T Lymphocytes in the Trigeminal Ganglia of Rabbits During Corneal HSV Infection

Transcription

1 Investigative Ophthalmology & Visual Science, Vol. 29, No. 11, November 1988 Copyright Association for Research in Vision and Ophthalmology T Lymphocytes in the Trigeminal Ganglia of Rabbits During Corneal HSV Infection Bryan M. Gebhardt and James M. Hill The results of this investigation reveal, for the first time, the presence of thymus-derived (T) lymphocytes in the trigeminal ganglia of rabbits undergoing primary corneal infection with herpes simplex virus type 1. Infiltration of T cells into the trigeminal ganglion was evident at 15 days after primary ocular infection but these cells were no longer present by 45 days after infection. Corneas and trigeminal ganglia of rabbits sacrificed at 3, 7, 12, 15, 20, 25, 30, 45, 70 and 90 days after infection were assayed for infectious virus and stained for viral antigen and immunoreactive T cells. Infectious virus and cells expressing viral antigens were present in the corneas and trigeminal ganglia during the acute phase (day 0-day 14) of the infection. T cell infiltration of the trigeminal ganglion was present as a perivascular infiltrate along with a sparse scattering of these cells among the nerve fibers. The perivascular infiltration is characteristic of viral infection of a tissue and was not seen in the sections of trigeminal ganglia obtained earlier than 15 days or in ganglia obtained 45 days or more after primary corneal infection. This investigation demonstrates conclusively that the neural ganglia are not completely shielded from the host immune response, as evidenced by the observation that immunocompetent T lymphocytes infiltrate the ganglia subsequent to the infection of a peripheral tissue such as the cornea of the eye. Invest Ophthalmol Vis Sci 29: ,1988 A consequence of primary corneal infection by herpes simplex virus type 1 (HSV-1) is the establishment of latency by the virus. 1 ' 2 Although still a subject of study, 3 " 5 it is accepted by most investigators that the neural ganglia whose axons innervate the eye are the primary site in which the virus resides in its latent form. 6 " 8 The present study is part of a continuing investigation of the immunological consequences 9 of primary corneal herpetic infection and the biology and pharmacology of neural latency and reactivation. 10 " 14 The HSV-1 genome, 15 viral transcripts 16 and viral RNA 1718 have been identified in the sensory ganglia of infected humans and animals. The viral genome may intercalate into the DNA of only a few cells of the neural ganglia, making it initially difficult to localize and quantitate the extent of viral colonization With the advent of more sensitive and efficient techniques it is now possible to detect picogram amounts of HSV DNA and RNA " 24 Integration From the Lions Eye Research Laboratories, LSU Eye Center, Louisiana State University Medical Center School of Medicine, New Orleans, Louisiana. Supported in part by PHS grants EY-O315O, EY-06311, EY-02672, EY-02389, and EY from the National Eye Institute, National Institutes of Health, Bethesda, Maryland. Submitted for publication: December 14, 1987; accepted June 13, Reprint requests: Bryan M. Gebhardt, PhD, LSU Eye Center, 2020 Gravier Street, Suite B, New Orleans, LA of the viral genome into the DNA of the host protects the virus from the host's resistance mechanisms, including cellular and humoral immunity If 5 however, there is continued production and maturation of even a small number of viral particles by latently infected neural ganglia, it might be expected that the infected host could, during latency, maintain immunity to the virus. Unfortunately, it has been extremely difficult to obtain precise details regarding the time course of establishment of neural latency by HSV-1 and the timing of the generation of immunity to the virus. It is paradoxical that HSV-1 can cause recurrent infection in the face of existing specific immunity to viral antigens unless the neural ganglia are truly sequestered sites from which infectious virus can escape to cause recurrent disease. The purpose of this study was to determine if, during the establishment of neural latency, HSV-1 alerts the immune system to its presence in the ganglion and elicits an intraganglionic cellular immune response. A hallmark of such a response would be the presence of T lymphocytes. The production of viral particles and/or viral glycoproteins by infected neurons might stimulate immunity and attract immune effector cells (T cells) into the ganglia if these viral products reach the ganglionic vasculature or if the viral infection causes ganglionic cells to emit chemotactic signals that attract effector cells. As indicated above, the physical state of HSV-1 during neural latency is uncertain. During the 1683

