CD137 costimulation is associated with reduced herpetic stromal keratitis and with developing normal CD8⁺ T cells in trigeminal ganglia

Abstract

Costimulatory interactions can be critical in developing immune responses to infectious agents. We recently reported that herpes simplex type 1 (HSV-1) infections of the cornea require a functional CD28-CD80/86 interaction to not only reduce the likelihood of encephalitis, but also to mediate herpetic stromal keratitis (HSK) following viral reactivation. In this same spirit we decided to determine the role that CD137 costimulation plays during HSK. Using both B6-CD137L^-/- mice, as well as antagonistic and agonistic antibodies to CD137 we characterize the immune response and to what extent CD137 plays an important role during this disease. Immune responses were measured in both the cornea and in the trigeminal ganglia where the virus forms a latent infection. We demonstrate that CD137 costimulation leads to reduced corneal disease. Interestingly, we observed that lack of CD137 costimulation resulted in significantly reduced CD8⁺ T expansion and function in the trigeminal ganglia. Finally, we showed that viruses that have been genetically altered to express CD137 display significantly reduced corneal disease, though they did present similar levels of trigeminal infection and peripheral virus production following reactivation of a latent infection. CD137 interactions lead to reduced HSK and are necessary to develop robust trigeminal CD8⁺ T cell responses.

Introduction

Corneal infection with herpes simplex virus one begins with proliferation first at the site of infection followed by infection of nerves found in the cornea and the subsequent retrograde transport of virus to the trigeminal ganglia [1–3]. Once in the ganglia, the virus replicates there, and some of these newly formed viruses return to the periphery [3], but within a week infectious virus typically can no longer be detected in the TG of wild-type mice as the virus establishes a latent infection of the neurons in the TG [2–6]. During latency, the viral genome is present, but few active virions are detected in these infected neurons, however they do express viral RNA, with latency associated transcripts being one of the most abundant in these latently infected neurons [5, 6]. This latent infection can be interrupted by various forms of stimuli including immunosuppressive events such as fever, menses, sunlight (UV), irradiation, stress, and trauma [7–11]. Following reactivation, new viruses are formed in the neurons of the trigeminal ganglia and the virus travels via anterograde axonal transport back to the epithelial surface. Once there the virus replicates and re-stimulates the existing host immune response which is the primary mechanism leading to recurrent herpetic stromal keratitis (r-HSK) [12–14]. This disease is a leading cause of infectious blindness in the Western world with one study determining a prevalence of HSV keratitis of 149/100000 people [15].

In humans, symptoms are not typically seen during primary infection with HSV-1 and those that do occur are mostly in children and the immunosuppressed adults. Thus, primary disease is most often clinically asymptomatic, although in 1–6 % of cases it presents as blepharo-conjunctivitis that heals without scarring [8].

As a consequence, the dominant form of clinical disease is the result of reactivation of virus. This recurrent disease in the cornea is an immunopathologic condition that is initiated by renewed presence of virus in the cornea which re-stimulates the immune response leading to inflammation of the cornea resulting in damage to the cornea. In humans, the inflammatory infiltrate in HSK is characterized by influx of a phenotypically diverse population of leukocytes consisting of lymphocytes, neutrophils, and mononuclear phagocytes [6, 7]. Animal studies have shown that the cell type found in greatest numbers in corneas displaying disease are neutrophils [16, 17]. In addition, many studies have conclusively determined that in the absence of T cells HSK does not occur [18–21].

The activation of T cells requires not only engagement of the TCR with the MHC +peptide complex, but also costimulatory interactions between the T cell and the antigen presenting cell. The dominant costimulatory interaction for naïve T cells is that between the T cell’s CD28 and the APC’s CD80 or CD86 surface antigen. We recently reported that if this costimulation does not occur then corneal disease also does not occur [22]. This indicates the primacy of this costimulation in the development of herpetic corneal disease. In addition to this costimulatory interaction, there are several other costimulatory interactions that have been investigated during primary HSK. These include CD40 with CD40L [23], OX40 with OX40L [24], and CD137 with CD137L [25]. We decided to revisit the role that CD137 costimulation plays. We made this decision based on the extensive literature that indicated that augmentation of this costimulation was seen to increase direct antiviral immune responses, particularly in the development and expansion of CD8 T cell responses [26–30] and in some cases even augmented antibody responses [31] and increased B cell survival and proliferation [32]. We have shown in other studies that in vivo primed CD8 T cells when transferred to infected mice results in significantly reduced corneal disease [33]. Furthermore, we reported that a vaccine that significantly prevents rHSK is at least partially associated with increased viral specific antibody responses [34], an observation seen with other vaccines that are effective in controlling primary HSK [35, 36]. Consequently, we set out to determine the consequences of ocular infection with HSV-1 on the development of both primary and recurrent HSK in mice where this interaction does not occur. We will present evidence demonstrating that in the absence of CD137 costimulation animals experience significantly worse corneal disease. In addition, CD8⁺ T cell immune responses in the trigeminal ganglia, are more significantly impaired, than seen with wild-type mice. Furthermore, mice infected with a CD137L expressing HSV-1 virus did not result in significant disease.

