in Herpetic Stromal Keratitis T Cells + Virus-Specific and Bystander ...

The Journal of Immunology

Protective and Pathological Roles of Virus-Specific and Bystander CD8ⴙ T Cells in Herpetic Stromal Keratitis1 Kaustuv Banerjee,* Partha Sarathi Biswas,* Udayasankar Kumaraguru,* Stephen P. Schoenberger,† and Barry T. Rouse2* Herpetic stromal keratitis (HSK), resulting from corneal HSV-1 infection, represents a T cell-mediated immunopathologic lesion. In T cell transgenic mice on a SCID or RAG knockout background, the T cells mediating lesions are unreactive to viral Ags. In these bystander models, animals develop ocular lesions but are unable to control infection. Transfer of HSV-immune cells into a CD8ⴙ T cell bystander model resulted in clearance of virus from eyes, animals survived, and lesions developed to greater severity. However, the adoptively transferred CD8ⴙ T cells were not evident in lesions, although they were readily detectable in the lymphoid tissues as well as in the peripheral and CNS. Our results indicate that viral-induced tissue damage can be caused by bystander cells, but these fail to control infection. Immune CD8ⴙ T cells trigger clearance of virus from the eye, but this appears to result by the T cells acting at sites distal to the cornea. A case is made that CD8ⴙ T cell control is expressed in the trigeminal ganglion, serving to curtail a source of virus to the cornea. The Journal of Immunology, 2004, 173: 7575–7583.

H

SV infection of the eye is a painful syndrome that often results in a blinding immunoinflammatory lesion termed herpetic stromal keratitis (HSK)3 (1–3). These lesions are judged to represent T cell-orchestrated immunopathological responses to infection (2–7). Overwhelming support for this mechanism comes from experimental studies in the mouse eye. It is evident that CD4⫹ T cells represent the crucial participants in HSK lesions, but the nature of target Ags recognized by the CD4⫹ T cell remains unclear. Some advocate that autoantigens unveiled in the corneal stroma by HSV infection represent principal targets (8). Others have indicated that T cells reactive with neither viral nor autoantigen may be the principal participants in lesions. This is referred to as the bystander activation model (9 –12). The best evidence for the latter mechanism comes from studies in T cell transgenic mice on a SCID or RAG⫺/⫺ background. Such animals develop HSV upon ocular HSV infection, even though their T cells have no demonstrable reactivity to viral or ocular tissues (9 –12). Curiously, in these models, the transgenic T cells appear able to mediate pathology but are incapable of controlling the HSV infection. In such circumstances, virus persists in the eye and also disseminates to the brain resulting in lethal encephalitis (9 –12). In the present report, we have sought to determine whether CD8⫹ T cells reactive with HSV could also enter the cornea after adoptive transfer and contribute to lesion expression in CD8⫹ TCR transgenic RAG⫺/⫺ mice. Unexpectedly, donor cells could not be unequivocally demonstrated in lesions that in such recipi-

ents progressed in severity. However, the adoptive transfers protected the CD8⫹ TCR recipients from lethal encephalitis and, despite their apparent absence in corneas, caused viral clearance from this tissue. Our results indicate that the bystander cells themselves are responsible for lesion expression but are ineffective at clearing virus. Immune cells, by contrast, do effectively mediate viral control but may cause these effects other than by acting in the site of tissue damage. The likely means by which CD8⫹ T cells function to control virus are discussed.

Materials and Methods Mice Five- to 6-wk-old female mice were used for the experiments. OT-1 RAG⫺/⫺ CD45.1 mice were produced in the laboratory of Stephen Schoenberger (La Jolla Institute of Allergy and Immunology, San Diego, CA). These mice are referred to in the text as OT-1 RAG⫺/⫺ mice. HSV-specific TCR transgenic mice (gBT-I.3- referred to in the text as gBT mice) were produced in the laboratory of Francis Carbone (University of Melbourne, Melbourne, Australia) (13). C57BL/6 (B6) mice were purchased from Harlan Sprague Dawley (Indianapolis, IN). Both B6 and gBT mice were CD45.2⫹. Animals were age and sex matched for all experiments. All manipulations involving immunocompromised mice were performed in a laminar flow hood. To prevent bacterial super infections, mice received prophylactic treatment, starting 1 day before corneal infection, with sulfatrim pediatric suspension (Barre National, Baltimore, MD) at the rate of 5 ml per 200 ml of drinking water. All experimental procedures were in complete agreement with the Association of Research in Vision and Ophthalmology resolution on the use of animals in research.

Virus *Comparative and Experimental Medicine, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996; and †Division of Immune Regulation, La Jolla Institute of Allergy and Immunology, San Diego, CA 92121 Received for publication July 28, 2004. Accepted for publication October 4, 2004. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1

This work was supported by National Institutes of Health Grant EY05093.

2

Address correspondence and reprint requests to Dr. Barry T. Rouse, M409 Walters Life Sciences Building, University of Tennessee, Knoxville, TN 37996-0845. E-mail address: [email protected] 3 Abbreviations used in this paper: HSK, herpetic stromal keratitis; TG, trigeminal ganglion; DLN, draining lymph node; p.i., postinfection.

Copyright © 2004 by The American Association of Immunologists, Inc.

HSV-1 RE (from Robert Hendricks when at the University of Illinois, Chicago, IL) and KOS strains were propagated and titrated on monolayers of Vero cells (CCL81; American Type Culture Collection, Manassas, VA) using standard protocols (14). HSV-1 RE was used for corneal infections, and HSV-1 KOS was used to immunize mice and elicit cutaneous delayedtype hypersensitivity reactions.

