 μg/ml gentamicin (Invitrogen, Carlsbad, CA) (complete media). Corneas were digested in 0.25 Wümsch units of Liberase TL enzyme (Roche Diagnostics, Indianapolis, IN) suspended in 500 μl of complete media at 37° C for 1 hr, and exposed to trituration every 15–20 min. Corneas, TG, and MLN were then filtered through a 40-μm nylon mesh filter (Thermo Fisher Scientific, Waltham, MA) prior to labeling. Cell suspensions were blocked with anti-CD16/32 (eBioscience, San Diego, CA), labeled with the indicated combination of Abs for 20–30 min, and washed in 1X PBS containing 1% BSA. All samples were analyzed on a MACSQuant 10 flow cytometer with MACSQuantify software (Miltenyi Biotec, Bergisch Gladbach, Germany). Isotype labeling and fluorescence minus one (FMO) controls were conducted to validate specific labeling and negate spectral overlap respectively (Supplemental Figure 1).

Cytokine/Chemokine Quantification

Corneas and TG from vaccinated mice were collected 7DPI following exsanguination. Uninfected mouse tissue served as baseline controls. Tissue was weighed upon extraction and placed in Next Advance GREEN bead lysis tubes (Averill Park, NY) containing PBS and 1X protease inhibitor cocktail (Santa Cruz Biotechnology, Dallas, TX). The samples were then homogenized in a Next Advance Bullet Blender Storm 24 homogenizer for 10 min, sonicated in a water bath for 10 min and subsequently analyzed for cytokine/chemokine content using customized kits for select analytes (Millipore, Billerica, MA) and a Bio-Plex suspension array system to detect and quantify analytes (Bio-Rad, Hercules, CA). The sample contents were normalized based on the wet weight of each cornea and reported in pg analyte/mg cornea. Samples were diluted 1:5 prior to analysis. The limit of detection of each analyte evaluated was as follows: eotaxin: 3.17 pg, G-CSF: 3.26 pg, GM-CSF: 20.1 pg, IFN-γ: 3.22 pg, IL-1α: 3.78 pg, IL-1β: 3.30 pg, IL-2: 3.27 pg, IL-3: 3.19 pg, IL-4: 3.23 pg, IL-5: 3.04 pg, IL-6: 3.25 pg, IL-7: 3.07 pg, IL-10: 3.26 pg, IL-12p40: 2.93 pg, IL-12p70: 2.77 pg, IL-13: 8.53 pg, IL-15: 2.63 pg, IL-17: 3.28 pg, CXCL10: 3.23 pg, CXCL1: 3.09 pg. LIF: 3.24 pg, CCL2: 2.99 pg, M-CSF: 3.58 pg, CXCL9: 3.05 pg, CCL3: 17.25 pg, CCL4: 19.11 pg, CXCL2: 20.29 pg, CCL5: 3.2 pg, TNF-α: 3.22 pg, and VEGFA: 3.24 pg.

Immunoprecipitation and HSV-1 protein identification using Ab from vaccinated mice

To determine the repertoire of HSV-1 proteins that are recognized by HSV-1 0ΔNLS vaccinated mouse serum, 1.2 × 10⁶ Vero cells were plated into each well of a 6-well plate and infected at a multiplicity of infection of 1.0 with HSV-1 McKrae. Twenty four hr post-infection the cells were collected, washed in 1.0 ml of PBS twice, and centrifuged at 300 x g, 5 min for each wash. Following the second wash, the supernatant was removed, and the cells were resuspended in 500 μl of 1% Triton X-100 detergent (lysis buffer) and placed on ice for 15–20 min. Following the incubation, cell lysates were clarified from cell debris by centrifugation (10,000xg, 10 min at 4° C). The supernatants from infected and uninfected Vero cells were incubated with 4 μl of serum from vaccinated mice and 100 μl of immunomagnetic protein G microbeads (Miltenyi Biotec) at 4° C for 30 min with gentle agitation every 5 min. The protein/Ab/microbead complex was then added onto μMACS magnetic bead columns (Miltenyi Biotec). The columns were washed 4X with 200 μl lysis buffer, and retained proteins were eluted with 50 μl 100 mM glycine, pH 2.5.

Trypsin digest of immunoprecipitated proteins was performed according to the filter-aided sample preparation (FASP) protocol (40). Briefly, the eluate was buffer exchanged to 8M urea, the proteins were reduced with 10mM DTT (dithiothreitol) and then alkylated with 10mM iodoacetamide. The peptides were eluted in 10mM ammonium acetate pH 8.0, dried and resuspended in 10mM ammonium formate pH 10.0. Liquid chromatography tandem mass spectrometry was performed by coupling a nanoAcquity UPLC (Waters Corp., Manchester, UK) to a Q-TOF SYNAPT G2S instrument (Waters Corp., Manchester, UK). Each protein digest (about 100ng of peptide) was delivered to a trap column (300 μm × 50 mm nanoAcquity UPLC NanoEase Column 5 μm BEH C18, Waters Corp, Manchester, UK) at a flow rate of 2 μl/min in 99.9% solvent A (10mM ammonium formate pH 10, in HPLC grade water). After 3 min of loading and washing, peptides were transferred to another trap column (180 μm × 20 nanoAcquity UPLC 2G-V/MTrap 5 μm Symmetry C18, Waters Corp, Manchester, UK) using a gradient from 1% to 60% solvent B (100% acetonitrile). The peptides were then eluted and separated at a flow rate of 200 nL/min using a gradient from 1% to 40% solvent B (0.1% FA in acetonitrile) for 60min on an analytical column (7.5 μm × 150 mm nanoAcquity UPLC 1.8 μm HSST3, Waters Corp, Manchester, UK). The eluent was sprayed via PicoTip Emitters (Waters Corp, Manchester, UK) at a spray voltage of 3.0 kV and a sampling cone voltage of 30 V and a source offset of 60 V. The source temperature was set to 70 °C. The cone gas flow was turned off, the nano flow gas pressure was set at 0.3 bar and the purge gas flow was set at 750 ml/h. The SYNAPT G2S instrument was operated in data-independent mode with ion mobility (HDMSe). Full scan MS and MS2 spectra (m/z 50 – 2000) were acquired in resolution mode (20,000 resolution FWHM at m/z 400). Tandem mass spectra were generated in the trapping region of the ion mobility cell by using a collisional energy ramp from 20 V (low mass, start/end) to 35 V (high mass, start/end). A variable IMS wave velocity was used. Wave velocity was ramped from 300 m/s to 600 m/s (start to end) and the ramp was applied over the full IMS cycle. A manual release time of 500 μs was set for the mobility trapping and a trap height of 15 V with an extract height of 0 V. The pusher/ion mobility synchronization for the HDMSe method was performed using MassLynx V4.1 and DriftScope v2.4. LockSpray of Glufibrinopeptide-B (m/z 785.8427) was acquired every 60 s and lock mass correction was applied post acquisition.