Methods

Mice

Investigations with mice conformed to the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. C57BL/6 (B6) were purchased from NCI. The B6.Cg-Tyr^c-J/J mice (albino B6 mice, B6-TyC) were originally acquired from Jackson Labs and bred in our colony. These mice are included in these studies because they display slightly greater corneal disease following UV-B reactivation than do standard B6 mice (data not shown). The B6.CD137L KO (B6-CD137L^-/-) [37] mice were obtained initially from Amgen and then maintained in our breeding colony with periodic genotyping to assure that they maintained their lack of CD137L expression.

Infection of mice

For all experiments described, we used age matched male and female mice from the aforementioned strains that were 6–10 weeks old. For primary disease we infected scarified corneas of C57BL/6 (B6) and B6-CD137^-/- mice with 2×10⁵ p.f.u. of the RE strain of HSV-1. This strain was used for primary HSK because it does not have the mortality associated with other more neurovirulent strains of HSV-1 [38]. Eye swabs were taken from these mice for 7 to 10 days following infection to monitor the persistence of viral shedding. For recurrent disease, mice were infected on the scarified cornea with 10⁶ p.f.u. HSV-1 McKrae strain as previously described [34]. Each mouse received an intraperitoneal (IP) injection of 0.5 ml pooled human serum (Sigma Chemicals, St. Louis MO; ED50 for virus neutralization=1 : 1600) concurrent with infection. Administration of human anti-HSV antibodies at the time of ocular infection has been shown to protect mice from death and corneal disease during primary infection, while allowing for the establishment of latency and subsequent reactivation of virus after corneal UV-B exposure. These human antibodies are undetectable at the time of UV-B irradiation, 5 weeks after primary infection. To confirm infection, only mice displaying infectious virus obtained from eye swabs taken 3 days post-infection were used for subsequent reactivation [21]

UV-B irradiation and virus reactivation

Mice were reactivated from latency as previously described [21, 22]. Briefly, the eyes of all latently infected mice were examined for corneal opacity prior to UV-B irradiation, and only animals with clear corneas were used, and mice with residual corneal disease from the primary infection are removed from the study. At least 5 weeks after primary infection, the eyes of latently infected and control mock-infected mice were exposed to 250 mJ cm⁻² of UV-B light using a TM20 Chromato-Vu transilluminator (UVP, Inc., San Gabriel, CA), which emits UV-B at a peak wavelength of 302 nm. Irradiated mice were swabbed with sterile cotton applicators from day 0 to day 7, unless otherwise indicated. The swab material was cultured on Vero cells, as described above, to detect recurrent virus shedding from the cornea. Reactivation was defined as the finding of any HSV-positive eye swab on any day after UV-B exposure, with day 0 swabs serving as a control [22].

Clinical evaluation

On the designated days after viral infection or UV-B reactivation, a masked observer, who is unaware of the experimental groups, examined the mouse eyes through a binocular dissecting microscope to score clinical disease. Stromal opacification was rated on a scale of from 0 to 4, where 0 indicates clear stroma, 1 indicates mild stromal opacification, 2 indicates moderate opacity with discernible iris features, 3 indicates dense opacity with the loss of defined iris detail except pupil margins, and four indicates total opacity with no posterior view. Corneal neovascularization was evaluated as described previously [21, 22] using a scale of from 0 to 8, where each of the four quadrants of the eye is evaluated for the density of vessels that have grown into them. Periocular disease was measured in a masked fashion on a semiquantitative scale as previously described [39]. In the later studies we included corneal blink reflex as a means of determining corneal nerve damage. This was accomplished by loosely holding the mouse and touching the cornea with the blunt tips of a surgical forceps without touching the eyelashes and whiskers. The cornea was divided into five areas (four quadrants and centre area), loss of blink reflex referred to the inability of the mouse to blink when an area was touched and was recorded as 0. Positive blink reflex referred to the ability to blink when an area of the cornea was touched and was recorded as 1. The total score of the five areas would be the final score of corneal blink reflex for a mouse. A score of 0 indicated a complete loss of corneal sensation such that the mouse failed to blink when any area of the cornea was touched. A score of 5 indicated retention of some degree of sensation such that the mouse blinked when any area of the cornea was touched [39]. Note: uninfected, UV-B irradiated control mice were used as a baseline for any effects due to UV-B irradiation.

