T Cells in the Uninjected Eye after Anterior Chamber Inoculation of Herpes Simplex Virus Type 1

Transcription

1 T Cells in the Uninjected Eye after Anterior Chamber Inoculation of Herpes Simplex Virus Type 1 Atsushi Azumi and Sally S. Atherton PURPOSE. TO investigate T cell infiltration in the posterior segment of the uninjected eye after uniocular anterior chamber inoculation of HSV-1. METHODS. The anterior chamber of one eye of euthymic BALB/c mice was injected with 1 X 10 4 plaque-forming units (PFU) to 2 X 10 4 PFU of herpes simplex virus type 1 (HSV-1; KOS strain). All mice were examined for retinitis on day 8 postinoculation (p.i.). Only mice with retinitis were retained and used in these experiments. Animals were killed on days 9, 11, 14, 21, 35, and 63 pi. The uninjected eyes were removed. Some of the uninjected eyes were sectioned and stained for CD4 + and CD8 + cells using the avidin-biotinylated enzyme complex method. Infiltrating cells were collected from the remaining uninoculated eyes and stained using rat anti-mouse monoclonal antibodies specific for CD4 + or CD8 + T cells, and the percentage of CD4 + and CD8 + T cells was determined by flow cytometry. RESULTS. At day 9 pi. (acute retinitis), T cells were observed in the uvea but not in the retina of the contralateral eye. CD4 + and CD8 + cells were observed in the sensory retina coincident with the onset of retinal necrosis (day 11 p.i.), and CD4 + and CD8 + T cells continued to be detected in the remnants of the retina up to and including day 63 p.i. The maximum percentage of both CD4 + and CD8 + T cells was observed at day 21 p.i. CONCLUSIONS. These results demonstrate that T cells enter the retina of the uninoculated eye during HSV-1 infection. The observation that T cells arrive in the sensory retina at the onset of retinal necrosis and not during acute retinitis and the peak of virus replication provides further evidence that T cells play a role in development of retinal necrosis. The result that T cells are observed in the uninjected eye as late as day 63 p.i. suggests that T cells might also have a role in the resolution phase of the disease. (Invest Ophthalmol Vis Sci. 1998;39:78-83) Inoculation of herpes simplex virus type 1 (HSV-1; KOS strain) into one anterior chamber of euthymic BALB/c mice results in virus infection of the contralateral optic nerve and retina in about 90% of infected mice beginning on day 7 postinoculation (p.i.). 1 " 3 The peak of virus replication in the contralateral eye occurs on day 9 or day 10 p.i., and the titer of infectious virus declines rapidly after this time. 4 The median virus titer in the uninjected eye of euthymic mice at day 14 is less than 100 plaque-forming units (PFU), 4 ' 5 and all virus is cleared from the uninjected eye by day 21. The infectious process in the retina of the uninoculated eye has been divided into three phases: acute retinitis, retinal necrosis, and resolution. 6 Acute retinitis is observed between days 7 and 9 p.i. 6 Because T cells are not observed in the contralateral retina at day 10 p.i. 7 and depletion of CD4 + or CD8 + T cells does not affect the severity of retinitis or the titer of virus in the contralateral eye at day 9 pi, 5 acute retinitis is caused primarily by virus replication in the retina of the uninjected eye. Retinal necrosis begins on day 10 p.i. and appears to require the participation of CD4 + T cells. Results from T cell From the Department of Cellular and Structural Biology, The University of Texas Health Science Center, San Antonio, Texas. Supported by National Institutes of Health grant EY Submitted for publication July 17, 1997; revised September 26, 1997; accepted October 4, Proprietary interest category: N. Reprint requests: Sally S. Atherton, Department of Cellular and Structural Biology, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX depletion studies demonstrated that the severity of retinal infection was significantly reduced and virus clearance was delayed at day 14 p.i. only in CD4-depleted mice. 5 The resolution phase begins on or about day 15 p.i. 6 Early in this phase, all remaining replicating virus is cleared from the uninjected eye. Later in this phase, the remnants of the retina are organized into a fibrocellular scar, and the ocular inflammation is gradually resolved. Based on the results of previous studies, it is hypothesized that T cells are recruited to the infected contralateral eye at or near the time of acute retinal necrosis. To test this hypothesis, immunohistochemistry and flow cytometry were used