2 1684 INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / November 1988 Vol. 29 acute phase of a comeal infection infectious virus is recoverable from the trigeminal ganglion 2728 and Green et al 6 ' 29 have reported the presence of viral immediate early polypeptides in the neurons of the trigeminal ganglion both during the acute phase and during latency. We report here the presence of T lymphocytes in the trigeminal ganglia of rabbits that have undergone corneal inoculation with HSV-1 and document that these cells are first present when the corneal infection is resolving and latency is still being established, but are no longer present when latency has been achieved and viral shedding has stopped. Animals Materials and Methods New Zealand White (NZW) rabbits of both sexes weighing 3-5 pounds each were used in these studies. Animal care and treatment in this investigation were in compliance with the ARVO Resolution on the Use of Animals in Research. Experimental Corneal Infection One hour prior to infection the rabbits were given 1 ml of 25 mg/ml chlorpromazine (Smith Kline Corp., Philadelphia, PA) intramuscularly. Immediately prior to infection the rabbit eyes were treated topically with two drops of proparacaine hydrochloride solution (E.R. Squibb and Sons, Inc., Princeton, NJ) to minimize ocular sensation during the application of infectious virus. A nontraumatic method of viral infection was employed. Twenty-five microliters of a suspension of infectious McKrae strain HSV-1 (1 X 10 7 PFU/ml) was applied to the ocular surface of both eyes of each rabbit. The lids were held closed and the eye gently massaged for 30 seconds after which the animal was returned to its cage. Control animals were treated exactly the same way except that they were mock infected with a 25 n\ volume of culture medium not containing infectious virus. All animals were examined every other day after infection (day 0) using the slit-lamp microscope and only those animals that gave definitive evidence of corneal infection were included in this study. Infected corneas had at least one epithelial dendritic lesion and the majority (44 of 50) had two or more dendrites. At each ocular examination and on the days animals were sacrificed, ocular swabs were taken for viral culture. The frequency of virus infection using this approach was 85%. All of the eyes that were determined to be infected by clinical examination were documented to be infected by an ocular swab and virus cultivation procedure. Only those rabbits that had an unequivocal corneal epithelial dendrite and from whom infectious virus could be cultured during the first 4 days of infection were included in this study. Rabbits were sacrificed by intravenous sodium pentobarbital (The Butler Co., Columbus, OH) euthanasia. Three infected and one uninfected rabbits were sacrificed at 3, 7, 12, 15, 17, 25, 30, 45, 55, 65 and 80 days after corneal infection. At the time the animals were sacrificed, they were bled for serum antibody studies and the corneas and trigeminal ganglia obtained for cocultivation and immunohistochemical study. Determination of Infectious Virus by Cultivation In Vitro Swabs of the ocular surface were performed as previously reported. 14 A sterile, Dacron-tipped applicator (American Hospital Supply/American Scientific Products, McGaw Park, IL) was rotated gently in the upper cul-de-sac of the orbit, then rotated gently across the cornea into the lower cul-de-sac and allowed to rest in the nasal fornix for 10 seconds. The lower part of the applicator was broken off and immediately placed into a 15 X 120 mm tissue culture tube containing a monolayer of primary rabbit kidney cells (PRK, Whittaker MA Bioproducts, Walkersville, MD) in 1 ml of Eagle's Minimum Essential Medium (E-MEM, Grand Island Biological Co., Grand Island, NY) supplemented with 7% fetal bovine serum (Grand Island Biological Co.). Bacterial and fungal growth was inhibited by adding 1 ml of a penicillin, streptomycin and Fungizone solution (Grand Island Biological Co.). The tubes were incubated for 24 hr at 37 C in an atmosphere of 5% CO 2, 95% air and then the swabs were squeezed against the tube wall and removed. Two milliliters of culture medium was added and the tubes incubated for 14 days. The cell monolayers were examined daily for the appearance of cytopathic effect (CPE). The corneas from infected and uninfected rabbits that were sacrificed at the times indicated above were dissected free of other ocular tissue by means of sterile scissors and forceps. Each cornea was hemisected and one-half immediately preserved in cold 95% ethyl alcohol for later embedding and sectioning. The other half of the cornea was minced in 1 ml of tissue culture medium and homogenized in a Brinkmann polytron. Cell and tissue debris was removed by centrifugation at 5000 g for 10 min. One-half milliliter aliquots of the supernate were added to PRK monolayers, and incubated for 24 hr at 37 C in a CO 2 incubator. Two milliliters of culture medium was added and the tubes reincubated for 14 days, during which the cultures were monitored for CPE.