Raw MS data were processed by PLGS (ProteinLynx Global Server, Waters Corp., Manchester, UK) for peptide and protein identification. MS/MS spectra were searched against the Uniprot HSV-1 proteome database (release date November 2, 2017 containing 1,776 un-reviewed sequences) with the following search parameters: full tryptic specificity, up to two missed cleavage sites; carbamidomethylation of cysteine residues was set as a fixed modification and N-terminal protein acetylation and methionine oxidation were set as variable modifications. Proteins reported were identified in ≥2 out of 7 samples per group.

Passive Immunization

Serum obtained from terminal cardiac punctured PBS (vehicle)- or 0ΔNLS (10³ or 10⁵ PFU)-vaccinated mice was pooled and administered i.p. (250 μl) to naive CD1 male or female mice 24 hr prior to HSV-1 (1,000 PFU/eye) challenge. Mechanosensory function of the cornea was assessed 1–7 DPI using a Cochet-Bonnet esthesiometer as previously described (12). Mice were monitored for cumulative survival and deaths recorded to 21 DPI. Mice were subsequently exsanguinated, and the corneas were assessed for opacity as previously described (41) and subsequently processed for neovascularization (39).

Statistics

Graphpad Prism 8 was used to analyze data for statistical significance (p<.05) as determined using statistical tests described in each figure. Data are presented as mean ± SEM.

RESULTS

HSV-1 0ΔNLS vaccine suppresses virus replication and prevents HSV-1-mediated mortality in a dose-dependent fashion

Previous studies investigating the efficacy of the HSV-1 0ΔNLS vaccine against ocular HSV-1 challenge were conducted using a dose of 10⁵ PFU for the primary immunization and boost (33–36). To formalize the minimal efficacious dose of vaccine to maximize the protective effect against subsequent challenge, mice were immunized with doses ranging from 10³-10⁵ PFU prior to challenge. Mice immunized with 10⁴-10⁵ PFU HSV-1 0ΔNLS showed significantly higher neutralizing antibody titers to HSV-1 compared to mice vaccinated with 10³ PFU HSV-1 0ΔNLS (Fig. 1A). However, all doses of HSV-1 0ΔNLS used to immunize mice were found to protect animals subsequently challenged with HSV-1 as measured by cumulative survival (Fig. 1B).

Figure 1: Mice vaccinated with 10³-10⁵ PFU HSV-1 0ΔNLS are protected against HSV-1-mediated mortality but show differences in antibody neutralization titers.

(A) Male and female mice (n=11–17/group) were subcutaneously immunized with 10³-10⁵ PFU HSV-1 0ΔNLS vaccine followed by an intramuscular boost three weeks later. Blood was collected thirty days post boost and assessed for neutralization titers to HSV-1. **p<0.01 compared to the vehicle (PBS) vaccinated group. ^Δp<.05 comparing 10⁴-10⁵ PFU HSV-1 0ΔNLS to the 10³ PFU HSV-1 0ΔNLS vaccinated group as determined by ANOVA and Scheffé multiple comparison test. (B) Mice vaccinated with 10³-10⁵ PFU HSV-1 0ΔNLS (n=17–19/group) were challenged with 10,000 PFU HSV-1 McKrae per eye 30 days following the final immunization. Cumulative survival was recorded over 30 days post infection. **p<.001 comparing the HSV-1 0ΔNLS vaccinated mice to vehicle (PBS) control vaccinated animals as determined by ANOVA and Wilcoxon test.

To further compare the dose response of the HSV-1 0ΔNLS vaccine in resistance against ocular HSV-1 challenge, the tear film, cornea and TG from mice immunized with 10³-10⁵ PFU HSV-1 0ΔNLS or vehicle (PBS) were evaluated for viral content. Mice vaccinated with any dose of HSV-1 0ΔNLS were found to shed less virus during acute infection (i.e., 3–7 DPI) compared to PBS-vaccinated mice (Fig. 2A). By comparison, only the 10⁴-10⁵ PFU dose of HSV-1 0ΔNLS reduced corneal virus titer whereas the 10³ PFU dose of HSV-1 0ΔNLS did not have a significant effect by 7 DPI (Fig. 2B). Similar to what was observed in viral shedding, mice immunized with any dose of the HSV-1 0ΔNLS vaccine showed reduction in virus replication in the TG compared to the PBS vaccinated control group (Fig. 2C). However, little to no infectious virus was recovered from the TG of mice vaccinated with 10⁴-10⁵ PFU HSV-1 0ΔNLS compared to the low (10³ PFU) dose of HSV-1 0ΔNLS in which over 50% of mice TG had detectable levels of infectious virus (Fig. 2C).

Figure 2: HSV-1 0ΔNLS vaccine suppresses viral replication and spread in the cornea and TG in a dose-dependent manner.

Male and female mice were subcutaneously immunized with 10³-10⁵ PFU HSV-1 0ΔNLS or vehicle (PBS) vaccine followed by an intramuscular boost three weeks later. Vaccinated mice (n=17–19/group) were challenged with 10,000 PFU HSV-1 McKrae per eye 30 days following the final immunization. (A) Mouse corneas (5–7 mice/group/time point) were swabbed from vaccinated mice at the indicated day (1–7) post infection and assayed for viral content by plaque assay. Data is presented as mean ± SEM; *p<.05, **p<.01 comparing the indicated group to PBS-vaccinated control. (B) Mouse corneas were harvested at day 7 PI, processed, and assayed for viral content by plaque assay. The experiment was repeated twice with 5–6 mice/group. *p<.05, **p<.01 comparing the indicated group to the PBS-vaccinated control group. (C) Mouse TG were harvested at day 7 PI, processed, and assayed for viral content by plaque assay. The experiment was repeated twice with 9–11 mice/group. **p<.01 comparing the indicated group to the PBS-vaccinated control group. ^Δp<.05 comparing the indicated groups to the low (10³ PFU 0ΔNLS) dose vaccinated group. Data were analyzed by ANOVA and Tukey’s post-hoc t-test.