In vivo antibody treatment

For blockade CD137, 150 µg α4-1BBL (CD137), clone TKS-1, IgG2a) (Bio X Cell) was administrated on day −1, day 1 and 3 days post-infection. Likewise, agonistic CD137 IgG1k antibody (3H3, Millipore Sigma) treatment was administered in the same fashion. Control mice received a similar amount of a rat IgG isotype control antibody (clone GL113). All antibodies were administrated i.p. in 400 µl PBS.

Flow cytometric analysis

Cells were isolated from corneas and trigeminal ganglia as previously described [22]. Briefly, corneas or trigeminal ganglia were excised at defined time points and incubated in PBS-EDTA at 37 °C for 15 min at 37 °C. Stromas were separated from overlying epithelium and digested in 84 U collagenase type 1 (Sigma-Aldrich, St. Louis, MO) per cornea for 2 h at 37 °C and then were triturated to form a single-cell suspension. Suspensions were filtered through a 40 -µm cell strainer cap (BD Labware, Bedford, MA) and washed and then stained. Suspensions were initially stained with: PerCP-conjugated anti-CD45 (clone 30-F11, from BioLegend, San Diego, CA) and this was used to gate cells of bone marrow origin. For cornea suspension, cells were further stained with Alexa Fluor700-Gr-1 (clone RB6-8C5) and APC F4/80 (clone BM8) (from BioLegend, San Diego, CA); FITC conjugated anti-CD4 (clone RM4–5), PE-conjugated anti-CD8α (clone 53–6.7) (from PharMingen) and PE-conjugated CD11b (clone M1/70) (from eBiosciences, San Diego, CA). The strategy for analysis was to initially gate on live cells and then the CD45⁺ cells. These CD45⁺ cells were further evaluated for T cell markers CD4 and CD8, or for macrophage markers F4/80⁺CD11b⁺GR-1^-, or neutrophil markers GR-1⁺,CD11b⁺F4/80^-. Cells isolated from the TG were stained for T cell markers (CD4 and CD8), for the HSV-1 gB tetramer to determine the number of antigen-specific T cells. Intracellular staining was performed following fixation and permeabilization (using BD Bioscience’s Cytofix/Cytoperm kit) the proteins measured were: IFNɣ (clone XMG1.2) (from BioLegend), CD107a (clone 1D4B) (from BD PharMingen), TNF-α (clone MP6-XT22) (from BD PharMingen) to determine expression of these proteins and thus the functional activity of the cells were established. Cells were then analysed on a flow cytometer (FACSAria with FACSDIVA data analysis software; BD Biosciences).

Creation of CD137L expressing HSV-1 virus

We constructed a McKrae strain of HSV-1 that expresses CD137L (Fig. 1). We chose the McKrae strain because it displays two very important phenotypes that other strains of HSV-1 lack. The first being that it is one of the very few that can be induced to reactivate in mice on a consistent basis [10, 22]. The second is that it is one of the most neurovirulent strains of HSV-1 and infections of B6 mice lead to ocular disease [22]. Therefore, this virus will be of great use not only in our research, but also for others who wish to study pathogenesis of HSV-1. This virus was constructed using a previously described strategy of homologous recombination that has been used to insert genes into the HSV-1 genome [40]. We created a plasmid that possesses flanking sequences from HSV-1, McKrae strain that will allow for insertion of the plasmid into the intergenic UL49.5 and UL50 location of the HSV-1 genome (Fig. 1a). In order to ensure strong expression of CD137L protein we also cloned a CMV promotor sequence to drive its expression. The insertion of this plasmid into the resulting HSV-1 virus was confirmed by PCR amplification of DNA isolated from plaque picks of the recombinant virus grown on VERO cells. This was accomplished by using a primer pair that overlaps the flanking sequence of HSV-1 with the CD137L DNA. Primers used were: (forward), 5′-AAA ACC TCC CAC ATC TCC CC-3′; (reverse), 5′-CTT GTG AAA CCC GAC AAC CC-3′ (made by Integrated DNA Technologies, Coralville, IA). The virus was then further subcloned twice to assure that the plasmid is stably inserted into the HSV-1 genome. We have since determined that this virus displays strong expression of CD137L (Fig. 1b). This was accomplished by infecting VERO cells at an MOI of 5 and harvesting cells at 18 h and staining with anti-CD137L (clone 14, rabbit anti-mouse CD137L, Invitrogen, Cat #MA5-29838). When the growth characteristics of this virus were compared to wild-type McKrae HSV-1 in VERO cells there were no differences observed in either a one step or multistep growth (data not shown).