3 No. 11 CORNEAL HSV INFECTION / Gebhordr and 1685 The right and left trigeminal ganglia were bisected dorsoventrally to yield right and left halves of equal size. One half was immediately preserved in cold 95% ethyl alcohol for subsequent immunohistochemistry. The other half was minced and sonicated in 1 ml of tissue culture medium for three 10 second pulses. Cell debris was removed by centrifugation as described above and two 0.5 ml aliquots were added to PRK monolayers and incubated for 24 hr at 37 C in a CO2 incubator. Five milliliters of culture medium was added and the tubes reincubated for 14 days, during which CPE was monitored by daily microscopic examination. Enzyme Linked Immunosorbent Assay (ELISA) for Serum Anti-HSV-1 Antibody McKrae strain HSV-1 was propagated in PRK cells. Seventy-five square centimeter tissue culture flasks were inoculated with 1 X 10 5 PFU/ml of virus and incubated for 36 hr. Mock-infected PRK cell monolayers served as a control. The cell monolayers were washed three times with sterile saline and the flasks were frozen and thawed three times. The cell debris was collected in 5 ml of sterile saline, centrifuged at 10,000 g for 20 min, and the supernate assayed for protein concentration using the technique developed by Bradford. 30 The concentrated virus and viral protein were stored frozen at -80 C until used. The ELISA was performed in Immunolon I, 96- well microtiter plates (Dynatech Laboratories, Alexandria, VA). Preliminary experiments established that infected cell lysates gave optimal results when used at a concentration of 10 ^g protein/ml to coat ELISA plate wells. A cell lysate from mock-infected cells was used at the same protein concentration to coat control plate wells. Two-tenths milliliter of HSV-1 antigen or control antigen was added to plate wells and the plates incubated overnight at 4 C. The plate wells were washed three times in Tris-buffered saline (TBS, 20 mm Trisbase, 500 mm NaCl) and the remaining protein binding sites blocked by adding 0.2 ml TBS supplemented with 3% bovine serum albumin (TBS-BSA, Sigma Chemical Co., St. Louis, MO). After 2 hr incubation at 4 C the plates were washed three times with TBS containing 0.5% Tween 20 (T-TBS, Biorad Laboratories, Inc., Richmond, CA) and either used immediately or stored at 4 C. Fifty microliters of each serum sample from uninfected rabbits or HSV-1 -infected rabbits was added to the wells of control and HSV-1 antigen-coated plates. The serum samples were diluted through 15 twofold dilutions (1:1-1:16,384) using a Titertek multidiluter (Flow Laboratories, Inc., McLean, VA). Each time an ELISA was performed a positive control serum (Rabbit anti-hsv-1, Accurate Chemical and Scientific Corp., Westbury NY) was included as a standard. A negative control serum sample and wells without any serum were included each time the assay was performed. The plates were incubated for 2 hr at 22 C, washed three times with T-TBS and 50 n\ of a 1:500 dilution of alkaline phosphatase-conjugated goat anti-rabbit IgG (Organon-Teknika-Cappel, Inc., Malvern, PA) was added to all wells. The plates were incubated for 1 hr at 22 C and washedfivetimes with T-TBS; 50 y\ of alkaline phosphatase substrate solution (5 mg para-nitrophenyl phosphate in 5 ml of 10% diethanolamine buffer, ph 9.8, Sigma Chemical Co.) was added to each well. The plates were incubated at 37 C for 1 hr and optical density readings at 405 nm were taken at 15 min intervals using a Titertek Multiskan plate reader (Flow Laboratories). The data shown were obtained at the 45 min time point, as this appeared to be optimal for these experiments. As employed here the ELISA should not be considered a quantitative assay, but rather a comparative assay in which known positive and negative controls were used for comparison with sera obtained from animals at different times after infection. Tissue Preparation and Immunohistochemical Staining A variety of tissue preparation and fixation methods were tested in preliminary investigations; the following technique was adopted because of its sensitivity, specificity and the low background staining obtained. Corneal tissue and neural ganglionic tissue to be used for immunohistochemical staining were obtained as quickly as possible after the animal was sacrificed by anesthetic overdose. The tissues