A critical aspect of HSV-1 infection is the establishment of latency as a result of invasion of the sensory neurons that reside in the TG during acute infection. To determine the success of permanently colonizing neurons in the TG, Cre-inducible tdTomato fluorescent reporter mice vaccinated with HSV-1 0ΔNLS (10³-10⁵ PFU) were challenged with HSV-1 encoding Cre recombinase under the infected cell protein (ICP)0 lytic gene promoter (37). Cells that survive the acute infection are permanently “tagged” and will express tdTomato. TG from mice immunized with 10⁴-10⁵ PFU HSV-1 0ΔNLS displayed significantly fewer labeled cells as measured by threshold area compared to mice vaccinated with 10³ PFU HSV-1 0ΔNLS at 30 days PI (Fig. 3A, 3B). These results are consistent with the genome copy number of HSV-1 recovered in the TG of latent-infected, vaccinated mice with a significant reduction recovered from mice immunized with 10⁴-10⁵ PFU HSV-1 0ΔNLS compared to the 10³ PFU dose 30 DPI (Fig. 3C). In this experiment, PBS-vaccinated mice did not survive and therefore, could not be assessed as a baseline positive control. However, a previous study found PBS-vaccinated tdTomato fluorescent reporter mice that did survive the acute infection to 30 DPI displayed a similar phenotype as that shown with 10³ PFU 0ΔNLS vaccinated animals (36). In summary, all doses of the HSV-1 0ΔNLS vaccine tested showed protection against ocular HSV-1 challenge in terms of cumulative survival and viral spread and replication in the TG although the lowest dose evaluated displayed no efficacy against HSV-1 replication in the cornea which correlated with a low neutralizing antibody titer and higher virus copy number found in the TG of latent infected mice.

Figure 3: HSV-1 0ΔNLS vaccine reduces viral load in the TG of immunized mice during latency.

Latent virus was analyzed in trigeminal ganglia from vaccinated mice following ocular infection. (A) Representative sections of confocal images from vaccinated mice (10³-10⁵ PFU HSV-1 0ΔNLS) expressing the Cre-inducible tdTomato reporter construct on the Rosa 26 locus on a C57BL/6 background 30 days post ocular challenge with 1 × 10⁴ PFU/eye Cre-expressing HSV1 SC16. Neurons successfully infected with the virus are permanently labeled and express the tdTomato reporter. Bar = 100 μm. (B) Summary threshold area of tdTomato expression by cells in TG sections (n=7/group) from immunized mice (n=3/group). Data is displayed as threshold area, ***p<.001 comparing 10³ PFU HSV-1 0ΔNLS vaccine to higher dose vaccines. (C) HSV-1 copy number in the TG of vaccinated mice (10³-10⁵ PFU 0ΔNLS) at day 30 PI (n=21–22/group). *p<.05, **p<.01 comparing the 10³ PFU dose to the higher doses of HSV-1 0ΔNLS. Data was analyzed by ANOVA and Tukey’s post-hoc t-test.

Leukocyte infiltration into the cornea and TG is stymied in HSV-1 0ΔNLS vaccinated mice in a dose-dependent manner

HSV-1 infection of the cornea elicits a robust cellular immune response initially with a massive onslaught of neutrophils and activation of mast cells followed by the infiltration of activated monocytes/macrophages and NK cells, and eventually CD4⁺ and CD8⁺ T cells (5, 42–44). We evaluated T and myeloid cell infiltration comparing vaccinated to non-vaccinated animals at 3 and 7 DPI in the cornea, the latter time point when non-vaccinated mice begin to succumb to infection (Fig. 1B). A representative flow plot comparing PBS-, 10³ PFU 0ΔNLS (low)-, and 10⁵ PFU 0ΔNLS (high)-vaccinated mice for T and B cells (Fig. 4A) and myeloid cells (Fig. 4D) is provided. At 3 DPI, the high dose vaccinated mice showed significantly fewer CD4⁺ and CD8⁺ T cells residing in the cornea compared to the PBS- or low dose-vaccinated mice (Fig. 4B). By 7 DPI, the high and low 0ΔNLS dose vaccinated mouse corneas retained fewer T cells than the PBS-vaccinated counterparts (Fig. 4C). A similar finding was observed in analysis of granulocytic populations at 3 (Fig. 4E) and 7 (Fig. 4F) DPI with either the high dose or high and low dose 0ΔNLS vaccinated mice possessing significantly fewer cells compared to the PBS-vaccinated control animals. Significantly fewer monocyte/macrophage populations were found in the cornea of the high and low dose 0ΔNLS vaccinated mice compared to the PBS-vaccinated controls at 3 (Fig. 4G) and 7 (Fig. 4H) DPI as well. Although modest in number, the corneas of infected mice contained significantly fewer CD19⁺ B lymphocytes in the high dose immunized mice compared to the vehicle-control vaccinated animals at 7 DPI (Fig. 4C) but not 3 DPI (Fig. 4B). As we are unable to detect B lymphocytes in the uninfected CD1 mouse cornea, the infiltration of these cells following infection is of interest relative to local antibody production. Currently, we have not been able to detect anti-HSV-1 antibody in the tear film of vaccinated mice prior to or following infection.

Figure 4: Diminished leukocyte infiltrate in the cornea of vaccinated mice in response to HSV-1 infection.

Male and female mice were subcutaneously immunized with 10³ PFU HSV-1 0ΔNLS, 10⁵ PFU HSV-1 0ΔNLS, or vehicle (PBS) vaccine followed by an intramuscular boost three weeks later. Mice (n=5/group) were challenged with 1,000 PFU HSV-1 McKrae per eye 30 days following the final immunization. Mice were euthanized at 3 or 7 DPI, and the corneas were removed and enzymatically processed to single cell suspensions. (A) Representative flow plot corneal digest for distribution of CD4⁺ and CD8⁺ T cells in vaccinated mice. (B) Absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes residing in the cornea of vaccinated mice 3 DPI. (C) Absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes residing in the cornea of vaccinated mice 7 DPI. Uninfected absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes were 9±3, 4±2, and 4±3 respectively. *p<.05, **p<.01, ***p<.001 comparing the indicated group to PBS-vaccinated group as determined by ANOVA and Tukey’s post hoc t-test. (D) Representative flow plot corneal digest for distribution of myeloid cells in vaccinated mice designated as 1: (CD45⁺CD11b⁺Ly6G^midLy6C⁺), 2: (CD45⁺CD11b⁺Ly6G^highLy6C^int) and 3: (CD45⁺CD11b⁺Ly6G⁻Ly6C^high). Absolute number of granulocytes (CD45⁺CD11b⁺Ly6G^midLy6C⁺) and (CD45⁺CD11b⁺Ly6G^highLy6C^int) at 3 (E) and 7 (F) DPI. Absolute number of monocytes (CD45⁺CD11b⁺Ly6G⁻Ly6C^high) at 3 (G) and 7 (H) DPI. *p<.05, **p<.01, ***p<.001 comparing the indicated group to PBS-vaccinated group or 10³ 0ΔNLS dose-vaccinated mice as determined by ANOVA and Tukey’s post hoc t-test.

Figure 7: Corneal neovascularization and opacity is significantly reduced in vaccinated mice in a dose-dependent fashion post HSV-1 challenge.