Fig. 1.

(a) Scheme for creating CD137L expressing HSV-1. Shows the location of the HSV-1 genome where the CD137L gene is inserted (intergenic region between 49.5 and 50). (b) Immunofluorescent staining of VERO cells at 16 h post-infection with an anti-CD137L antibody.

Q-PCR for HSV-1 latent genomes of trigeminal ganglia

Quantitation of HSV-1 genomes in the trigeminal ganglia of infected mice were performed as previously described [22]. Briefly, 6- to 8-week-old B6 and B6-CD137KO mice were infected as described above. Trigeminal ganglia were harvested on the indicated days post-infection, and genomic DNA was extracted using the DNeasy blood and tissue kit from QIAGEN (Cat #69506). The number of latent genomes per trigeminal ganglion was determined using primers to the thymidine kinase (tk) gene of HSV-1 designed to amplify a 70 bp fragment. The sequence of these primers is as follows: Tk (forward), 5′-CCAAAGAGGTGCGGGAGTTT-3′; Tk (reverse), 5′-CTTAACAGCTGTCAACAGCGTGCCG-3′. A standard curve of purified HSV-1 infectious chromosomal DNA was diluted in the background of mouse DNA in ten-fold dilutions from 10⁶ to 10¹ copies and used as a standard curve for determination of the total genome copy number in latently infected trigeminal ganglia. Adipsin gene was used as internal reference as follows, mouse adipsin forward primer was 5′-AGTGTGCGGGGATGCAGT-3′; reverse primer was 5′-ACGCGAGAGCCCCAGGTA-3′. Trigeminal ganglion DNA harvested from uninfected mice was used to generate mouse DNA standards from 10⁵ to 10¹ copies, from which the total copy number of mouse adipsin in each of the infected trigeminal ganglion samples was determined. Each value for tk copy number was normalized to the lowest value of the mouse adipsin copy number, and the number of copies of genome per trigeminal ganglion was then expressed on a log scale.

Statistics

All statistical analyses were performed with the aid of Sigma Stat for Windows, version 2.0 (Jandel, Corte Madera, CA). The log rank test was used to compare disease scores. Student’s unpaired t-test was used to compare virus titre data, genome copies, LAT +cells and flow cytometry numbers.

Results

To better understand the role of CD137 costimulation on the development of corneal disease following infection with HSV-1, we infected both C57BL/6 and B6-CD137L KO mice with the RE strain of HSV-1. As seen in Fig. 2, mice lacking CD137 demonstrated significantly increased corneal opacity and neovascularization scores when compared to wild-type mice at all time points. This indicated that mice lacking an intact CD137 costimulatory pathway experienced significantly worse corneal disease than their wild-type B6 counterparts.

Fig. 2.

B6-CD137L KO mice display increased acute HSK when compared to wild-type B6 mice when infected with 2×10⁵ p.f.u. of HSV-1, RE strain. B6-CD137L KO mice (n=20) were compared with wild-type B6 mice (n=20) for corneal opacity (a) and corneal neovascularization (b). B6-CD137L KO displayed significantly greater corneal opacity and neovascularization for all time points measured (p values ranged from <0.01–0.001 by Rank Sum analysis).

Next, we tested whether CD137 costimulation plays a role in recurrent HSK by establishing a latent HSV-1 infection in both C57BL/6 and CD137L KO mice with the McKrae strain of HSV-1. At 5 weeks post-infection, when latency is fully established and no mice were shedding virus into the tear film, we reactivated the virus by irradiating their eyes with UV-B light. We fully expected that latently infected CD137L KO mice, following UV-B induced reactivation, would demonstrate greater corneal disease as had been seen during acute infection (Fig. 2). The data from these studies confirmed this hypothesis as CD137LKO mice did present increased recurrent HSK as evidenced by greater opacity, greater neovascularization than did wild-type B6 mice (Fig. 3). This further established that mice lacking an intact CD137 costimulatory pathway were not able to control HSV-1 induced recurrent HSK as well as wild-type mice did. Interestingly, the pattern of virus shedding was indistinguishable between these two strains following reactivation (Fig. 4) and the degree of viral latency as determined by viral genome copy numbers also did not show differences between these two strains of mice (data not shown).