to determine the pattern of T cell infiltration into the uninoculated contralateral eye as well as the percentage of T cells in the contralateral eye during all phases of the infection. METHODS Mice Adult (S:6 weeks) euthymic female BALB/c mice (Ia d ) were purchased from Taconic (Germantown, NY). Mice were housed in accordance with National Institutes of Health Guidelines, and all animal procedures conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Sodium pentobarbital (0.65 mg/10 g body weight) was used as the anesthetic agent for all ocular inoculations and for clinical observations. Investigative Ophthalmology & Visual Science, January 1998, Vol. 39, No. 1 Copyright Association for Research in Vision and Ophthalmology

2 IOVS, January 1998, Vol. 39, No. 1 T Cells in the Uninjected Eye 79 Virus The KOS strain of HSV-1 was used in these experiments. Vims stocks were prepared, aliquoted, and stored at 70 C as described previously. 8 The titer of stock vims was determined by plaque assay on Vero cells and was 2.5 X 10 8 PFU/ml. For each experiment, a fresh aliquot of stock virus was thawed and diluted to the appropriate concentration immediately before ocular inoculation. Immunohistochemistry Mice were killed and perfused with phosphate-buffered saline (PBS). The spleens and contralateral eyes were removed, embedded in OCT compound (Miles, Elkhart, IN), and sectioned at 8 /xm using a cryostat. All sections were dried in air and kept at 4 C overnight before staining at room temperature. Sections were fixed in acetone (5 minutes) and incubated with 3% (vol/vol) normal goat serum (15 minutes). Excess serum was blotted, and the sections were incubated for 45 minutes with biotinylated rat monoclonal antibodies (mabs) specific for mouse CD4 + T cells or CD8 + T cells (Life Technologies, Gaithersburg, MD). All mabs were diluted with 3% (w/v) bovine serum albumin in PBS (bovine serum albumin solution) to a concentration of 2.5 to 50 jag/ml. The sections were washed twice with PBS and placed in 0.5% (vol/vol) hydrogen peroxide in methanol for 20 minutes. The sections were washed once with PBS and incubated with avidin- biotinylated ABC enzyme complex reagent (Vector Laboratories, Burlingame, CA) for 30 minutes. The sections were then washed three times with PBS and 0.5 mg/ml 3,3'-diaminobenzidine tetrahydrochloride (DAB) solution (Sigma Chemical, St. Louis, MO) supplemented with 0.3% (wt/vol) nickel chloride and 0.01% (vol/vol) hydrogen peroxide was added. Color development was monitored microscopically. All sections were covinterstained with hematoxylin and examined microscopically. Cell Collection and Flow Cytometry Mice were killed and perfused with PBS, and the contralateral eyes and contralateral superficial cervical lymph nodes were collected. Lymph node cells were collected by placing the lymph nodes on loo-ptm nylon mesh and teasing the tissue gently with forceps. The cell suspension was washed twice with PBS and resuspended in PBS. For ocular cell collection, the conjunctiva and extraocular muscles were removed from the enucleated eye. To minimize cell loss, the eye was positioned on a 1-cm 2 nylon mesh square placed over the opening of a 1.5-ml microcentrifuge tube filled with complete Dulbecco's modified Eagle's medium with 5% serum. The lens, iris, and ciliary body were removed. The retina and the choroid were dissected from the posterior segment, and the posterior uveal tract and the sensory retina were carefully teased on the nylon mesh until onlyfibroustissue remained. The cell suspension was pelleted and resuspended in PBS. After resuspension, aliquots of each sample were transferred to fresh microcentrifuge tubes. The cells were repelleted, the supernatant was aspirated, and the cells were resuspended in the residual PBS and incubated for 30 minutes on ice with 5 jul of fluorescein isothiocyanate-conjugated rat anti-mouse CD4 mab (Life Technologies), 5 /xl of fluorescein isothiocyanate-conjugated rat anti-mouse CD8 mab (Becton-Dickinson, Mountain View, CA), or 5 jixl of bovine serum albumin solution. The monoclonal antibodies were diluted with bovine serum albumin solution; the concentration of the anti-cd4 mab was 0.5 mg/ml, and the <D LJA Log Relative Fluorescence Lymph node Normal eye Inflamed eye Inflamed eye CD4 depleted Inflamed eye CD8 depleted FIGURE 1. Flow cytometry results demonstrating gate selection to detect CD4 + and CD8 + T cells isolated from eyes with HSV-1 retinitis. Cells were isolated from lymph nodes and eyes as described under Methods and analyzed by