Male and female mice were subcutaneously immunized with 10³-10⁵ PFU HSV-1 0ΔNLS or vehicle (PBS) vaccine followed by an intramuscular boost three weeks later. Mice (n=6/group) were challenged with 10,000 PFU HSV-1 McKrae per eye 30 days following the final immunization. Thirty days post infection, the mice were initially evaluated for opacity using a masked observer and then subsequently euthanized. The corneas were removed and processed for whole mount staining for blood and lymphatic vessels. (A-C) Representative z-stacked corneal images depicting blood (red) and lymphatic (green) vessels comparing low (A, 10³ PFU), medium (B, 10⁴ PFU), and high (C, 10⁵ PFU) HSV-1 0ΔNLS vaccinated mouse corneas. (D) Metamorph quantification of corneal area covered by lymphatic or blood vessels comparing each group of vaccinated animals. *p<.05 comparing the high vaccine dose to the low vaccine dose for lymphatic vessels. **p<.01 comparing the high and medium vaccine dose to the low vaccine dose for blood vessels. (E) Opacity score of corneas from each groups of vaccinated mice. ^ΔΔp<.01 comparing the indicated groups to the low vaccine dose group. **p<.01 comparing the indicated groups to the vehicle (PBS)-treated group as determined by ANOVA and Tukey’s post hoc t-test, n=6 mice/group.

We also evaluated T and myeloid cell infiltration comparing vaccinated to non-vaccinated animals at 3 and 7 DPI in the TG. A representative flow plot comparing PBS-, 10³ PFU 0ΔNLS (low)-, and 10⁵ PFU 0ΔNLS (high)-vaccinated mice for T and B cells (Fig. 5A) and myeloid cells (Fig. 5D) is provided. Unlike the cornea, only the high dose-vaccinated mice possessed significantly fewer CD4⁺ T cells compared to the PBS control-vaccinated group at 3 DPI (Fig. 5B) whereas both CD4⁺ and CD8⁺ T cell numbers were significantly reduced in the TG of high 0ΔNLS dose-vaccinated mice at 7 DPI (Fig. 5C). The granulocyte infiltration was not impacted by the 0ΔNLS vaccination at 3 DPI with similar numbers of CD11b⁺Ly6G⁺Ly6C^+/int cells (Fig. 5E). However, at 7 DPI the granulocyte numbers were significantly reduced in the TG of the high 0ΔNLS dose-immunized mice compared to the PBS control-vaccinated animals (Fig. 5F). By comparison, monocyte/macrophage (CD45⁺CD11b⁺Ly6G⁻Ly6C^high) numbers were significantly lower in the TG of the high 0ΔNLS dose-immunized mice compared to the PBS control-vaccinated animals at 3 (Fig. 5G) and 7 (Fig. 5H) DPI. No differences were found comparing the low dose-vaccinated mice to the PBS-vaccinated control group. It should also be noted there was no change in the B lymphocyte number that resides in the TG comparing uninfected levels to that following infection suggesting that unlike the cornea, B lymphocytes do not traffic to the TG of infected mice (Fig. 5B & 5C).

Figure 5: Diminished leukocyte infiltrate in the trigeminal ganglia of vaccinated mice in response to HSV-1 infection.

Male and female mice were subcutaneously immunized with 10³ PFU HSV-1 0ΔNLS, 10⁵ PFU HSV-1 0ΔNLS, or vehicle (PBS) vaccine followed by an intramuscular boost three weeks later. Mice (n=5/group) were challenged with 1,000 PFU HSV-1 McKrae per eye 30 days following the final immunization. Mice were euthanized at 3 or 7 DPI, and the TG were removed and enzymatically processed to single cell suspensions. (A) Representative flow plot TG digest for distribution of CD4⁺ and CD8⁺ T cells in vaccinated mice. (B) Absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes residing in the TG of vaccinated mice 3 DPI. (C) Absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes residing in the cornea of vaccinated mice 7 DPI. Uninfected absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes were 51±11, 42±5, and 102±20 respectively. *p<.05, comparing the indicated group to PBS-vaccinated group as determined by ANOVA and Tukey’s post hoc t-test, n=5 mice/group.. (D) Representative flow plot corneal digest for distribution of myeloid cells in vaccinated mice designated as 1: (CD45⁺CD11b⁺Ly6G^midLy6C⁺), 2: (CD45⁺CD11b⁺Ly6G^highLy6C^int) and 3: (CD45⁺CD11b⁺Ly6G⁻Ly6C^high). Absolute number of granulocytes (CD45⁺CD11b⁺Ly6G^midLy6C⁺) and (CD45⁺CD11b⁺Ly6G^highLy6C^int) at 3 (E) and 7 (F) DPI. Absolute number of monocytes (CD45⁺CD11b⁺Ly6G⁻Ly6C^high) at 3 (G) and 7 (H) DPI. *p<.05 and **p<.01 comparing the indicated group to PBS-vaccinated group as determined by ANOVA and Tukey’s post hoc t-test.

Next, we surveyed the draining (mandibular) lymph nodes (MLN) of vaccinated mice at 3 and 7 DPI to determine if the results found in the infected tissue mirrored that found in the organized lymphoid tissue most responsible for the generation of the adaptive immune response during acute infection. The results of surveying the MLN at these time points found the cell numbers reflected a very similar profile to that reported for the cornea for T lymphocytes (Fig. 6A) and myeloid cells (Fig. 6D). Specifically, CD4⁺ and CD8⁺ T cells (Fig. 6B, 6C) as well as the granulocyte (Fig. 6E, 6F) and monocyte/macrophage (Fig. 6G, 6H) populations were significantly reduced in the MLN of the high 0ΔNLS dose-vaccinated mice compared to the PBS- and low dose- vaccinated groups at 3 and 7 DPI. In addition to T cells, B lymphocyte numbers in the MLN of the high 0ΔNLS dose-vaccinated animals were considerably lower than the low dose- or PBS-vaccinated groups at 3 (Fig. 6B) and 7 (Fig. 6C) DPI. A similar profile was found in the cornea, TG, and MLN in mice vaccinated with 10⁴ PFU 0ΔNLS as that with 10⁵ PFU 0ΔNLS at 7 DPI (data not shown). The 3 DPI time point was not conducted with 10⁴ PFU 0ΔNLS-vaccinated animals. With some exceptions, the effectiveness of the vaccine in terms of reduction in HSV-1 found in a given tissue is inversely correlated with the leukocyte infiltrate and the lack of expansion of cells within the draining lymph nodes. We surmise the overall attenuated cellular response to infection in the surveyed tissue of vaccinated mice is due to greater control of viral replication and therefore, less antigen available to drive local and regional immune activation post challenge.