Corneal HSV infections and clinical observation Corneal infections of all mice groups were conducted under deep anesthesia induced by avertin (Sigma-Aldrich, St. Louis, MO). Mice were scarified on their corneas with a 27-gauge needle, and a 4-␮l drop containing 5 ⫻ 106 pfu of HSV-1 RE was applied to the eye and gently massaged with the eyelids. The eyes were examined on different days postinfection (p.i.) with 0022-1767/04/$02.00

CD8⫹ T CELLS IN THE PATHOGENESIS OF HSK

7576 a slit lamp biomicroscope (Kowa, Nagoya, Japan), and the clinical severity of keratitis of individually scored mice was recorded. The scoring system was as follows: ⫹1, mild corneal haze; ⫹2, moderate corneal opacity or scarring; ⫹3, severe corneal opacity, but iris invisible; ⫹4, opaque cornea; and ⫹ 5, necrotizing stromal keratitis.

Virus recovery and titration Eye swabs were taken from infected corneas (four eyes/group) using sterile swabs soaked in DMEM containing with 10 IU/ml penicillin and 100 ␮g/ml streptomycin. Swabs were put in sterile tubes containing DMEM and stored at ⫺80°C. For detection of virus, samples were thawed and vortexed. Duplicate 200-␮l aliquots of dilutions of each sample were plated on Vero cells grown to confluence in 24-well plates at 37°C in 5% CO2 for 1 h 30 min. Medium was aspirated, and 500 ␮l of 2⫻ DMEM containing 1% low-melting point agarose was added to each well. Titers were calculated as log10 pfu/ml as per standard protocol (14).

Adoptive transfer of HSV-immune CD8⫹ T cells To generate HSV-immune CD8⫹ T cells, B6 and gBT mice were injected with 5 ⫻ 106 pfu of HSV-1 KOS into the footpad. Single-cell suspensions of pooled spleens and popliteal lymph nodes were prepared from immunized mice 7– 8 days later, and CD8⫹ T cells were purified using a mouse CD8 subset column (R&D Systems, Minneapolis, MN). By flow cytometry analysis, the purified population consisted of 80% CD8⫹ T cells with no detectable CD4⫹ T cells. Approximately 9% (B6) and 50% (gBT) of the CD8⫹ T cells produced IFN-␥ upon restimulation to the gB498 –505 immunodominant epitope of HSV, measured by intracellular IFN-␥ production (measured before purification from cell samples taken from pooled splenocyte and DLN single-cell suspensions; see below for details). Ocularly infected OT-1 RAG⫺/⫺ animals received an i.v. injection of 5 ⫻ 106 purified cells at 24 (referred to in the figures as OT-1 RAG⫺/⫺: CD8 24 h) and 72 h (referred to in the figures as OT-1 RAG⫺/⫺: CD8 72 h; cells from gBT mice are referred to as OT-1 RAG⫺/⫺: gBT CD8 72 h) p.i.

analysis, eyes were enucleated at indicated time points and frozen in OCT compound (Miles, Elkart, IN). Six-micrometer-thick sections were cut, air dried, and fixed in acetone:methanol (1:1) at ⫺20°C for 10 min. For detection of viral Ags, sections were blocked with 5% BSA in PBS containing 1/200 dilution of Fc block (2 h) and incubated with rabbit anti-HSV serum (10 min) (DakoCytomation, Carpenteria, CA) followed by biotinylated goat anti-rabbit Ab (20 min) (Biogenex, San Ramon, CA). Sections were then treated with HRP-conjugated streptavidin for 45 min (1/1000 dilution; Jackson Immunoresearch Laboratories, West Grove, PA) followed by 3,3⬘-diaminobenzidine substrate (Biogenex) and counterstained with hematoxylin (Richard Allen Scientific, Kalamazoo, MI). For immunofluorescence, endogenous biotin was blocked in acetone: methanol-fixed frozen sections with endogenous biotin blocking kit (Molecular Probes, Eugene, OR). This was followed by blocking with 5% BSA-PBS-0.05% Tween 20 containing 1/200 dilution of Fc Block (clone 2.4G2; BD Pharmingen) for 2 h. Abs used for staining were anti-CD8FITC (5 ␮g/ml) (clone 53-6.7; BD Pharmingen), anti-CD4-FITC (clone RM4-5; BD Pharmingen), and biotin-labeled anti-CD45.2 (10 ␮g/ml) (clone 104; BD Pharmingen). An appropriate isotype control (clone G155178; BD Pharmingen) was used for the anti-CD45.2 Ab. Ab dilutions were made in 1% BSA-PBS. After incubating overnight at 4°C, slides were washed thoroughly in PBS, and streptavidin-conjugated Alexa Fluor 546 (1 ␮g/ml) (Molecular Probes) in 1% BSA-PBS was added for 1 h at room temperature. Slides were mounted with Vectashield without propidium iodide (Vector Laboratories, Burlingame, CA). For slides single stained with FITC-labeled Abs, propidium iodide was used as a counterstain (Vectashield with propidium iodide; Vector Laboratories). Images were captured with a Leica SP2 laser scanning confocal microscope (Leica, Deerfield, IL).

Statistical analysis Wherever specified, data obtained were analyzed for statistical significance by a one-tailed standard Student’s t test.