flow cytometry. Cells from a normal lymph node stained for CD4 (A) and for CD8 (B), cells from a normal eye of an uninfected mouse stained for CD4 (C) and for CD8 (D), cells from an uninoculated eye with HSV-1 retinitis at day 11 p.i. stained for CD4 (E) and CD8 (F), cells from the uninoculated eye of a CD-4-depleted mouse with retinitis stained for CD4 (G) and CD8 (H), and cells from the uninoculated eye of a CD8-depleted mouse with retinitis stained for CD4 (I) and CD8 (J). The dotted line indicates unstained cells, and the solid line indicates stained cells. concentration of the anti-cd8 mab was 0.2 mg/ml. Cells were washed once with PBS containing 50 /xg/ml propidium iodide (PI; Sigma Chemical) resuspended in 0.25 ml to 0.5 ml of PBS with 5% (vol/vol) fetal calf serum, and passed through nylon mesh before flow cytometry. 9 Flow cytometry was done on a FACStar plus (Becton Dickinson, Mountain View, CA) using a single excitation wavelength of 488 nm. A gate to select mononuclear cells and to exclude dead cells and debris was set at the time of analysis using a single cell suspension stained with PI alone. To verify specificity of gating, cells from a normal lymph node, from a normal eye, and from an uninoculated eye with HSV-1 retinitis at day 11 p.i. were isolated, stained for CD4 + and CD8 + T cells, and analyzed (Figs. 1A through IF). As an additional control, at day 11 p.i., cells from the uninoculated eye of a CD4-depleted mouse with HSV-1 retinitis and from a CD8-depIeted mouse with HSV-1 retinitis were isolated and stained for CD4 + and CD8 + T cells (Figs. 1G through 1J). For the studies on T cell depletion, mice were depleted of CD4 + or CD8 + T cells using a previously described method. 5 To determine the frequency

3 80 IOVS, January 1998, Vol. 39, No. 1 Azumi and Atherton For these experiments, adjacent serial sections from mice with clinical evidence of retinitis at day 8 p.i. were stained with anti-cd4 or anti-cd8 mab from day 9 p.i. through day 63 p.i. At day 9 pi., CD4 + and CD8 + T cells were observed in the iris and ciliary body (not shown) and in the choroid (Figs. 3A, 3B). T cells are not seen in the sensory retina even though the retina is infected with virus and became disorganized at that time.'' At day 11 p.i., T cells were observed in the sensory retina, and the number of T cells in the choroid appeared to be increasing (Figs. 3C, 3D). By day 14 p.i., fewer T cells were observed in the choroid, whereas the number of CD4 + and CD8 + T cells in the retina still appeared to be increasing (Figs. 3E, 3F). By day 21 p.i., the border between the sensory retina and the choroid became indistinct, and the retina began to be replaced by fibrous tissue (Figs. 4A, and 4B). At that time, the maximum number of T cells was observed in the posterior segment. By day 35 p.i., fibrous tissue replacement of the retina continued, and the number of CD4 + and CD8 + T cells was decreased (Figs. 4C, 4D). Also, at that time, the uninjected eye was becoming phthisical (not shown). By day 63 pi-, the retina was completely replaced by fibrous tissue that contained both CD4"1" and CD8 + T cells within its interstices (Figs. 4E, 4F), Flow Cytometry FIGURE 2. Imniunohistochemistry results illustrating CD4 + and CD8 + cells, respectively, in the spleen of a normal uninfected mouse (A, B), tack of reaction of BALB/c spleen cells with an anti-iak anti-body (C), and absence of both CD4 + and CD8 + T cells in the uvea and retina of an eye from an uninfected mouse (D, E). Original magnification: 33X. of CD4 + or CD8 + T cells, single color analysis was performed, and results from each eye were depicted on a single-parameter fluorescence histogram. Experimental Plan On day 0, all mice were inoculated in one anterior chamber with 1 X 104 PFU to 2 X 104 PFU of HSV-1 (KOS).' Because a small fraction (<10%) of mice inoculated with this dose of virus do not develop retinitis,1"3 on day 8 p.i., all mice were anesthetized and the contralateral eyes were examined for retinitis by indirect ophthalmoscopy. 