Figure 6. Absence of lymphocyte expansion in draining lymph nodes of HSV-1 infected mice vaccinated with high dose HSV-1 0ΔNLS.

Male and female mice were subcutaneously immunized with 10³ PFU HSV-1 0ΔNLS, 10⁵ PFU HSV-1 0ΔNLS, or vehicle (PBS) vaccine followed by an intramuscular boost three weeks later. Mice (n=5/group) were challenged with 1,000 PFU HSV-1 McKrae per eye 30 days following the final immunization. Mice were euthanized at 3 or 7 DPI, and the mandibular lymph nodes (MLN) were removed and processed to single cell suspensions. (A) Representative flow plot MLN distribution of CD4⁺ and CD8⁺ T cells in vaccinated mice. (B) Absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes residing in the MLN of vaccinated mice 3 DPI. (C) Absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes residing in the MLN of vaccinated mice 7 DPI. Uninfected absolute number of CD4⁺ and CD8⁺ T cells and CD19⁺ B lymphocytes were 386,139±13,075, 143,432±3,572, and 117,423±11,108 respectively. *p<.05, **p<.01, ***p<.001, comparing the high 0ΔNLS-vaccinated dose group to PBS-vaccinated group as determined by ANOVA and Tukey’s post hoc t-test. (D) Representative flow plot MLN distribution of myeloid cells in vaccinated mice designated as 1: (CD45⁺CD11b⁺Ly6G^midLy6C⁺), 2: (CD45⁺CD11b⁺Ly6G^highLy6C^int) and 3: (CD45⁺CD11b⁺Ly6G⁻Ly6C^high). Absolute number of granulocytes (CD45⁺CD11b⁺Ly6G^midLy6C⁺) and (CD45⁺CD11b⁺Ly6G^highLy6C^int) at 3 (E) and 7 (F) DPI. Absolute number of monocytes (CD45⁺CD11b⁺Ly6G⁻Ly6C^high) at 3 (G) and 7 (H) DPI. *p<.05, **p<.01 comparing the high 0ΔNLS-vaccinated dose to PBS-vaccinated group as determined by ANOVA and Tukey’s post hoc t-test.

Select cytokine and chemokines levels expressed in infected cornea and TG are dramatically reduced in HSV-1 0ΔNLS vaccinated mice

Chemokines including CCL2, CCL3, CCL5, CXCL1, and CXCL10 are expressed early in the cornea following HSV-1 infection (45–47). Neutralization or loss of select chemokines or their cognate receptor results in aberrant and often times, loss of leukocyte infiltration during acute corneal HSV-1 infection (38, 46, 48–53). Therefore, to determine if the reduction in corneal leukocyte infiltration in response to HSV-1 infection in the vaccinated mice correlated with chemokine expression, the expression of 10 chemokines was evaluated by suspension array. The results show a significant drop in CCL2, CCL3, CCL4, CCL5, CCL11, CXCL9, and CXCL10 in HSV-1-infected, vaccinated mouse corneas in accordance with the vaccine dose administered (Table I). Pro-inflammatory factors including IL-6 and IFN-γ were also found to be reduced as was IL-10 and VEGF-A in the vaccinated mouse cornea in a dose-dependent fashion (Table 1). Other chemokines and inflammatory molecules investigated including CXCL1, CXCL2, IL-1α, IL-1β, G-CSF, LIF, M-CSF, and TNF-α were all reduced in the vaccinated animals but the levels did not achieve significant (p<.05) differences in large part due to sample variation (Table I).

Table I.

Cytokine/Chemokine Expression in Cornea of HSV-1 0ΔNLS Vaccinated Mice^a

Cytokine/Chemokine	Vehicle	103 PFU 0ΔNLS	104 PFU 0ΔNLS	105 PFU 0ΔNLS
Eotaxin/CCL11	125 +/− 52	44 +/− 20*	13 +/− 6**	16 +/− 7**
G-CSF	1077 +/− 795	88 +/− 50	24 +/− 21	0 +/− 0
IFN-γ	63 +/− 19	11 +/− 8**	2 +/− 2**	0 +/− 0**
IL-1α	133 +/− 85	55 +/− 18	44 +/− 17	59 +/− 25
IL-1β	77 +/− 53	10 +/− 7	2 +/− 2	0 +/− 0
IL-6	158 +/− 78	28 +/− 18**	3 +/− 3**	0 +/− 0**
IL-10	16 +/− 9	2 +/− 2*	0 +/− 0*	0 +/− 0*
IP-10/CXCL10	3539 +/− 868	1474 +/− 731*	343 +/− 154**	260 +/− 111**
KC/CXCL1	1947 +/− 1261	494 +/− 399	417 +/− 290	46 +/− 28
LIF	106 +/− 67	40 +/− 34	3 +/− 3	2 +/− 2
MCP1/CCL2	3069 +/− 1466	728 +/− 514*	235 +/− 106*	96 +/− 57**
M-CSF	18 +/− 11	12 +/− 6	7 +/− 4	5 +/− 5
MIG/CXCL9	1188 +/− 422	240 +/− 71**	83 +/− 29**	80 +/− 28**
MIP-1α/CCL3	412 +/− 221	96 +/− 64*	41 +/− 13*	39 +/− 14*
MIP-1β/CCL4	618 +/− 275	118 +/− 54*	28 +/− 14**	20 +/− 13**
MIP-2/CXCL2	3751 +/− 3401	1599 +/− 1490	58 +/− 40	50 +/− 31
RANTES/CCL5	99 +/− 33	22 +/− 5*	18 +/− 6**	10 +/− 6**
TNF-α	16 +/− 10	2 +/− 2	0 +/− 0	0 +/− 0
VEGF-A	24 +/− 14	10 +/− 5	3 +/− 2*	0 +/− 0*

^aCorneas were collected at day 7 post infection, and processed for cytokine/chemokine content by suspension array. Numbers reflect pg/mg of each indicated analyte ± SEM, n=5–6 samples/group.

^**p<.01,

^*p<.05 comparing the indicated group to the vehicle control as determined by ANOVA and Dunnett’s multiple comparison test. Uninfected corneas had no detectable analyte expressed except those noted here (pg/mg ± SD, n=2/analyte): eotaxin, 7±2; IL-1α, 12±12; IL-10, 5±5; CXCL10, 15±15; CXCL1, 39±24

T cell and myeloid cell infiltration into the TG was also muted in the higher vaccinated dose animals following HSV-1 infection. Such results were reflected by the expression of cytokines and chemokines. Specifically, CCL2, CCL3, CCL4, CCL5, CCL11, CXCL9, CXCL10, IFN-γ, and LIF expression were also significantly reduced in the TGs from the higher dose (10⁴-10⁵ PFU HSV-1 0ΔNLS) vaccinated mice compared to the vehicle or lower dose immunized animals (Table II). Although expression was modest, TNF-α levels in the TG were significantly reduced in all vaccinated mice compared to the vehicle control vaccinated group (Table II). Other immune mediators including CXCL1, G-CSF, IL-6, and M-CSF were lower in the higher vaccinated mouse TG but due to variability, did not reach significance (Table II). Neither IL-1β or VEGF-A was detectable above background (Table II) whereas CXCL2 and IL-10 showed high background levels (in uninfected animals) and therefore, were not included in the analysis (data not shown). Taken together, the measurement of analytes in the cornea and TG reveal a strong correlation with the reduction of leukocyte infiltration, control of viral replication in these tissues, and antibody neutralization titers observed in the high dose vaccinated mice 7 DPI.