were immediately submerged in cold 95% ethyl alcohol for a period of approximately 18 to 24 hr. Next, the tissues were submerged overnight in a 30% solution of sucrose. Subsequently, the tissues were frozen in optimum cutting temperature (OCT, Miles Laboratories, Naperville, IN) compound and sectioned as thinly as possible in the cryostat. By trial and error, we found that tissue sections obtained from corneas and ganglia treated in this way retained morphology that was adequate for the localization of viral antigens and infiltrating cells. The avidin-biotin immunoperoxidase staining procedure was modified from that originally described by Hsu et al. 31 Tissue sections of corneas and trigeminal ganglia to be stained were first incubated for 30 min in a solution of 100% methanol containing 0.3% hydrogen peroxide. The slides were rinsed in phosphate buffered saline (PBS) and incubated for 30 min in 3% horse serum. The monoclonal antibody

4 1686 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / November 1988 Vol. 29 Table 1. Presence of infectious (cell-free) HSV in the cornea and trigeminal ganglia* Time after infection (days) Tearfilm\ Sampling site Cornea.% 2/6 Trigeminal ganglion^ 4/6 4/6 5/6 5/6 2/6 1/6 Fifty NZW rabbits were infected with the McKrae strain of HSV-1. Slit lamp examinations and ocular surface swabs were performed every other day after infection (day 2, 4, 6, 8, etc.) for 30 days and then weekly thereafter. Three infected rabbits and one uninfected rabbit were sacrificed at 3, 7, 12, 15, 17, 20, 25, 30,45, 55, 65 and 80 days after infection for determination of infectious virus in the cornea and trigeminal ganglia and for immunohistochemical staining for virus infected cells and T lymphocytes. t Ocular swabs were tested for infectious virus on PRK cell monolayers. The data are given as the number of virus positive samples over the number of samples tested. $ Portions of the corneas and trigeminal ganglia were minced, homogenized, and tested for infectious virus on PRK monolayers. specific for rabbit T lymphocytes was obtained from a cell culture supernatant of a hybridoma cell line designated LI 1/135 (American Type Culture Collection, Rockville, MD). This hybridoma was produced by Jackson et al 32 by fusing cell surface glycoproteins of rabbit T cells with the P3X63Ag8-Ul myeloma. The LI 1/135 hybridoma secretes an IgG, antibody that binds to a T cell differentiation marker on mature ConA- and alloantigen-responsive T lymphocytes from rabbit blood and lymphoid tissues. 32 All tissue sections to be stained for rabbit T lymphocytes were incubated for 1 hr in undiluted tissue culture culture supernatant containing the anti-t cell antibody. The tissue sections were thoroughly washed for 3-5 min intervals in PBS and then incubated for an additional 30 min in biotinylated horse antimouse IgG antibody (Vector Laboratories, Burlingame, CA). The tissue sections were again washed for 30 min and incubated in an avidin-biotin complex (ABC) reagent for 1 hr followed by washing in PBS. After incubation in the ABC reagent, the slides were washed as above and all sections were incubated in a 1 mg/ml solution of diaminobenzidine (Sigma Chemical Co.) in PBS, containing 50 n\ of an 8% solution of nickel chloride and 0.03% hydrogen peroxide. Incubation of the slides in this solution for a period of 10 min resulted in the development of a brown to brownish-black color around the membranes of cells to which the monoclonal anti-t lymphocyte antibody had bound. The slides were then washed in PBS, counterstained in methyl green stain in acetate buffer, dehydrated in 100% ethyl alcohol, cleared in toluene and coverslipped with Eukitt mounting medium (Calibrated Instruments, Inc., Ardsley, NY). All slides were examined and representative photographs taken. Cryostat sections of corneas and trigeminal ganglia were also stained for the presence of viral antigens using a direct immunoperoxidase staining technique. Six micron sections affixed to gelatin-coated slides were hydrated in PBS for 5 min and incubated for 1 hr in a 1:20 dilution of peroxidase-conjugated rabbit anti-hsv-1 (Accurate Chemical and Scientific Corp.). The tissue sections were rinsed in three 5 min washes of PBS and incubated for 10 min in the diaminobenzidine substrate solution prepared as described