Cellular analysis and flow cytometry Surface staining for the detection of adoptively transferred CD8ⴙ T cells. Single-cell suspensions were prepared from spleen and draining lymph nodes (DLN) of mice. The DLN that were used for cellular analysis were the mandibular and the superficial cervical lymph nodes. Whole brains were minced, and single-cell suspensions were made by passing minced tissue through a 70-␮m nylon cell strainer. Isolated trigeminal ganglions (TGs) were pooled, minced, and digested in 1 mg/ml collagenase/dispase (Roche Diagnostics, Mannheim, Germany) for 1 h 30 min followed by two washes in PBS. Surface staining of cells was conducted in flow cytometry buffer (1⫻ PBS with 3% FCS and 0.1% sodium azide). Viable cells (106) were blocked with Fc Block (clone 2.4 G2; BD Pharmingen, San Diego, CA). Cells were double stained with FITC-labeled antiCD8␣ (clone 53-6.7; BD Pharmingen) and biotin-labeled anti-CD45.1 (clone A20; BD Pharmingen) (host cells) or anti-CD45.2 (clone 104; BD Pharmingen) (donor cells). Biotin-labeled Abs was detected by further incubation of cells with a 1/200 dilution of streptavidin-PerCP (BD Pharmingen). Events were collected on FACScan (BD Biosciences, San Jose, CA) and analyzed using CellQuest version 3.0 (BD Biosciences).

Quantification of IFN-␥ production by intracellular staining. To enumerate IFN-␥-producing CD8⫹ T cells, intracellular cytokine staining was performed as previously described (10). In brief, 106 splenocytes or DLN cells were cultured in flat-bottom 96-well plates. Cells were left untreated, stimulated with HSV gB498–505 peptide (SSIEFARL), OVA257–264 peptide (SIINFEKL) (1 ␮g/106 cells), or treated with PMA (10 ng/ml) and ionomycin (500 ng/ml) and incubated for 6 h at 37°C in 5% CO2. Brefeldin A (10 ␮g/ml) and IL-2 (50 U/ml) was added for the duration of the culture period. After this period, cell surface staining was performed as described above. This was followed by intracellular cytokine staining using a cytofix/ cytoperm kit (BD Pharmingen) in accordance with the manufacturer’s recommendations. PE-labeled anti-IFN-␥ Ab was used for intracellular cytokine staining (clone XMG1.2; BD Pharmingen). Events were collected on FACScan (BD Biosciences) and analyzed using CellQuest version 3.0 (BD Biosciences). Histopathology, immunohistochemistry, and immunofluorescence Eyes were enucleated and fixed in 10% buffered neutral formalin and embedded in paraffin. Sections (6-␮m thick) were cut, deparaffinized, and stained with H&E. For immunohistochemistry and immunofluorescence

Results

HSV-immune CD8⫹ T cells protect from lethal herpetic encephalitis and from HSK when present early after infection As recorded in Fig. 1A (right), ocular infection of OT-1 RAG⫺/⫺ transgenic and control B6 mice with HSV resulted in HSK lesions. However, all of the transgenic mice developed severe encephalitis and had to be sacrificed by day 12 p.i. (Fig. 1A, left). None of the control B6 animals succumbed to infection (Fig. 1A, left). At the time of sacrifice, an HSK incidence of ⬃80% was seen in the infected OT-1 RAG⫺/⫺ mice (Fig. 1B). When visualized with a slit lamp microscope, typical corneal opacification and growth of blood vessels was seen from the limbus (Fig. 1C, left). Histopathologically, the lesions showed inflammatory cell infiltrates and pathology consistent with typical HSK lesions (Fig. 2A). Replicating virus was detectable in eye swabs from OT-1 RAG⫺/⫺ animals at day 8 p.i. (Fig. 3A). At day 12 p.i., ocular lesions from OT-1 RAG⫺/⫺ mice, but not those of B6 animals (not shown), had demonstrable viral Ags present both in the corneal epithelium and stroma (Fig. 3, B and C). In contrast, in control B6 animals, as with other immunocompetent models (15), Ags were detectable in the corneal epithelium until day 5 p.i., and by day 6 p.i. viral was not recoverable from ocular swabs (data not shown). Whereas, in the CD8⫹ transgenic mice, the ocular inflammatory cells were of the CD8⫹ phenotype (Fig. 4A), in the B6 controls almost all of the T cells expressed the CD4⫹ phenotype (Fig. 4B). It was surmised that CD8⫹ T cells from immune B6 animals should be demonstrable in the HSK lesions if such were adoptively transferred into the ocularly infected OT-1 RAG⫺/⫺ mice. To measure this, infected OT-1 RAG⫺/⫺ mice were given HSV-immune B6 CD8⫹ T cell transfers at either 24 or 72 h p.i., and recipients were followed clinically over the ensuing 25 days. Animals were subsequently sacrificed to analyze tissues histochemically for the presence of host and donor cells in HSK lesions. The adoptive