Fig. 3.

B6-CD137L KO mice displayed significantly greater recurrent HSK when compared to wild-type B6 mice. Eyes of mice were infected with 10⁶ p.f.u. of HSV-1, McKrae strain. Six weeks following infection mice were irradiated with UV-B to reactivate the latent infection. B6-CD137L KO mice (n=20) were compared with wild-type B6 mice (n=20) for corneal opacity (a) corneal neovascularization (b) and blepharitis (c) along with UV irradiated uninfected controls (n=5 for both strains). P values for the comparison of the two strains ranged from 0.05 to 0.001 by Rank Sum analysis.

Fig. 4.

Latently infected B6 and B6-CD137L KO mice were swabbed for 7 days following UV-B reactivation. Results indicate that there were no differences between B6 (n=20) and B6-CD137L (n=20) mice. These results were compared for statistical significance by Student’s t-test.

To further investigate the CD137 costimulatory pathway, we took advantage of factors that either impair this pathway or promote this pathway. It had been previously shown that an antagonistic anti-CD137 acted to block the ability of CD137 to engage CD137L and thus prevent costimulation. This antagonistic antibody had been used to block CD137 costimulation of MCMV-specific CD8⁺ T cells [40]. When mice were treated with this antagonistic anti-CD137 (clone TKS-1), the disease was significantly increased when compared to control antibody treated mice following UV-B reactivation at most time points measured (Fig. 5). In counterpoint to these studies we used the agonistic antibody (clone 3H3) to determine if promotion of the CD137 costimulatory pathway led to reduced corneal disease [41]. When mice were treated with this agonistic antibody, corneal disease was significantly inhibited compared to control antibody treated animals again at most time points measured (Fig. 6). These data confirm our earlier results indicating that the CD137 pathway helped to better control recurrent HSK following UV-B induced reactivation.

Fig. 5.

B6 mice were latently infected and then treated with a CD137 antagonistic antibody at the time of UV-B reactivation. B6 mice were treated with either TKS-1 (n=18) or with an isotype matched antibody (n=17) at the time of UV-B reactivation and clinical disease monitored for 5 weeks. Results indicated that mice treated with TKS-1 displayed significantly greater corneal disease at days 21 and 28. Data represents mean±S.E.M. for these groups and significance was determined by rank sum analysis, asterisk represents P<0.05–0.01 by Rank Sum analysis.

Fig. 6.

B6 mice were latently infected and then treated with a CD137 agonistic antibody at the time of UV-B reactivation. B6 mice were treated with either 3H3 (n=15) or with an isotype matched antibody (n=15) at the time of UV-B reactivation and clinical disease monitored for 5 weeks. Results indicated that mice treated with H3H displayed significantly reduced corneal disease from Day 21 onward. Data represents mean±S.E.M. for these groups and significance was determined by rank sum analysis, asterisk represents P<0.05–0.01 by Rank Sum analysis.

In order to better characterize the cellular response that occurs in wild-type versus CD137L KO mice, we isolated cells from the corneas of reactivated mice 17 days following UV-B induced reactivation. As seen in Fig. 7, mice lacking the CD137 costimulatory pathway had significantly greater numbers of CD45⁺ cells infiltrating their corneas (P<0.02). When the CD45⁺ cells were further analysed, we saw significantly more cells bearing markers of neutrophils (GR-1, P<0.02) and CD4⁺ T cells (P<0.05) in the CD137L KO mice than in wild-type B6 mice (Fig. 7). These data further support the hypothesis that the CD137 pathway is involved in restricting inflammatory responses to the cornea.

Fig. 7.

Flow cytometric analysis of corneas from CD137L KO mice displayed great numbers of inflammatory cells than corneas from B6 mice. We reactivated B6 (n=11) and B6-CD137L KO mice (n=12) and harvested their corneas on Day 17 following UV-B irradiation. Cells were subjected to flow cytometric analysis for the indicated markers. Results indicate that B6-CD137L KO mice had a greater total inflammatory infiltrate (CD45⁺ cells) and a corresponding increase in neutrophils (GR-1⁺) and to a lesser extent CD4⁺ T cells. Data represents mean±S.E.M. for individually analysed corneas from these groups. Results were analysed by Student’s t-test and significance indicated by asterisk (P<0.05–0.02).