Table II.

Cytokine/Chemokine Expression in Trigeminal Ganglion of HSV-1 0ΔNLS Vaccinated Mice^a

Cytokine/Chemokine	Vehicle	103 PFU 0ΔNLS	104 PFU 0ΔNLS	105 PFU 0ΔNLS
Eotaxin/CCL11	1435 +/− 363	633 +/− 157	188 +/− 82*	94 +/− 26**
G-CSF	562 +/− 268	389 +/− 231	12 +/− 12	0 +/− 0
IFN-γ	902 +/− 27	402 +/− 192	38 +/− 26*	0 +/− 0**
IL-1β	3 +/− 3	4 +/− 4	0 +/− 0	0 +/− 0
IL-6	744 +/− 275	613 +/− 290	0 +/− 0	0 +/− 0
IP-10/CXCL10	15,194 +/− 2,777	11,528 +/− 3,919	1,648 +/− 865*	436 +/− 373**
KC/CXCL1	505 +/− 203	399 +/− 238	27 +/− 27	6 +/− 4
LIF	155 +/− 52	71 +/− 31	4 +/− 4*	0 +/− 0**
MCP1/CCL2	3,634 +/− 1,241	1,097 +/− 563	79 +/− 48*	12 +/− 9**
M-CSF	48 +/− 13	48 +/− 20	7 +/− 7	0 +/− 0
MIG/CXCL9	2,310 +/− 423	1,655 +/− 457	444 +/− 276*	103 +/− 92**
MIP-1α/CCL3	1,964 +/− 375	865 +/− 388	75 +/− 35*	54 +/− 30**
MIP-1β/CCL4	3,238 +/− 656	1,380 +/− 646	41 +/− 41**	0 +/− 0**
RANTES/CCL5	430 +/− 76	263 +/− 74	28 +/− 21**	0 +/− 0***
TNF-α	33 +/− 16	0 +/− 0*	0 +/− 0*	0 +/− 0*
VEGF-A	0 +/− 0	0 +/− 0	6 +/− 6	0 +/− 0

^aTrigeminal ganlia were collected at day 7 post infection, and processed for cytokine/chemokine content by suspension array. Numbers reflect pg/mg of each indicated analyte ± SEM, n=8–10 samples/group.

^***p<.001,

^**p<.01,

^*p<.05 comparing the indicated group to the vehicle control as determined by ANOVA and Dunnett’s multiple comparison test. Uninfected corneas had no detectable analyte expressed except those noted here (pg/mg ± SEM, n=3/analyte): eotaxin, 54±54; CXCL10, 16±16; CXCL1, 39±24

Corneal neovascularization is greatly diminished in high dose HSV-1 0ΔNLS vaccinated mice

Corneal neovascularization is a hallmark of herpetic stromal keratitis in rodents and humans alike (8). Previously, we reported a loss of vision in mice as a result of HSV-1 infection could be linked to severe corneal opacity and gross neovascularization (36). Therefore, we investigated the impact of vaccine doses on the severity of corneal blood and lymphatic vessel genesis at 30 DPI, a time when maximum angiogenesis is evident (54). Our results reflect the dose-response efficacy of the vaccine with the higher vaccine doses (10⁴-10⁵ PFU HSV-1 0ΔNLS) preventing corneal hem- and lymph-angiogenesis in comparison to the low (10³ PFU HSV-1 0ΔNLS) vaccine dose which showed a similar profile to vehicle-vaccinated mice (33) (Fig. 7A–D). Since the vehicle-vaccinated mice do not survive out to 30 DPI to a significant degree (Fig. 1B), we assessed corneal opacity in the animals at 7 DPI via slit lamp exam using a masked observer. Similar to what is observed with neovascularization, mice that received the higher vaccine doses displayed minimal corneal opacity compared to mice that received low dose or vehicle vaccine (Fig. 7E). Collectively, these results concur with the data above demonstrating the efficacy of the HSV-1 0ΔNLS vaccine against ocular viral challenge is greatly attenuated at immunization doses below 10⁴ PFU.

Serum from mice immunized with the high HSV-1 0ΔNLS vaccine dose immunoprecipitates HSV-1 proteins not recognized using the low HSV-1 0ΔNLS vaccine dose

Since the high vaccine dose demonstrated superior efficacy in protecting mice against HSV-1 in comparison to mice immunized with the low dose, we hypothesized viral proteins may be uniquely recognized by antibody from the antiserum of high dose vaccinated mice. To test this hypothesis, antiserum obtained from high and low dose immunized mice was evaluated for recognition of HSV-1 antigens. A total of 15 HSV-1 proteins were recognized by antiserum from high dose immunized mice significantly above that recognized by low dose vaccinated animals (Table III). Furthermore, 2 additional HSV-1 proteins (UL18/VP23 and UL35/VP26) were recognized by the high and low HSV-1 0ΔNLS immunized mice that were significantly above the level displayed by serum from the vehicle-vaccinated mice (Table III). Additional HSV-1 proteins were precipitated by the HSV-1 0ΔNLS vaccinated mice but the abundance of recognition did not reach significance compared to the vehicle-immunized mice due to the variability from animal to animal.

Table III.