The Journal of Immunology transfers conferred protection from encephalitis, with transfer recipients surviving without any signs of CNS disease for at least 25 days (longest time tested) (Fig. 1A, left). OT-1 RAG⫺/⫺ mice receiving cells at 72 h p.i. developed progressive HSK lesions (Fig. 1A, right). By day 12, 45% of eyes from such animals showed clinically evident HSK lesions, and average disease scores were similar to those in mice that did not receive cells (Fig. 1B). However, in mice that received cells at 24 h p.i., lesion development was slower and average scores were 2-fold lower than in mice that did not receive cells or received them at 72 h p.i. (Fig. 1B). HSK lesions were then monitored in mice that survived encephalitis. In mice that received the cells at 72 h p.i., and showed clinically evident signs of HSK development at day 12 p.i., lesions developed further or persisted throughout the course of the study (Fig. 1, A and B), and at day 25, which was the last time point analyzed, the HSK incidence was 64% (Fig. 1B). The corneas of many of such eyes had severe pathology, with necrosis and scarification and the development of blood vessels reaching the center of the cornea as well as massive infiltration of inflammatory cells (Figs. 1C, right, and 2B). In contrast to these findings, lesion development remained slow, or regressed, in mice that received cells at 24 h (Fig. 1, A and B), and at the end of the study period the incidence was greater than 3-fold lower (Fig. 1B). Eyes from this group showed very mild pathological changes at day 25 p.i. (Fig. 2C). These results show a protective role of immune CD8⫹ T cells in terms of an effect on lesion expression that varied according to the time of adoptive transfer. Chronic corneal lesions contain mainly HSV-unreactive CD8⫹ T cells Sections from OT-1 RAG⫺/⫺ mice given adoptive transfers of HSV-immune B6 CD8⫹ T cells (around 9% of such cells produced IFN-␥ in response to the immunodominant HSV-1 gB498 –505 epitope; see Materials and Methods) were analyzed for the presence of donor T cells at day 25 p.i., relying on the congenic differences in the expression of the CD45 molecule to identify phenotype. Serial ocular sections were stained with anti-CD8 and anti-CD45.2 (donor cells). As revealed by this assay and evident from Fig. 4C, visually, a majority of the CD8⫹ T cells were of the host origin (i.e., non-HSV specific). A few, very rare, donor CD8⫹ cells were found in the central and paracentral cornea (Fig. 4C). At this time, very few of the CD8⫹ T cells in the DLN and the spleen of such mice were of donor origin (Fig. 5). The corneas of mice receiving cells at 24 h p.i. that eventually failed to develop HSK lesions had no demonstrable CD8⫹ T cells (not shown). Additional experiments were done in which HSV-immune CD8⫹ T cells from gBT mice were adoptively transferred 72 h p.i. into OT-1 RAG⫺/⫺ mice at 72 h p.i. Greater than 98% of CD8⫹ T cells in gBT mice have TCRs recognizing the immunodominant HSV-1 gB epitope (13). At the time of adoptive transfer, 50% of these cells were activated and produced IFN-␥ upon stimulation with the gB epitope-peptide (see Materials and Methods). As with the earlier B6 transfers, gBT CD8⫹ cells protected OT-1 RAG⫺/⫺ from lethal encephalitis, and animals developed HSK lesions with an average score of 3 (not shown). Examination of lesions at 12 day p.i (9 days post-adoptive transfer) revealed the presence of host CD8⫹ T cells (Fig. 4D) but donor HSV-reactive CD8⫹ T cells were not demonstrable (Fig. 4D). However, these donor cells were readily demonstrable, in spleens and DLN (total number of recovered CD8⫹ T cells in DLN approximately double that seen in the spleen, data not shown) (Fig. 5). Moreover, in both spleen and DLN, these donor cells were functional as judged by their ability

7577 to produce IFN-␥ in response to the HSV gB epitope stimulation in vitro (Fig. 5). Thus chronic HSK lesions contain mainly HSVunreactive cells. During this time frame, HSV-reactive T lymphocytes are rarely present.

Protection and viral clearance by CD8⫹ T cells in the OT-1 RAG⫺/⫺ mice As mentioned previously, ocularly infected OT-1 RAG⫺/⫺ mice continuously express virus in the cornea. Replicating virus was recoverable from corneal swabs at day 8 p.i. (Fig. 3A) and until death (not shown). In addition, abundant viral Ags were demonstrable by immunohistochemistry, with such Ags being principally in the stroma after the first few days (Fig. 3B). In contrast, adoptive transfer of B6-immune CD8⫹ T cells at either 24 or 72 h p.i. resulted in viral clearance as judged by failure to detect replicating virus in eye swabs from a majority of eyes (in each case three of four eyes tested) at day 8 p.i. (Fig. 3A). Furthermore, viral Ags were undetectable in eyes examined at day 12 p.i. (Fig. 3, D and E). Because, as described above, donor-immune cells could not be demonstrated in lesions, viral clearance must have been mediated other than by direct effect of the CD8⫹ T cells in the site of infection. Additional experiments were performed to help explain how virus-specific CD8⫹ T cells cleared virus from the eye and mediated protection from encephalitis. In such experiments, CD8⫹ T cells were taken from HSV-immune gBT mice (activated and producing IFN-␥, see above), and these were adoptively transferred into ocularly infected OT-1 RAG⫺/⫺ animals 72 h p.i. After adoptive transfer, recipients were killed 9 days later when ocular lesions were apparent. Whereas the gBT cells were undetectable in ocular lesions (see previous section), they were readily detectable by immunohistochemistry (Fig. 4E) and by flow cytometry (Fig. 5) in the TG. Host CD8⫹ T cells were also present in the TG. Curiously, the donor HSV-specific gBT cells outnumbered host CD8⫹ T cells by 3:1 (Fig. 5). In addition, CD8⫹ T cells were also present in the brain as judged by flow cytometry (Fig. 5). In this tissue, the ratio of virus-reactive to nonreactive cells in the brain was almost 6:1 (Fig. 5). The presence of immune CD8⫹ T cells in the peripheral and CNS likely means that they are controlling infection at these sites as documented by others (16).