Next, we performed similar studies evaluating the CD45⁺ infiltrate in trigeminal ganglia of mice at different times following primary infection. We had hypothesized that there might be significantly greater numbers of CD45⁺ cells in the trigeminal ganglia of CD137L KO mice. However, we were surprised to find that the number of CD45⁺ cells in the trigeminal ganglia of CD137L KO mice were fewer than 50 % of that found in wild-type mice, and in particular CD8⁺ cells were much fewer in trigeminal ganglia from mice with defective CD137 costimulatory pathways at all three time points that were evaluated (Fig. 8). A similar pattern was also noted when these cells were monitored for gB tetramer expression (data not shown). When these CD8⁺ cells were further monitored for their percentage of activation markers, they were also seen to express a lower percentage of TNFα, IFNɣ, and CD107a in CD137L KO mice than in wild-type B6 mice for all markers at all time points measured (Fig. 9). We interpreted these results as suggesting that these mice did not mount as strong an immune response in the trigeminal ganglia and thus when virus was reactivated from latency, there would be a greater percentage of latently infected neurons that would be reactivating and thus potentially release more virus that would migrate to the periphery. However, data from previous studies had already demonstrated that CD137L KO mice did not display viral litres in corneal tear film that were different from that seen in wild-type B6 mice (Fig. 4).

Fig. 8.

B6-CD137L KO mice display significantly fewer CD8⁺ T cells, fewer gB tetramer⁺ cells (data not shown) with less functional capacity in latently infected trigeminal ganglia at days 7 and 14. Trigeminal ganglia were removed from latently infected B6 (n=10) and B6-CD137L KO (n=8) mice at the indicated time points and were disaggregated into single-cell suspensions and stained with antibodies against: CD45, CD8α, gB-specific tetramer, CD107A, TNFα, and IFNɣ. Cells were then analysed by flow cytometry. Data represents mean±S.E.M. for individually analysed trigeminal ganglia from these groups. Results were analysed by Student’s t-test and all values were significantly less from B6-CD137L KO cells (P<0.01–0.005).

Fig. 9.

The CD8+T cells from trigeminal ganglia of infected B6-CD137L KO were compared to that in B6 mice. Results demonstrate that all activation markers (IFNγ, CD107a, and TNFα) were expressed on significantly fewer CD8⁺ T cells from b6-CD137L KO mice than were such markers on B6 mice at all time three time points measured. Trigeminal ganglia were removed from latently infected B6 (n=10) and B6-CD137L KO (n=8) mice at the indicated time points and were disaggregated into single-cell suspensions and stained with antibodies against: CD45, CD8α, CD107A, TNFα, and IFNɣ. Cells were then analysed by flow cytometry. Data represents mean±S.E.M. for individually analysed trigeminal ganglia from these groups. Results were analysed by Student’s t-test and significance was observed for all comparisons (P<0.05 to 0.005).

Next, we decided to evaluate whether increased CD137L expression in the context of infection would alter the course of disease. To test this, we developed a recombinant HSV-1 virus strain that expressed CD137L (See Fig. 1). As seen in Fig. 10, wild-type B6 mice infected with McKrae-CD137L recombinant HSV-1 failed to develop significant corneal disease as indicated by significantly lower corneal opacity, corneal neovascularization and blink response scores following UV-B-induced reactivation when compared to mice infected with wild-type HSV-1 following UV-B induced reactivation. Similar results were also seen when albino B6-TyC mice were infected with these viruses (Fig. 11). Interestingly, infection with this recombinant virus led to similar levels of latency when compared to wild-type virus infection in B6 mice (Fig. 12a). Likewise, the amount of virus shed by recombinant virus in wild-type B6 mice following UV-B reactivation was similar following acute infection to that seen with the wild-type virus (Fig. 12b). These data suggest that primary infection with HSV-1-CD137L virus had similar viral dynamics to that of wild-type virus. However, the disease profile suggests that the immune response generated in these animals infected with HSV-1-CD137L virus was fundamentally different from that seen with infection of wild-type with HSV-1.

Fig. 10.