6 Only mice with retinitis on day 8 p.i. were retained and killed on days 9, 11, 14, 21, 35, and 63 pi- for flow cytometry (four or five mice per time point) or for immunohistochemistry (one or two mice per time point). The experiment was repeated, and the results of both studies were combined. As a result of the extent of retinal destruction and difficulty in selecting representative sections that were equivalent from eye to eye at all time points, it was not possible to quantify the number of T cells in each eye using histopathologic methods. Therefore, flow cytometry was used to confirm the histopathologic observations and to determine the percentage of CD4 + and CD8 + T cells in the uninoculated eye during all phases of the infectious process. Similar to the histopathologic observations, flow cytometry revealed an increasing number of CD4+ and CD8 + T cells from day 9 p.i. through day 21 p.i. with a decreasing number of CD4 + and CD8 + T cells on day 35 p i. and day 63 p i. (Fig. 5). The average percentages of CD4 + T cells were 5.6%, 16.4%, 23-0%, 36.1%, 18.2%, and 5.2% and of CD8 + T cells were 6.6%, 14.0%, 27.7%, 33-9%, 31.7%, and 12.0% at days 9, 11, 14, 21, 35, and 63 p.i., respectively. The maximum percentage of both CD4 + and CD8 + T cells was observed at day 21 p.i. The percentage of CD4 + and CD8 + T cells at day 21 p.i. in the uninjected eye of a mouse that did not develop retinitis was 0.7% and 0.5%, respectively (not shown). These results provide flow cytometric confirmation of the histopathologic findings shown in Figures 2D and 2E. In contrast to the flow cytometry results in the uninoculated eye, the percentage of CD4 + and CD84" T cells in the draining lymph nodes from HSV-1-infected mice remained relatively constant during the course of the experiment (Fig. 5). DISCUSSION RESULTS Immunohistochemistry Initial studies were done to verify the ability of the anti-cd4 and anti-cd8 mabs to detect T cells in the spleen of a normal uninfected mouse (Figs. 2A, 2B). As shown in Figure 2C, an adjacent section of the spleen did not react with an isotypematched rat anti-mouse lak (Life Technologies). In addition, the retina of a normal eye from an uninfected mouse did not contain either CD4 + or CDS"1" T cells (Figs. 2D; 2E). After uniocular anterior chamber inoculation of the KOS strain of HSV-1, the retina of the uninoculated eye becomes infected with virus, and acute retinitis is observed in the uninoculated eye beginning on day 7 p.i.' 1 ' 6 ' 1 ' Acute retinitis is followed by retinal necrosis, and retinal necrosis is followed by resolution, during which the inflammatory process in the uninoculated eye is resolved and the remnants of the retina become organized into a fibrocellular scar.6 Although an unequivocal role for T cells in protection of the retina of the injected eye has been demonstrated, 212 " 14 the role of T cells in the pathogen-

4 T Cells in the Uninjected Eye IOVS, January 1998, Vol. 39, No. 1 ; * \ FIGURE 3- Retinal cross-sections illustrating recruitment of T cells (dark-stained cells at arrow tips) to the contralateral uvea and retina after anterior chamber inoculation of HSV-1. (a) CD4 + cells, day 9 p.i.; (b) CD8"1" cells, day 9 p.i.; (c) CD4 + cells, day 11 p.i.; (d) CD8 + cells, day 11 p.i.; (e) CD4 + cells, day 14 p.i.; and (f) CD8 + cells, day 14 p.i. The choroid is indicated by an asterisk. Original magnification: 132X. _ S: : ' FIGURE 4. Retinal cross sections illustrating persistence of T cells (dark-stained cells at arrow tips) in the posterior segment of the contralateral eye during the resolution phase after anterior chamber inoculation of HSV-1. (a) CD4 + cells, day 21 p.i.; (b) CD8 + cells, day 21 p.i.; (c) CD4 + cells, day 35 p.i.; (d) CDS"*" cells, day 35 p.i.; (e) CD4 + cells, day 63 p.i.; and (f) CD8 + cells, day 63 pi. The area of the fibrocellular scar is indicated by a bracket in O-F. Original magnification: 132X. 81

5 82 Azumi and Atherton IOVS, January 1998, Vol. 39, No. 1 B CD4 {%) Day11p.i. 0- o CO O ?40- co Q Day63p.i. ' ' * 0- ' I I FIGURE 5. Flow cytometry results after uniocular anterior chamber inoculation of HSV-1 from day 9 pi. through day 63 p.i. (A-F). Each closed circle represents the percentage of CD4 + and CD8 + cells in a single eye with retinitis, and each open circle represents the percentage of CD4 + and CD8 + cells in the draining lymph nodes from a mouse with retinitis. esis of retinal infection and subsequent destruction is still under investigation. Results of a previous study by Zaltas and co-workers, 7 to determine timing and location of inflammatory cells in the uninoculated eye after anterior chamber injection of HSV-1, showed that at day 10 p.i., the infiltrate in the contralateral retina comprises macrophages but no T cells. At this time, T cells were observed only in the choroid and not in the retina of the uninjected eye. The latest time point examined in the studies was day 10 p.i. (acute retinitis). Results from a later study in which mice were depleted of CD4 +, CD8 +, or both CD4 + and CD8 + T cells indicated that neither CD4 + nor CD8 + cells influence the course of acute retinitis or the extent of viral replication. 