Discussion Viral infections frequently result in tissue damage that may be especially evident during the recovery phase. Usually T cells are responsible for the inflammation, and it is commonly assumed that the immunopathology is the consequence of viral Ag-specific CD4⫹ or CD8⫹ T cells reacting with virus-infected cells or cells that take up and express viral-encoded epitope peptides (3, 17–22). However, at least in some systems, T cells with no demonstrable reactivity to viral Ags contribute to tissue damage. This mechanism may be a dominant one as in the system used in the present report. In this model, ocular lesions occur in HSV-infected TCR transgenic RAG⫺/⫺ mice whose T cells recognize peptide with no evident cross-reactivity with viral peptides (9 –12, 23). We demonstrate that if viral Ag-reactive CD8⫹ T cells were adoptively transferred to such animals, few if any donor cells were detectable in ocular lesions. However, the donor cells did play a notable role in viral pathogenesis. Accordingly, recipient mice were protected from lethal

7578


FIGURE 1. (Legend continues)


FIGURE 2. Inflammatory cells infiltrate corneas of OT-1 RAG⫺/⫺ mice showing HSK lesions. Photographs of representative corneas of OT-RAG⫺/⫺ mice at day 12 (A), OT-1 RAG⫺/⫺: CD8 72 h at day 25 (B), and OT-1 RAG⫺/⫺: 24 h at day 25 (C). Paraffin-embedded eyes were sectioned, and H&E staining was conducted on 6-␮m sections. Magnification, ⫻200.

infection, lesions progressed to greater severity, and virus was cleared from them even though donor cells were seemingly absent. Viral clearance from the eye may have been affected by antiviral cytokines released by CD8⫹ T cells at extra-ocular sites or perhaps more likely by CD8⫹ T cells in the TG that curtailed a source of infection to the cornea. Control of infection in the brain was presumably mediated by the immune CD8⫹ T cells acting at that site, as has been well-demonstrated to occur by others (16). HSV infection of the cornea only results in HSK in mice that have T cell immunocompetence (5–7, 24). The lesion is an excellent example of a virus-induced immunopathological reaction. In immunocompetent mice, the predominant T cells detectable in the

7579

FIGURE 3. Early clearance of virus from corneas of mice receiving immune CD8⫹ T cells. A, Viral titers were determined on day 8 p.i. by agarose overlay method from four eyes for each group and expressed as log10 pfu/ml. Results expressed as mean virus titer ⫾ SD. ⴱ, p ⬍ 0.05 (OT-1 RAG⫺/⫺ vs either of the two groups receiving immune CD8⫹ T cells); ⴱⴱ, data shows mean virus titer ⫾ SD for four eyes, of which three failed to yield replicating virus. B–E, Immunohistochemistry for the detection of viral Ags (day 12 p.i.) in frozen sections of eyes from OT-1 RAG⫺/⫺ (B and C), OT-1 RAG⫺/⫺: CD8 24 h (D), and OT-1 RAG⫺/⫺: CD8 72 h (E). Arrows show the presence of viral Ag. The substrate was diaminobenzidine, and counterstaining was done with hematoxylin. Original magnification, ⫻200.

ocular lesions express the CD4⫹ phenotype, and these are assumed to be the main orchestrators (1, 4, 5). Establishing the identity of target Ags in the cornea that drive the immunopathology has

FIGURE 1. HSK severity and survival in OT-1 RAG⫺/⫺ mice receiving HSV immune CD8⫹ T cells. Mice were infected on scarified corneas with 5 ⫻ 106 pfu of HSV-1 RE. Adoptive transfer of HSV immune CD8⫹ T cells was conducted via tail vein at 24 h and 72 h p.i. Data compiled from two separate experiments consisting of four animals in each group. A, Survival in mice receiving immune CD8⫹ T cells and kinetics of HSK development. Mice were monitored for development of lesions by a slit lamp ophthalmoscope (see Materials and Methods) at the indicated time points. Results show the mean ⫾ SD for 12 (OT-1 RAG⫺/⫺ and OT-1 RAG⫺/⫺: CD8 24 h), 14 (OT-1 RAG⫺/⫺: CD8 72 h), and 16 eyes (B6). ⴱ, Significant difference (p ⬍ 0.05) compared with OT-1 RAG⫺/⫺: CD8 24 h eyes. ⴱⴱ, Significant difference (p ⬍ 0.05) compared with B6 eyes. B, Results show HSK lesion scores for day 12 and day 25 p.i. OT-1 RAG⫺/⫺ mice received an adoptive transfer of 5 ⫻ 106 immune CD8⫹ T cells at 24 h (E) and 72 h (F) p.i. Control OT-1 RAG⫺/⫺ mice (Œ) succumb to viral encephalitis on day 12 p.i. Shown for comparison is disease development in C57BL/6 mice (⽧). Each dot represents the HSK score from one eye. The horizontal bars and figures in parentheses show the mean for the groups. C, Eyes of OT-1 RAG⫺/⫺ mice, not receiving immune CD8s and succumbing to encephalitis by day 12 p.i., show evidence of opacification and angiogenesis (arrows). Mice receiving immune cells at 72 h p.i. and surviving encephalitis show necrotic lesions of the central cornea along with angiogenesis at day 25 p.i.