above. The sections were rinsed in PBS, counterstained with methyl green and coverslipped for microscopic examination. A total of 72 trigeminal ganglia from infected rabbits obtained between days 3 and 80 after infection and 22 ganglia from uninfected rabbits were examined following immunohistochemical staining for the presence of T lymphocytes and virus-infected neurons. Five pairs of anterior-posterior cryostat sections atfivedifferent levels of individual ganglia were stained and examined for T cells and HSV-1 antigen expressing cells. Results Demonstration of Infectious Virus in the Cornea and Trigeminal Ganglia On days 3, 7, 12, 15, 17, 25, 30, 45, 55, 65 and 80, four rabbits (three infected and one uninfected) were sacrificed and portions of their corneas and trigeminal ganglia homogenized and incubated on PRK cells for the determination of cell-free infectious virus. Ocular swabs were obtained from all remaining eyes every other day for the first 30 days after infection and weekly thereafter for 50 days. The last four animals were sacrificed 80 days after infection. Table 1 summarizes the results of assays for the presence of cell-free virus during acute and latent infection in the tears, corneas and trigeminal ganglia. Not unexpectedly, we found infectious virus in the corneas of rabbits early in the acute infection, but were largely unable to demonstrate virus by 15 days after primary coraeal inoculation. This was true for both the ocular swabs and the homogenates of corneal tissue. Similarly, cell-free virus was found in the trigeminal ganglia by day 4 after primary corneal infection and continued to be present up to day 25 after which it was no longer possible to find infectious virus.

5 No. 11 CORNEAL HSV INFECTION / Gebhordr and 1687 Fig. 1. The serum antibody response to HSV-1 following a corneal infection was measured by ELISA. Serum from uninfected and infected rabbits was collected on the days indicated and the ELISA antibody levels determined. Each data point (squares and solid line) represents the mean antibody level for three infected rabbits. Positive (triangles and dashed line) and negative (filled circles and dot and dash line) serum controls were included in each assay. The T cells present in TG -H Days after Infection data shown were obtained after 45 min incubation at 37 C. The horizontal line indicates the window of time (15-45 days) during which T lymphocytes were found in the trigeminal ganglion (TG). The mean optical density values of the serum samples obtained at 30 and 80 days after infection were found to be statistically significantly different from each other by analysis of variance (P > 0.001). The intervening mean values did not differ significantly from each other or the 30 and 80 day points. Analysis of Serum Anti-HSV-1 Antibody Titer In this investigation we found a lag of approximately 12 days before the serum IgG anti-hsv response became significantly elevated above the level of the controls (Fig. 1). Following this lag in synthesis of specific HSV antibody, the antibody level, as reflected by optical density readings in ELISA, rose relatively rapidly and reached a peak by approximately 30 days after primary infection. Furthermore, it was noted that the antibody level underwent a slow decay over the subsequent 50 days of the experiment (Fig. 1). These results not only demonstrate the delay in antibody production following corneal infection, but also indicate how long the antibody titer is sustained following the time at which infectious virus can be demonstrated in either a peripheral tissue such as the cornea, or a more central tissue such as the trigeminal ganglion. Immunoperoxidase Staining for HSV-1 Antigens and T Lymphocytes We found that there is a relatively narrow window of time during which immunocompetent T lymphocytes are present in the ganglia of infected animals. From 0 to 12 days after corneal infection T lymphocytes were not seen in parenchyma of the trigeminal ganglia, and no margination of T cells in the blood vessels associated with this organ were seen. Seventeen days after topical corneal infection, a significant perivascular infiltrate of T lymphocytes and a scattering of T lymphocytes within the nerve fibers innervating the ganglion were seen (Fig. 2). The continued presence of T cells was seen in trigeminal