5 However, these studies (which included samples from mice at day 14 p.i.) suggested that CD4 + T cells are involved in the transition from acute retinitis to retinal necrosis. Together, the results of these two previous studies suggest that examination of later time points is needed. Therefore, the present study used both immunohistochemistry and flow cytometry to examine uninoculated eyes from day 9 p.i. until day 63 p.i. These new studies confirm early (by day 9 p.i.) recruitment of T cells to the choroid of the uninoculated eye and provide additional evidence in support of the idea that T cells are more important during the necrosis and resolution phases of the disease than during early, acute retinitis. 0 In these studies (which used only uninoculated eyes with retinitis), both CD4 + and CD8 + T cells were observed in the uvea of the uninoculated eye as early as day 9 pi., but T cell infiltration into the sensor)' retina was not observed until the onset of acute retinal necrosis. Despite massive virus replication in an uninoculated eye with retinitis, 6 ''' T cell recruitment into the sensory retina of an uninoculated eye with retinitis was delayed until after the peak of virus replication. Because vascular occlusion has been observed in human patients with acute retinal necrosis, 15 " 17 early occlusion of the retinal artery may prevent T cell ingress into the sensory retina in the mouse. Alternatively, even though T cells were not observed in the retina until day 11 p.i., many inflammatory cells and a large amount of inflammation were observed in the uninjected eye coincident with increasing virus replication in this eye. 4 ' 7 Perhaps, virus infection of the retina downregulated T cell adhesion molecules on the endothelium of the retinal vasculature such that entrance of T cells into the contralateral retina was delayed until several days after the onset of virus replication. Not surprisingly, the results of flow cytometry and immunohistochemistry studies of the contralateral eye were complementary. However, even though the results of the previous depletion studies, along with the results from this study, support the idea that CD4 + cells are involved in the pathogenesis of retinal necrosis, the role of CD8 + T cells in one or more phases of the HSV-1 retinal infection has not been denned. Previous studies in CD8-depleted mice demonstrated that depletion of CD8 + T cells does not influence either the course of acute retinitis (evaluated from histopathologic evidence) or virus replication (assessed by plaque assay). 5 However, because the previous studies on T cell depletion were short-term (up to day 14 p.i.) experiments, 5 it is possible that, even though CD8 + T cells are not involved in acute retinitis, virus clearance, or the transition to retinal necrosis, they could be involved in resolution of the ocular inflammation and/or formation of the fibrocellular scar. HSV-1-specific precursor cytotoxic T cells are present in the uninoculated eye at clay 10 p.i. (the beginning of retinal necrosis). 18 Although the identity of these cells was not established, it is possible that these precursor cytotoxic T cells are CD8 + T cells that are activated later in the eye to become functional cytotoxic T lymphocytes and contribute to resolution by clearing cells expressing viral antigens and/or assisting with consolidation of the retinal debris. It might also be postulated that CD8 + T cells are merely present in the contralateral eye because of passive diffusion through a disrupted blood-retina barrier. This explanation appears to be unlikely for two reasons. First, the percentage of CD8 + T cells in the blood of an HSV-1-infected mouse did not vary appreciably over time (data not shown). Second, the percentage of CD8 + T cells in the cellular infiltrate increased until day 21 p.i. and then declined slowly until the end of the experiment (day 63 p.i.). Neither of these scenarios would be expected if the percentage of T cells in the eye merely reflected the composition of mononuclear cells in the circulation. In conclusion, these studies support the hypothesis that T cells are involved in the later stages (necrosis and resolution) of HSV-1 infection of the retina of the uninoculated eye after uniocular anterior chamber inoculation of virus. Studies are now in progress to define the mechanism(s) by which CD4 + T cells and, perhaps, CD8 + T cells contribute to this disease process.