7580


FIGURE 4. CD8⫹ T cells in corneas and TG. Frozen sections (6 ␮m) of representative eyes of OT-1 RAG⫺/⫺ (A) and B6 (B) mice at day 12 p.i. were stained with FITC-labeled anti-CD8␣ mAb and antiCD4 mAb, respectively. Propidium iodide was used as a counterstain. Scale bar, 80 ␮m. Frozen sections (6 ␮m) of eyes of OT-1 RAG⫺/⫺: CD8 72 h at day 25 p.i. (C) and of eyes (D) and TG (E) of OT-1 RAG⫺/⫺: gBT CD8 72 h at day 12 p.i. were incubated with FITClabeled anti-CD8␣ (green) and biotin-labeled antiCD45.2. As a negative control for the anti-CD45.2 Ab, sequential sections of TG from OT-1 RAG⫺/⫺: gBT CD8 72 h at day 12 p.i. were incubated with FITClabeled anti-CD8␣ (green) and bition-labeled mouse IgG2a Ab (F). After incubation with streptavidin-Alexa Fluor 546 (red), images were captured using a confocal microscope. The differential interference contrast (DIC) image shows the location of the epithelium (epi) and stroma (str) of the cornea. Scale bars, 80 ␮m.

proven difficult. It is likely they are multiple and not confined to epitope-peptides derived from viral-encoded proteins (23). In fact, as demonstrated in this report, as well as by others (9 –12), HSK can be induced in circumstances where none of the T cells in lesions express demonstrable reactivity to viral Ags. This so-called bystander model has been shown in several different TCR transgenic SCID or RAG⫺/⫺ mice, all of which lack demonstrable cross-reactivity to HSV or respond immunologically to HSV upon immunization (9 –12). The model used in current studies had CD8⫹ T cells reactive with an OVA peptide. These mice developed corneal lesions containing transgenic CD8⫹ T cells after ocular HSV infection, but they were unable to control the infection. In consequence, viral Ag remained present in lesions and animals succumbed to lethal encephalitis within 2 wk of infection. Conceivably, viral Ag, which likely represents replicating virus, drives proinflammatory mediator production that in turn attracts T cells to the eye and activates them to generate tissue-damaging products. If viral replication-induced events occur favorably within the first few days after infection, the stage is set for the successful development of HSK lesions (9). The kinetics of virus clearance after HSV-immune CD8⫹ T cell transfer has been measured previously and occurs within 3 days after transfer (25). Thus early viral clearance, as would have occurred in mice receiving the cells

at 24 h p.i., diminished lesion expression, presumably because there is an inadequate source of proinflammatory mediators to attract and activate the incoming T cells. However, in mice that received cells at 72 p.i. this proinflammatory milieu was induced optimally because virus was allowed to replicate for greater than 3 days. Put together, it appears that for optimal HSK induction in our bystander model viral replication is required for at least 6 – 8 days p.i. Thus the latter and not the former mice developed HSK lesions. Such ideas require further evaluation, particularly the identification of the mediators involved in the bystander activation process. Although CD8⫹ T cells are not prominent components of HSK in immunocompetent mice (1), such cells were the only cell type evident in OT-1 RAG⫺/⫺ corneal lesions. It was expected that in mice receiving HSV-immune CD8⫹ T cells at 72 h p.i. and developing HSK donor cells would gain access to lesions and perhaps add to their severity. In fact demonstrating donor T cells in lesions was extremely difficult both with B6 CD8⫹ T cells and CD8⫹ T cells from gBT mice, a TCR transgenic mouse strain that recognizes an immunodominant epitope peptide of HSV (13). Nevertheless, the adoptively transferred cells were readily detectable in sites other than the eye such as the lymphoid tissue, brain, and most interestingly the inflamed TG. These observations raise


7581

FIGURE 5. Donor CD8⫹ T cells in spleens, DLN, and the nervous system. Single-cell suspensions of spleen, DLN, TG, and brain of OT-1 RAG⫺/⫺ mice were analyzed for the presence of donor CD8⫹ T cells at day 25 and day 12 p.i. by flow cytometry. The ability of the detectable donor CD8s to produce IFN-␥ in response to the HSV immunodominant CD8 epitope gB498 –505 was measured by intracellular analysis (see Materials and Methods). All CD8⫹ T cells in a naive OT-1 RAG⫺/⫺ mouse were CD45.1⫹, and none were CD45.2⫹ (first row). Figures in the upper right quadrant of each plot represent the percentage of CD8⫹ ⌻ cells staining with the respective CD45 marker or intracellular IFN-␥ (for the latter gated on CD8⫹CD45.2⫹ cells). Plots for spleen, DLN, and brain represent data from one of three mice analyzed. Plots for TG represent data from a pool of three TGs.