ganglia obtained from rabbits as late as 45 days after corneal infection. Twenty-eight of the 30 trigeminal ganglia from infected animals examined between 17 and 45 days after infection contained a T cell infiltrate; such cells were not found in uninfected controls. Neurons expressing HSV-1 antigens were found in three of the 30 ganglia and in a total of ten of 72 ganglia examined. By 55 days T lymphocytes were no longer evident within the tissue of the ganglia and the perivascular infiltrates were gone as well. We were particularly interested in determining how the presence of viral antigen in the cornea and trigeminal ganglia correlated with the presence of T lymphocytes in the ganglion. We found that the frequency of infected neurons as demonstrated by immunoperoxidase staining for HSV-1 antigens was very low. All of the neuron cell bodies in five cryostat sections from each of 72 ganglia were carefully examined for viral antigens as revealed by immunoperoxidase staining. Neurons expressing HSV-1 antigens were found in ten of the 72 ganglia examined and in these ten, we found only one to three positive cell bodies in the five sections examined; none were found in the 22 ganglia from uninfected rabbits. Quantitation of the frequency of infected neurons per ganglion or animal was not possible by immunohistochemical staining. The majority of ganglia from infected rabbits were negative by immunoperoxidase staining. It is possible that cell bodies expressing HSV-1 antigens were missed because of the limited number of sections examined. In other studies not reported here we have conducted thorough and exhaustive immunohistochemical analysis of many

6 1688 INVESTIGATIVE OPHTHALMOLOGY G VI5UAL SCIENCE / November 1988 Vol. 29 Fig, 2. Immunoperoxidase stain of a trigeminal ganglion revealing the presence of T lymphocytes. (A) Margination of T cells (arrows) around a ganglionic capillary (C). (B) Scattered T cells (arrows) in the ganglionic parenchyma. Methyl green counterstain (X125)..1, sections from individual ganglia from infected rabbits and have found very few neuron cell bodies that contain viral antigens. Figure 3 is a photomicrograph of a neuron expressing HSV-1 antigen in the trigeminal ganglion of a rabbit 25 days after primary infection. As can be seen, in this section there is a single cell body that expresses HSV-1 antigen surrounded by a field of other cell bodies not expressing these antigens (Fig. 3). We were able to demonstrate the presence of viral antigen in corneal sections during the same interval of time that it was possible to culture infectious virus from this tissue. Figure 4 is a photomicrograph of a cross section of a dendritic lesion of a rabbit cornea infected seven days previously with McKrae strain HS V-1. The infected epithelial cells of this dendritic lesion clearly express viral antigen as demonstrated by the immunoperoxidase staining technique. Discussion Neither T cells nor HSV antigen expressing cells were found in the ganglia of uninfected rabbits. We

7 No. 11 CORNEAL H5V INFECTION / Gebhardr and Hill 1689 Fig. 3. Immunoperoxidase stain for HSV-1 antigens in the trigeminal ganglion section of a rabbit infected 25 days previously reveals the infrequent neuron cell body (arrow) expressing HSV-1 antigens. Other neuron cell bodies (CB) in this photomicrograph did not stain for viral antigens. Methyl green counterstain (XI25). have found that immunocompetent T lymphocytes are present in a distribution characteristic of tissues undergoing virus invasion33 and that the period of time during the establishment of latency at which the T cells are present is relatively short. Immunoglobulin positive B lymphocytes have not been identified in companion sections found to contain T cells. We have established that the T lymphocytes are present in and around the vessels nourishing the trigeminal ganglion and are sparsely distributed among the nerve routes of the ganglion only in rabbits in which the corneal infection had been successfully established and which were thus predisposed to trigeminal ganglion colonization although this was not Fig. 4. Immunoperoxidase stain of a corneal epithelial defect demonstrating the presence of HSV-1 antigens in epithelial cells (arrows). Dark brown-black cells are HSV-1 infected. Methyl green counterstain (X125).