B6 mice infected with CD137L-McKrae virus displayed significantly reduced recurrent HSK when compared to wild-type McKrae virus. Eyes of mice were infected with 10⁶ p.f.u. of either HSV-1, McKrae strain or CD137L-McKrae. Each group of mice consisted of 12 mice and were compared for corneal opacity (a) corneal neovascularization (b) and corneal blink response (c) along with UV irradiated uninfected controls (n=5). Significant differences were noted for all time points except for Day 7 in panel C. The P values for the comparison of the two viral strains ranged from 0.05 to 0.001 by Rank Sum analysis.

Fig. 11.

B6-TyC mice infected with CD137L-McKrae virus displayed significantly reduced recurrent HSK when compared to wild-type McKrae virus. Eyes of mice were infected with 10⁶ p.f.u. of either HSV-1, McKrae strain or CD137L-McKrae. Each group of mice consisted of ten mice and were compared for corneal opacity (a) corneal neovascularization (b) and corneal blink response (c) along with UV irradiated uninfected controls (n=5). Significant differences were noted for all time points measured. The P values for the comparison of the two viral strains ranged from 0.05 to 0.001 by Rank Sum analysis.

Fig. 12.

B6 mice infected with CD137L expressing HSV-1 strain McKrae virus displayed similar levels of genome loads in trigeminal ganglia and viral shedding following UV-B reactivation compared to wild-type McKrae virus. B6 mice were infected with either wild-type McKrae (n=10) or with McKrae-CD137L (n=20) virus. The trigeminal ganglia were removed and tested for the presence of DNA corresponding to viral thymidine kinase (tk) by real-time PCR analysis. Prior to this, mice were subjected to daily swabbing to detect virus in the tear film and this tear film was evaluated for viral litres following UV-B induced reactivation. Statistical analysis by Student’s t-test did not reveal any differences between genome copy levels nor viral litres from eye swabs using these two viruses.

Discussion

Recurrent HSV-1 leading to Herpetic Stromal Keratitis (HSK) is a leading cause of infectious blindness in the developed world [12, 13]. Following primary corneal infection by HSV-1, latency is established and can be disrupted by various stressors leading to reactivation. In humans, primary disease is typically without clinical symptoms, while most cases demonstrating clinical disease are a result of reactivation of a latent HSV-1 infection [8, 12, 13, 40]. Recurrent infection results in virus leaving the trigeminal ganglia and returning to the cornea where it initiates an immunopathologic condition that results in corneal inflammation, which if severe, will result in chronic damage that might be permanent. Previous studies have demonstrated that recurrent HSK (rHSK) requires intact T cells [18–21] and an inflammatory infiltrate that consists of both neutrophils [16, 42] and macrophages [43–45]. We have reported that a necessary requirement for rHSK is an intact CD28 costimulatory pathway, that results in the development of a strong CD4⁺ T cell response which are required for disease [18–21]. We have shown that mice who do not genetically possess that pathway do not develop corneal disease [22]. However, these mice are susceptible to encephalitis following UV-B reactivation, which can lead to death [22]. Thus, indicating that development of a T cell response during primary infection is a balancing act between that which controls potentially lethal effects of such infection and the development of rHSK following viral reactivation.

The effects of several other costimulatory pathways have previously been evaluated for their role in primary HSK (pHSK). Including CD40 with CD154 [23], OX40 with OX40L [24] and CD137 with CD137L [25]. Our present study revisited the role that CD137 mediated interactions have during both pHSK and during rHSK. It should be noted that previous work evaluating this interaction indicated that during pHSK CD137 mediated responses were associated with an increased disease phenotype when that response was blocked by antagonistic monoclonal antibodies [25].