7582 two so far unsolved issues. First, how can the failure of HSVspecific CD8⫹ T cells to invade HSK lesions, but not lesions in the peripheral nervous system, be explained? It may relate to the differential requirement for homing molecules on cells that access the eye vs other sites. This issue is under investigation. The second issue is to explain how the adoptively transferred immune CD8⫹ T cells succeed in triggering virus Ag clearance from the eye because the donor cells themselves appear absent in the tissue site. This could mean that viral control is largely mediated by soluble factors generated by immune CD8⫹ T cells at extraocular site such as the DLN. We were in fact able to show that at least in gBT recipients lymph node cells stimulated with the cognate viral peptide induced IFN-␥, a cytokine known to inhibit HSV replication (26 –29). This issue of viral clearance by cytokines is being further studied using anti-sera to neutralize candidate molecules involved as well as adoptive transfers of donor CD8⫹ T cells from cytokine knockout animals. Alternatively, the CD8⫹ T control of virus in lesions could result from their activity in the peripheral nervous system. Thus characteristically following HSV ocular infection, virus rapidly gains access to nerve endings and passes by retrograde transport to the TG (30, 31). Here it may replicate in some cells and induce a florid inflammatory response that, after a time, is dominated by CD8⫹ T cells (32–35). Virus that has replicated in the TG returns to the cornea and the skin surrounding it through innervating nerves by anterograde transport (36). Elegant work by the Hendricks group has shown that the CD8⫹ T cells are involved in antiviral control perhaps by a mechanism that does not cause infected cell destruction (33, 35, 37). Because in our adoptive transfer studies we could readily detect immune donor gBT cells in the TG, it is tempting to speculate that these cells functioned in the TG to curtail replication and stop virus transporting to the corneal stroma. This hypothesis requires further evaluation perhaps using virus strains that are unable to reactivate in the TG and transport to tissues such as the eye. In conclusion, our results, and those from previous studies (9 –12, 25), indicate that virus-induced immunopathology may be the result of a bystander mechanism capable of tissue damage but not control of the inciting infection. The latter requires specific T cells, although these may act without the need to enter the tissue site. However, whether this mechanism is operative in the immunologically normal mouse or in humans is presently unclear and needs further evaluation. Both virus-reactive CD4⫹ and CD8⫹ T cells have been expanded from human corneas (38 – 40) and virus-reactive CD4⫹ T cells have been detected in infected mouse corneas (8, 9). Such cells could be capable of both immunopathology and anti-viral effects. Recent results with the immunocompetent mouse model does indicate, however, that though both virus nonspecific and specific CD4⫹ T cells infiltrate the cornea, the former albeit dominates the latter during the clinical phase (9), implicating the involvement of bystander cells in immunopathology. It will of interest to evaluate whether a similar mechanism also operates in other viral-induced immunopathologies.

Acknowledgments We express our thanks to John Dunlap (Microscopy Facility, University of Tennessee, Knoxville, TN) for confocal microscopy and Nancy M. Sawtell (Cincinnati Childrens Hospital Medical Center, Cincinnati, OH) for assistance with the isolation of TG. The help of Amy Cupples and Ericka Blackwell is gratefully acknowledged.


References 1. Streilein, J. W., M. R. Dana, and B. R. Ksander. 1997. Immunity causing blindness: Five different paths to herpes stromal keratitis. Immunol. Today 18:443. 2. Hendricks, R. L. 1999. Immunopathogenesis of viral ocular infections. Chem. Immunol. 73:120. 3. Deshpande, S., K. Banerjee, P. S. Biswas, and B. T. Rouse. 2004. Herpetic eye disease: Immunopathogenesis and therapeutic measures. Expert Rev. Mol. Med. 2004:1. 4. Doymaz, M. Z., and B. T. Rouse. 1992. Herpetic stromal keratitis: An immunopathologic disease mediated by CD4⫹ T lymphocytes. Invest. Ophthalmol. Vis. Sci. 33:2165. 5. Mercadal, C. M., D. M. Bouley, D. DeStephano, and B. T. Rouse. 1993. Herpetic stromal keratitis in the reconstituted scid mouse model. J. Virol. 67:3404. 6. Metcalf, J. F., D. S. Hamilton, and R. W. Reichert. 1979. Herpetic keratitis in athymic (nude) mice. Infect. Immun. 26:1164. 7. Metcalf, J. F., and H. E. Kaufman. 1976. Herpetic stromal keratitis-evidence for cell-mediated immunopathogenesis. Am. J. Ophthalmol. 82:827. 8. Zhao, Z. S., F. Granucci, L. Yeh, P. A. Schaffer, and H. Cantor. 1998. Molecular mimicry by herpes simplex virus-type 1: Autoimmune disease after viral infection. Science 279:1344. 9. Deshpande, S., M. Zheng, S. Lee, K. Banerjee, S. Gangappa, U. Kumaraguru, and B. T. Rouse. 2001. Bystander activation involving T lymphocytes in herpetic stromal keratitis. J. Immunol. 167:2902. 10. Banerjee, K., S. Deshpande, M. Zheng, U. Kumaraguru, S. P. Schoenberger, and B. T. Rouse. 2002. Herpetic stromal keratitis in the absence of viral antigen recognition. Cell Immunol. 219:108. 11. Gangappa, S., S. P. Deshpande, and B. T. Rouse. 2000. Bystander activation of CD4⫹ T cells accounts for herpetic ocular lesions. Invest. Ophthalmol. Vis. Sci. 41:453. 12. Gangappa, S., J. S. Babu, J. Thomas, M. Daheshia, and B. T. Rouse. 1998. Virus-induced immunoinflammatory lesions in the absence of viral antigen recognition. J. Immunol. 161:4289. 13. Mueller, S. N., W. Heath, J. D. McLain, F. R. Carbone, and C. M. Jones. 2002. Characterization of two TCR transgenic mouse lines specific for herpes simplex virus. Immunol. Cell Biol. 80:156. 14. Spear, P. G., and B. Roizman. 1972. Proteins specified by herpes simplex virus. V. Purification and structural proteins of the herpesvirion. J. Virol. 9:143. 15. Babu, J. S., J. Thomas, S. Kanangat, L. A. Morrison, D. M. Knipe, and B. T. Rouse. 1996. Viral replication is required for induction of ocular immunopathology by herpes simplex virus. J. Virol. 70:101. 16. Anglen, C. S., M. E. Truckenmiller, T. D. Schell, and R. H. Bonneau. 2003. The dual role of CD8⫹ T lymphocytes in the development of stress-induced herpes simplex encephalitis. J. Neuroimmunol. 140:13. 17. Rouse, B. T. 1996. Virus-induced immunopathology. Adv. Virus Res. 47:353. 18. Selin, L. K., and R. M. Welsh. 2004. Plasticity of T cell memory responses to viruses. Immunity 20:5. 19. Sissons, J. G., A. J. Carmichael, N. McKinney, J. H. Sinclair, and M. R. Wills. 2002. Human cytomegalovirus and immunopathology. Springer Semin. Immunopathol. 24:169. 20. Varga, S. M., and T. J. Braciale. 2002. RSV-induced immunopathology: Dynamic interplay between the virus and host immune response. Virology 295:203. 21. Welsh, R. M., and L. K. Selin. 2002. No one is naive: The significance of heterologous T-cell immunity. Nat. Rev. Immunol. 2:417. 22. Whitton, J. L. 2002. Immunopathology during coxsackievirus infection. Springer Semin. Immunopathol. 24:201. 23. Deshpande, S. P., S. Lee, M. Zheng, B. Song, D. Knipe, J. A. Kapp, and B. T. Rouse. 2001. Herpes simplex virus-induced keratitis: Evaluation of the role of molecular mimicry in lesion pathogenesis. J. Virol. 75:3077. 24. Newell, C. K., S. Martin, D. Sendele, C. M. Mercadal, and B. T. Rouse. 1989. Herpes simplex virus-induced stromal keratitis: Role of T-lymphocyte subsets in immunopathology. J. Virol. 63:769. 25. Gangappa, S., S. P. Deshpande, and B. T. Rouse. 1999. Bystander activation of CD4⫹ T cells can represent an exclusive means of immunopathology in a virus infection. Eur. J. Immunol. 29:3674. 26. Cantin, E., B. Tanamachi, and H. Openshaw. 1999. Role for ␥ interferon in control of herpes simplex virus type 1 reactivation. J. Virol. 73:3418. 27. Mikloska, Z., and A. L. Cunningham. 2001. ␣ and ␥ interferons inhibit herpes simplex virus type 1 infection and spread in epidermal cells after axonal transmission. J. Virol. 75:11821. 28. Vollstedt, S., S. Arnold, C. Schwerdel, M. Franchini, G. Alber, J. P. Di Santo, M. Ackermann, and M. Suter. 2004. Interplay between ␣/␤ and ␥ interferons with B, T, and natural killer cells in the defense against herpes simplex virus type 1. J. Virol. 78:3846. 29. Vollstedt, S., M. Franchini, G. Alber, M. Ackermann, and M. Suter. 2001. Interleukin-12- and ␥ interferon-dependent innate immunity are essential and sufficient for long-term survival of passively immunized mice infected with herpes simplex virus type 1. J. Virol. 75:9596. 30. LaVail, J. H., W. E. Johnson, and L. C. Spencer. 1993. Immunohistochemical identification of trigeminal ganglion neurons that innervate the mouse cornea: Relevance to intercellular spread of herpes simplex virus. J. Comp. Neurol. 327:133. 31. Shimeld, C., S. Efstathiou, and T. Hill. 2001. Tracking the spread of a lacZ-tagged