6 IOVS, January 1998, Vol. 39, No. 1 T Cells in the Uninjected Eye 83 Acknowledgments The authors acknowledge the excellent assistance of Charles A. Thomas, III with flow cytometric analysis of ocular and lymph node cells and careful preparation of the manuscript by Lita Chambers. References 1. Whittum JA, McCulley JP, Niederkorn JY, Streilein JW. Ocular disease induced in mice by anterior chamber inoculation of herpes simplex vims. Invest Ophthalmol Vis Sci. 1984;25: Metzger EE, Whittum-Hudson JA. The dichotomy between herpes simplex virus type 1-induced ocular pathology and systemic immunity. Invest Ophthalmol Vis Sci. 1987;28: Zierhut M, Soukasian S, Zhao T-Z, Tamesis RR, Foster CS. Depletion of T-lymphocytes subsets in murine herpes-simplex-virus retinitis. GerJ Ophthalmol. 1992; 1: Atherton SS, Streilein JW. Two waves of virus following anterior chamber inoculation of HSV-1. Invest Ophthalmol Vis Sci. 1987; 28: Azumi A, Cousins SW, Kanter MY, Atherton SS. Modulation of murine HSV-1 retinitis in the uninoculated eye by CD4 + lymphocytes. Invest Ophthalmol Vis Sci. 1994;35: Cousins SW, Gonzalez A, Atherton SS. Herpes simplex retinitis in the mouse: clinicopathologic correlations. Invest Ophthalmol Vis Sci. 1989:30: Zaltas MM, Opremcak EM, Hemady R, Foster CS. Immunohistopathologicfindingsin herpes simplex virus chorioretinitis in the von Szily model. Invest Ophthalmol Vis Sci. 1992;33: Matsubara S, Atherton SS. Spread of HSV-1 to die suprachiasmatic nuclei and retina in T cell depleted mice. / Neuroimmunol. In press. 9. Cousins SW, Streilein JW. Flow cytometric detection of lymphocyte proliferation in eyes with immunogenic inflammation. Invest Ophthalmol Vis Sci. 1990;31: Niederkorn JY, Streilein JW, Shadduck JA. Deviant immune responses to allogeneic tumors injected intracamerally and subcutaneously in mice. Invest Ophthalmol Vis Sci, 1981;20: Vann VR, Atherton SS. Neural spread of herpes simplex virus after anterior chamber inoculation. Invest Ophthalmol Vis Sci. 1991; 32: Whittum-Hudson JA, Farazdaghi M, Prendergast RA. A role for T lymphocytes in preventing experimental herpes simplex virus type 1-induced retinitis. Invest Ophthalmol Vis Sci, 1985;26: Atherton SS, Altman NH, Streilein JW. Histopathologic study of herpes virus-induced retinitis in athymic BALB/c mice: evidence for an immunopathogenic process. Curr Eye Res. 1989;8: Azumi A, Atherton SS. Sparing of the ipsilateral retina after anterior chamber inoculation of HSV-1: requirement for either CD4 + or CD8 + T cells. Invest Ophthalmol Vis Sci. 1994;35: Willerson D Jr, Aaberg TM, Reeser FH. Necrotizing vaso-occlusive retinitis. Am J Ophthalmol, 1977;84: Fisher JP, Lewis ML, Blumenkranz M, et al. The acute retinal necrosis syndrome. Ophthalmology. 1982;89: l Duker JS, Nielsen JC, Eagle RC, Bosley TM, Granadier R, Benson WE. Rapidly progressive acute retinal necrosis secondary to herpes simplex virus, type 1. Ophthalmology. 1990;97:l Streilein JW, Igietseme JU, Atherton SS. Evidence that precursor cytotoxic T cells mediate acute necrosis in HSV-1-infected retinas. Curr Eye Res. 1991;10S:81-86.