Our initial decision to revisit the CD137 costimulatory pathway was due to the increased appreciation that this pathway had been shown to be involved in the activation of various cells of the immune system. In particular, the CD137 costimulatory pathway has been demonstrated to be useful in developing protective immune responses in both infectious disease [28–30] and cancer [26, 46, 47]. Mechanistically, these studies have revealed that CD137 interactions lead to increased CD8⁺ T cell responses [27–29] as well as occasionally increased antibody responses [31, 32]. This has been illustrated in a report by Dharmadhikari et al. who reported that CD137L expressing dendritic cells not only polarize CD8⁺ T cells to a Tc1 phenotype but do so in patients infected with cancer-associated viruses [48]. More recently, such costimulation has been shown to activate a subset of γδ⁺ T cells that participate in antiviral activity against influenza virus [49]. Consequently, we wished to evaluate disease pathology and characterize the cellular infiltrate found in mouse with or without an intact CD137 pathway. Our data is consistent with the notion that this pathway is associated with reduced corneal disease as mice deficient in CD137L displayed an increased disease phenotype when compared with wild-type B6 mice during both pHSK and rHSK. The observations in respect to pHSK were in contrast to what had been previously published [25]. We believe that this might be due, at least in part, that our results are based on using B6 mice that are incapable of expressing CD137L, while their observations were in a BALB/c pHSK model infected with RE. They targeted CD137 either with an antagonistic antibody or used a BALB/c mouse that lacked CD137, not CD137L. These differences could be the reason for these disparate results between what we report here and what they demonstrated then. It should be noted that this same group later demonstrated that enhanced CD137 costimulation results in increased IL-13 expression, which in this non-ocular model of infection limited polarized Th1-mediated inflammation [50]. Since HSK is considered primarily a Th1-mediated response [19, 51, 52], the fact that CD137 costimulation limits the development of Th1 responses argues that it is highly possible that such costimulation would result in reduced corneal disease, which is what our data demonstrates. It is interesting to note that patients with genetic defects that lead to the complete loss of CD137 expression results in significantly increased susceptibility to Epstein-Barr virus infection [53]. In these patients this appears due to not only a lack of T cell signalling, but also significantly reduced B cells and the associated hypogammaglobulinemia [53]. Thus, even in humans, this interaction has importance in both anti-viral T and B cell responses.

The T cell response that we noted in trigeminal ganglia in which CD137L KO mice displayed not only significantly fewer CD8⁺ T cells and gB tetramer⁺ cells but were functionally impaired. Since it has been reported that these cells are critical in limiting the ability of virus to enter into productive virus production from latency [54, 55], we thought that there would be significantly more viral genomes in these infected mice and that they would reactivate more robustly than one would expect from trigeminal ganglia of wild-type latently infected mice. Our data, however, did not support this hypothesis. While there were slightly more viral genomes in B6-CD137L KO mice the amount was not statistically significant. In a similar fashion, viral shedding following UV-B induced reactivation did not result in significantly greater litres of virus isolated from tear film swabs. Thus, one question might be, why the KO mice have worse disease? Two things must be kept in mind when interpreting the data that we observed. First, we do not see a direct correlation between the amount of virus found in tear film swabs and clinical disease. We would argue that this could be due the fact that most of the virus released into tear film likely comes from corneal epithelial cells being infected with the virus. However, the clinical disease that we are measuring is HSK and therefore primarily stromal damage. Virus produced in stromal cells is unlikely to be directly released into tear film. Consequently, we hypothesize that it is the presence of active viral infection of the stromal layer of the cornea that gives rise to the disease of rHSK.

In furtherance of our evaluation of CD137-mediated costimulation during herpetic ocular infection, we created an HSV-1 virus of the McKrae strain that expresses CD137L. Previously it had been reported that a lentivirus expressing both a tumour antigen receptor along with CD137 was successfully used to treat a patient with chronic lymphocytic leukaemia resulting in remission of the patient [56]. This strategy of coupling a T cell antigen receptor along with the CD137 costimulatory molecule that would more effectively engage tumour cells expressing the antigen the T cell receptor is specific for initiated the idea that we could exploit this idea in our model of HSK. Consequently, our thinking was that this would allow any cell infected with this recombinant virus to potentially participate in CD137-mediated costimulation, particularly since these cells would also be presenting viral antigen to potentially HSV-1-responsive T cells. This thinking led to our hypothesis that this recombinant virus would lead to altered corneal disease following reactivation of a latent infection with this recombinant virus. Results indicated that this was exactly the case as latently infected mice did not present with significant rHSK disease. This observation was made in the context of similar rates and levels of virus shed into tear film samples from mice infected with the recombinant virus as to what is measured from reactivated wild-type virus infected mice. Furthermore, the recombinant virus had a similar level of viral genomes in trigeminal ganglia during latency as to that in wild-type virus infected mouse trigeminal ganglia. We are currently exploring potential mechanisms responsible for the lack of disease in mice infected with the recombinant virus. Furthermore, we are also investigating the efficacy of this virus, when used as a vaccine, to significantly reduce rHSK when administered therapeutically into latently infected mice.

We believe that we have presented compelling evidence that CD137 co-stimulated immune responses have an ameliorating effect on HSK. Future studies will more fully characterize the cells (T cells, B cells, dendritic cells, and natural killer cells [57, 58]) as well as factors that are boosted as a result of such costimulation.