32.

33.

34.

35.

36.

herpes simplex virus type 1 between the eye and the nervous system of the mouse: Comparison of primary and recurrent infection. J. Virol. 75:5252. Liu, T., Q. Tang, and R. L. Hendricks. 1996. Inflammatory infiltration of the trigeminal ganglion after herpes simplex virus type 1 corneal infection. J. Virol. 70:264. Liu, T., K. M. Khanna, X. Chen, D. J. Fink, and R. L. Hendricks. 2000. CD8⫹ T cells can block herpes simplex virus type 1 (HSV-1) reactivation from latency in sensory neurons. J. Exp. Med. 191:1459. Shimeld, C., J. L. Whiteland, S. M. Nicholls, E. Grinfeld, D. L. Easty, H. Gao, and T. J. Hill. 1995. Immune cell infiltration and persistence in the mouse trigeminal ganglion after infection of the cornea with herpes simplex virus type 1. J. Neuroimmunol. 61:7. Khanna, K. M., R. H. Bonneau, P. R. Kinchington, and R. L. Hendricks. 2003. Herpes simplex virus-specific memory CD8⫹ T cells are selectively activated and retained in latently infected sensory ganglia. Immunity 18:593. Summers, B. C., T. P. Margolis, and D. A. Leib. 2001. Herpes simplex virus type

7583

37.

38.

39.

40.

1 corneal infection results in periocular disease by zosteriform spread. J. Virol. 75:5069. Liu, T., K. M. Khanna, B. N. Carriere, and R. L. Hendricks. 2001. Gamma interferon can prevent herpes simplex virus type 1 reactivation from latency in sensory neurons. J. Virol. 75:11178. Koelle, D. M., S. N. Reymond, H. Chen, W. W. Kwok, C. McClurkan, T. Gyaltsong, E. W. Petersdorf, W. Rotkis, A. R. Talley, and D. A. Harrison. 2000. Tegument-specific, virus-reactive CD4 T cells localize to the cornea in herpes simplex virus interstitial keratitis in humans. J. Virol. 74:10930. Verjans, G. M., L. Remeijer, C. M. Mooy, and A. D. Osterhaus. 2000. Herpes simplex virus-specific T cells infiltrate the cornea of patients with herpetic stromal keratitis: No evidence for autoreactive T cells. Invest. Ophthalmol. Vis. Sci. 41:2607. Verjans, G. M., L. Remeijer, R. S. van Binnendijk, J. G. Cornelissen, H. J. Volker-Dieben, S. G. Baarsma, and A. D. Osterhaus. 1998. Identification and characterization of herpes simplex virus-specific CD4⫹ T cells in corneas of herpetic stromal keratitis patients. J. Infect. Dis. 177:484.