8 1690 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / November 1988 Vol. 29 proven by cocultivation in this study. The portions of the trigeminal ganglia from these same animals examined immunohistochemically during the time at which latency was being established (15 to 30 days) revealed the presence of immunocompetent T lymphocytes which were presumably infiltrating the ganglion in response to the presence of viral antigens. Such lymphocytes were not present in the ganglia of uninfected rabbits, nor in the ganglia of rabbits that were more than 46 days beyond primary coraeal infection. The disappearance of T lymphocytes from the trigeminal ganglia of animals that had become latent for HSV-1 suggests that viral antigen expression and perhaps the presence of cell-free virus dips below the threshold level necessary for T cell signaling. Studies are in progress to determine if and when the T lymphocytes return during spontaneous or induced reactivation. It is possible that the cellular character of the intraganglionic immune response is different from the relatively "pure" T cell infiltrate seen during primary colonization. Further study will provide information on this point also. In parallel immunohistochemical analyses of rabbit trigeminal ganglionic tissue, we attempted to localize the presence of viral antigens in virus-infected neurons. Using both polyclonal antibodies and monoclonal antibodies specific for HSV-1 antigens, we found a low frequency of infected neurons. We conclude that in rabbits undergoing topical corneal infection by HSV-1, and in whom neural latency is established, the number of neurons that are actually successfully colonized is small and the extent of proliferation and virus spread and infection in the trigeminal ganglion is limited. This conclusion is supported by the recent studies using in situ and dot hybridization with cloned HSV probes, 7 ' 34 " 36 which indicate the sparse infection of trigeminal ganglion cells during primary infection. Using a DNA probe one study found that significantly fewer rabbit neurons hosted an acute viral infection, as compared to the neurons of the mouse ganglion. 7 We conclude that in the rabbit the extent of virus invasion and proliferation in the trigeminal ganglion is extremely limited. This conclusion is in agreement with those of other investigators who speculated that the number of ganglionic neurons that are latently infected is extremely low and that latency can be maintained without the expression of significant cytopathology in theganglion Nevertheless, the revelation that T lymphocytes enter the trigeminal ganglion during the time that neural latency is being established indicates that the immune system of the host is activated in the attempt to limit the spread of the virus. Work is in progress to determine if T lymphocytes reenter the ganglion during adrenergic reactivation of latent HSV-1. Key words: cornea, herpes simplex virus, infection, T lymphocytes, trigeminal ganglion References 1. Knotts FB, Cook ML, and Stevens JG: Latent herpes simplex virus in the central nervous system of rabbits and mice. J Exp Med 138:740, Nesburn AB and Green MT: Recurrence in ocular herpes simplex infection. Invest Ophthalmol 15:515, Tullo AB, Easty DL, Shimeld C, Stirling PE, and Darville JM: Isolation of herpes simplex virus from corneal discs of patients with chronic stromal keratitis. Trans Ophthalmol Soc UK 104:159, Stulting RD and Abghari SZ: Recovery of herpes simplex virus type 1 from latently infected ocular tissues of inbred mice. ARVO Abstracts. Invest Ophthalmol Vis Sci 28(Suppl):177, Cook SD and Brown SM: Herpes simplex virus type 1 latency in rabbit corneal cells in vitro: Reactivation and recombination following intratypic superinfection of long term cultures. J Gen Virol 68:813, Green MT, Courtney RJ, and Dunkel EC: Detection of an immediate early herpes simplex virus type 1 polypeptide in trigeminal ganglia from latently infected animals. Infect Immunol 34:987, Stroop WG and Schaefer DC: Herpes simplex virus type 1 invasion of the rabbit and mouse nervous systems revealed by in situ hybridization. Acta Neuropathol 74:124, Margolis T, Togni