HSV-1 Protein Recognition by Antiserum from Vaccinated Mice^a

HSV-1 Protein	High Dose (105 PFU)	Low Dose (103 PFU)	Vehicle
UL39/ICP6/RR1	6,793,035 ± 1,303967ΔΔ	352,954 ± 54,369	372,251 ± 25,242
UL19/MCP/VP5	6,696,422 ± 890,026ΔΔ	1,843,337 ± 869,108	535,912 ± 151,606
UL29/ICP8/DBP	5,597,621 ± 1,263,758ΔΔ	130,614 ± 90,589	94,761 ± 50,909
UL27/gB	4,453,171 ± 884,210ΔΔ	106,422 ± 63,346	22,788 ± 11,578
UL22/gH	2,223,914 ± 550,125ΔΔ	12,540 ± 12,540	0 ± 0
UL18/TRX2/VP23	2,335,474 ± 487,355**	967,491 ± 392,761*	330,052 ± 60,015
UL40/RR2	1,773,914 ± 513,782ΔΔ	85,050 ± 24,622	89,785 ± 21,998
UL44/gC	1,329,966 ± 253,999ΔΔ	20,216 ± 9,945	0 ± 0
US6/gD	1,026,939 ± 191,329ΔΔ	0 ± 0	0 ± 0
UL38/TRX1/VP19c	827,451 ± 263,020*	219,067 ± 125,881	28,706 ± 9,146
US8/gE	826,764 ± 169,833ΔΔ	72,489 ± 18,495	41,676 ± 9,301
UL35/SCP/VP26	779,290 ± 127,433**	239,978 ± 160,633*	28,879 ± 22,879
UL1/gL	554,211 ± 147,467ΔΔ	1,386 ± 1,386	721 ± 721
UL48/VP16	584,616 ± 145,499ΔΔ	99,682 ± 61,382	32,467 ± 15,242
UL31/NEC1	584,149 ± 204,252Δ	105,703 ± 28,173	63,740 ± 23,269
UL26/VP24/21	414,514 ± 125,223*	130,212 ± 72,319	41,565 ± 9,047
UL12/NUC	372,549 ± 87,858ΔΔ	65,219 ± 29,728	83,159 ± 35,251
UL25/CVC2	309,826 ± 75,562Δ	104,840 ± 49,050	51,411 ± 25,687
RS1/ICP4	277,967 ± 73,595Δ	67,645 ± 18,463	29,843 ± 12,568

^aSerum from naive WT or HSV-1 0ΔNLS immunized mice (high and low dose) was used to immunoprecipitate viral encoded proteins from HSV-1 infected Vero cells. Precipitated proteins were analyzed by mass spectrometry. Proteins derived from HSV-1 were identified by cross-referencing derivative peptide ions with a reference sequence database. Numbers reflect matched peptide abundance/intensity per protein ± SEM by antiserum from HSV-1 0ΔNLS or PBS (vehicle) vaccinated mice (n=7/group from 3 independent experiments).

^ΔΔp<.01,

^Δp<.05 comparing the high dose to the other groups of vaccinated mice;

^**p<.01,

^*p<.05 comparing the indicated group to the vehicle group as determined by ANOVA and Scheffé multiple comparison test.

Passive immunization and cornea pathology

In order to further characterize the relevance of differences in protein recognition of sera from mice immunized with the high (10⁵ PFU) versus low (10³ PFU) dose of the 0ΔNLS vaccine, naive mice were passively immunized with antiserum from the high and low dose vaccinated animals and subsequently challenged with HSV-1. Relative to mice receiving sera from PBS-vaccinated (naive control) animals, recipients of sera from either dose of 0ΔNLS vaccine were found to be resistant to HSV-1 challenge in terms of cumulative survival (Fig. 8A). However, there were marked differences between mice receiving sera from high versus low 0ΔNLS dose vaccine in terms of mechanosensory function (Fig. 8B), opacity (Fig. 8C), and neovascularization including lymphangiogenesis (Fig. 8D–8G). Specifically, there was minimal corneal sensation loss in the high dose sera recipients following HSV-1 infection compared to the other two groups with a 44% loss in low dose sera recipients and 96% loss in the naïve sera recipients by 7 DPI (Fig. 8B). Likewise, HSV-1-infected mice receiving sera from high dose vaccinated animals possessed cornea opacity levels similar to uninfected mice and lower than mice receiving sera from low dose- or PBS-vaccinated, HSV-1-infected animals (Fig. 8C). Equally revealing is the genesis of blood and lymphatic vessels in the cornea in response to HSV-1. High dose sera recipients displayed little to no cornea neovascularization as a result of HSV-1 challenge whereas there was notable vessel growth in the cornea from low dose sera recipients albeit lower than naive sera recipients (Fig. 8D–8G). Collectively, the results clearly demonstrate differences between recipients of sera from high versus low dose 0ΔNLS vaccinated mice in terms of corneal pathology that likely relates back to coverage of antigen recognition by antibody from these vaccinated animals.

Figure 8. Passively immunized mice with sera from high 0ΔNLS dose-vaccinated mice protects against HSV-1-mediated corneal pathology.

Sera (250 μl) from male and female mice (n=10/group) that were immunized and boosted with 10³ PFU HSV-1 0ΔNLS, 10⁵ PFU HSV-1 0ΔNLS or vehicle (PBS) was administered ip to naive recipients 24 hr prior to infection with HSV-1 (1,000 PFU/cornea). (A) Mice were monitored for cumulative survival out to 21 DPI. **p<.01 comparing the recipients of sera from 0ΔNLS vaccinated groups to the PBS-vaccinated sera recipients as determined by the Mantel-Cox test. (B) Over the course of the first 7 days post infection, cornea sensation was evaluated using a Cochet-Bonnet esthesiometer comparing the recipients of sera from 0ΔNLS vaccinated groups to each other and to the PBS sera recipients. **p<.01 comparing the recipients of sera from 0ΔNLS vaccinated mice to the recipients of sera from PBS-vaccinated mice, and ^p<.05 comparing the recipient of sera from the 10⁵ 0ΔNLS vaccinated mice to that of recipients of sera from the 10³ 0ΔNLS vaccinated animals as determined by ANOVA and Tukey’s t-test. Only 5 mice from the PBS sera recipient groups could be evaluated at 7 DPI due to mortality. (C) The corneas of passively immunized mice infected with HSV-1 were surgically removed from exsanguinated animals that survived out to 21 DPI and assessed for opacity measuring the optical density at 500 nm wavelength in a 30×30 matrix over the cornea surface. Uninfected mouse corneas served as the baseline control (dotted line). (D) The corneas from C were then stained for lymphatic (LYVE-1) and blood (CD31) vessels. Metamorph quantification of corneal area containing LYVE-1⁺ and CD31⁺ vessels. **p<.01, *p<.05 comparing the PBS serum immunized group to all other groups. ##p<.01 comparing the 10⁵ 0ΔNLS passively immunized mice to the PBS- and 10³ 0ΔNLS passively-immunized mice as determined by ANOVA and Tukey’s t-test. Representative images for (E) PBS-, (F) low 10³ 0ΔNLS-, and (G) high 10⁵ 0ΔNLS-passively immunized mice are shown

DISCUSSION

In the present study, we compared different doses of the HSV-1 0ΔNLS vaccine in mice subsequently challenged with HSV-1 to define the minimum effective dose that affords the host protection against ocular HSV-1 infection. Whereas all doses were found to be highly effective in terms of cumulative survival, there was a distinct difference in the lack of efficacy of the low dose (10³ PFU) vs higher doses (10⁴-10⁵ PFU) of the HSV-1 0ΔNLS vaccine in nearly all other aspects of protection measured in the infected tissue including virus replication and spread, establishment of latency, inflammation including cytokine and chemokine expression and leukocyte infiltration, and corneal pathology including opacity and neovascularization. These findings were inversely correlated to the neutralizing antibody titer from the vaccinated mice; the higher the antibody titer, the lower the inflammatory profile and reduction in virus replication and spread. It is worth noting the expansion of the lymphoid population observed in the vehicle or low dose vaccinated mice was not evident in mice that received the higher doses which we interpret to suggest antigen abundance is limited in high dose vaccinated mice likely due to control of the infection by antibody. This result is not to say T cells are not involved in this process as there is ample evidence by numerous investigators using mutant HSV-1 as vaccines that T cells play a role either directly as effector cells and/or facilitate the antibody response against HSV-1 (24, 55, 56). However, the present study was focused on antibody as a means to further delineate differences in vaccine doses that might be explained by antigen recognition.

Since there was a striking difference in vaccine efficacy measuring corneal keratitis comparing the high to low dose vaccine, we considered the possibility that unique HSV-1 antigens might be recognized by antiserum from the high dose vaccinated mice not recognized by low dosed vaccinated animals. Analysis of isolated proteins by mass spectrometry from immunoprecipitation runs identified 15 virus-encoded proteins recognized by the antiserum from the high dose vaccinated group significantly beyond that of the antiserum from the low dose vaccinated or naive (vehicle) vaccinated groups. One family of recognized proteins included six HSV-1 encoded glycoproteins (g)B, gH, gC, gD, gE, and gL. Of the two glycoproteins with the highest reactivity score to the antiserum, both recombinant gB and gH or derived peptides used as immunogens or antagonists have been reported to suppress HSV replication, block anterograde or retrograde spread of the virus, and/or prevent virus-associated disease through a robust T cell or antibody response to the antigen (23, 29, 31, 57, 58). Recombinant gC and gD have also been evaluated as prototypical vaccines against HSV-1 administered prophylactically or therapeutically with reported success (19–21, 25, 59). Similar to the other HSV-1 glycoproteins, gE and gL alone or in combination with other HSV-1 glycoproteins when used as vaccines have been reported to protect mice from a lethal HSV-1 challenge (60–62).

The non-structural intracellular proteins primarily recognized by the antiserum from high dose HSV-1 0ΔNLS vaccinated mice which includes pUL39/ICP6, pUL29/ICP8, pUL40/RR2, pUL31/NEC1, pUL12/NUC, and Rs1/ICP4 are all important elements in virus replication. pUL31, a component of the nuclear egress complex, is thought to promote the selective process in infectious virus particle assembly along with pUL17 and pUL25 (63). ICP8 is a single-stranded DNA binding protein critical for efficient annealing of complementary DNA and therefore, essential for DNA replication during infection (64). pUL12 exonuclease is important in the packaging of viral DNA into infectious virus in which mutants in the UL12 gene show a significant loss in the production of infectious virus (65). One of the more intriguing outcomes in the proteomic analysis was the recognition of pUL39 and pUL40, the former the most highly recognized protein by the antiserum from the high dose vaccinated mice. Early work reported a pUL39 null mutant replicated poorly in vitro and in vivo and did not cause ocular disease following cornea infection (66). Human and mouse data also suggest this protein is a specific target of CD8⁺ T cells residing in the TG following infection (67, 68). The other subunit of the ribonucleotide reductase, pUL40, is recognized by T cells from asymptomatic seropositive patients infected with HSV-2 and has been found to be a highly protective immunogen when used to vaccinate guinea pigs against subsequent challenge with HSV-2 (69). The other non-structural intracellular protein, ICP4, found to be selectively recognized by antiserum from high dose vaccinated mice has been found to be instrumental in driving vascular endothelium growth factor A-induced corneal neovascularization in response to HSV-1 infection (70). This recognition and predicted neutralization of ICP4 are consistent with the reduced level of corneal neovascularization in the high dose vaccinated mice compared to the low dose-vaccinated animals following ocular HSV-1 challenge (Fig. 7).

A third group of HSV-1 proteins recognized by the antiserum from high dose vaccinated mice includes the tegument and capsid proteins pUL19/VP5, pUL48/VP16, and pUL25/CVC2. These proteins along with the other capsid proteins recognized by either high or low dose vaccinated mice including pUL18/VP23, UL38/VP19c, UL35/VP26, and UL26/VP24/21 are critical in HSV-1 replication whether it be transactivation of viral immediate early lytic genes (VP16) or incorporation and assembly of capsid proteins into the capsid shell (71–73). Since VP5, VP19c, and pUL25 contribute to the long-range axonal transport of the infectious unit of the virion to the neuronal cell bodies through an assembled capsid-associated tegument complex (74), it is quite possible the antiserum recognizes only a few epitopes on a single protein entity that results in the immunoprecipitation of the majority of the complex. Consequently, one or more of these proteins may not be a contributing member of the protective immune repertoire of antigens recognized by the antiserum from the high dose 0ΔNLS vaccinated mice. Another possibility is the complex itself forms a structural epitope recognized by the antibody that contributes to protection from virus spread, replication, and establishment of latency. Thus, analysis of single viral-encoded proteins that collectively form the capsid associated tegument protein complex would afford little to no protection when used as a vaccine as the “protective” epitope if it is only formed by the associated protein complex. While this may be true for antibody recognition, epitopes representing capsid or tegument proteins have been found to generate a robust CD8⁺ T cell response with polyfunctional effector T cells that elicit a protective immune response against HSV-1 keratitis (30, 75). The importance of HSV-1 protein recognition by the antisera is underscored by the passive immunization results that show quantifiable corneal pathology is reduced or absent in high 0ΔNLS dose-vaccinated mice compared to the low dose- or PBS-vaccinated control following HSV-1 challenge (Fig. 8). Even though survival was similar between the high and low 0ΔNLS-vaccinated mice and significantly above the PBS-vaccinated control group, the difference in cornea pathology between the 0ΔNLS immunized groups demonstrates the need to incorporate a more encompassing approach in evaluating the success of a vaccine to protect against an ocular pathogen, a criteria often overlooked or only accomplished subjectively from most labs in the HSV-1 field. In the case of the current investigation, further studies are required to clearly elucidate those specific viral-encoded tegument and capsid proteins as well as other identified viral proteins that are contributors to the HSV-1 0ΔNLS vaccine efficacy.

Abbreviations used:

TG

trigeminal ganglia

PFU

plaque forming unit

g

glycoprotein

DPI

days post infection

ICP

infected cell protein

MLN

mandibular lymph node