B, La Vail J, and Dawson CR: Identifying HSV infected neurons after ocular inoculation. Curr Eye Res 6:119, Gebhardt BM, Cutrone C, and Kaufman S: Cell-mediated immunity in HSV keratitis: Modulating effect of serum antibodies. In Advances in Immunology and Immunopathology of the Eye, O'Connor GR and Chandler JW, editors. New York, Masson, 1985, pp Kwon BS, Gangarosa LP, Burch KD, deback J, and Hill JM: Induction of ocular herpes simplex virus shedding by iontophoresis of epinephrine into rabbit cornea. Invest Ophthalmol Vis Sci 21:442, Hill JM, Kwon BS, Shimomura Y, Colborn GL, Yaghmai F, and Gangarosa LP: Herpes simplex virus recovery in neural tissues after ocular HSV shedding induced by epinephrine iontophoresis to the rabbit cornea. Invest Ophthalmol Vis Sci 24:243, Shimomura Y, Dudley JB, Gangarosa LP, Sr, and Hill JM: HSV-1 quantitation from rabbit neural tissues after epinephrine-induced reactivation. Invest Ophthalmol Vis Sci 26:121, Berman EJ and Hill JM: Spontaneous ocular shedding of HSV-1 in latently infected rabbits. Invest Ophthalmol Vis Sci 26:587, Hill JM, Rayfield MA, and Haruta Y: Strain specificity of spontaneous and adrenergically induced HSV-1 ocular reactivation in latently infected rabbits. Curr Eye Res 6:91, Rock DL and Fraser MW: Detection of HSV-1 genome in central nervous system of latently infected mice. Nature 302:523, Galloway DA, Fenoglio CM, and McDougall JK: Limited transcription of the herpes simplex virus genome when latent in human sensory ganglia. J Virol 41:686, 1982.

9 No. 11 CORNEAL HSV INFECTION / Gebhordr and Hill Galloway DA, Fenoglio C, Sherchuk M, and McDougall JK: Detection of herpes simplex RNA in human sensory ganglia. Virology 95:265, Stevens JG, Wagner EK, Devi-Rao GB, Cook ML, and Feldman LT: RNA complementary to a herpesvirus a gene mrna is prominent in latently infected neurons. Science 235:1056, Cantin EM, Puga A, and Notkins AL: Molecular biology of herpes simplex virus. In Concepts in Viral Pathogenesis, Notkins AL and Oldstone MB, editors. New York, Springer-Verlag, 1984, pp Mellerick DM and Fraser NW: Physical state of the latent herpes simplex virus genome in a mouse model system: Evidence suggesting an episomal state. Virology 158:265, Fung JC, Shanley J, and Tilton RC: Comparison of the detection of herpes simplex virus in direct clinical specimens with herpes simplex virus-specific DNA probes and monoclonal antibodies. J Clin Microbiol 22:748, Stroop WG and Schaefer DC: Severity of experimentally reactivated herpetic eye disease is related to the neurovirulence of the latent virus. Invest Ophthalmol Vis Sci 28:229, Deatly AM, Spivack JG, Lavi E, and Fraser NW: RNA from an immediate early region of the type 1 herpes simplex virus genome is present in the trigeminal ganglia of latently infected mice. Proc Natl Acad Sci USA 84:3204, Puga A and Notkins AL: Continued expression of a poly(a) + transcript of herpes simplex virus type 1 in trigeminal ganglia of latently infected mice. J Virol 61:1700, Stroop WG and Schaefer DC: Production of encephalitis restricted to the temporal lobes by experimental reactivation of herpes simplex virus. J Infect Dis 153:721, Green MT, Dunkel EC, and Pavan-Langston D: Effect of immunization and immunosuppression on induced ocular shedding and recovery of herpes simplex virus in infected rabbits. Exp Eye Res 45:375, Klein RJ: The pathogenesis of acute, latent and recurrent herpes simplex virus infections. Arch Virol 72:143, Wander AH, Bubel HC, and McDowell SG: The pathogenesis of herpetic ocular disease in the guinea pig. Arch Virol 95:197, Green MT, Dunkel EC, and Courtney RJ: Detection of herpes simplex virus induced polypeptides in rabbit trigeminal ganglia. Invest Ophthalmol Vis Sci 25:1436, Bradford M: A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248, Hsu SM, Raine L, and Fanger H: The use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase technique: A comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 29:577, Jackson S, Chused TM, Wilkinson JM, Leiserson WM, and Kindt TJ: Differentiation antigens identify subpopulations of rabbit T and B lymphocytes. J Exp Med 157:34, Sobel RA, Collins AB, Colvin RB, and Bhan AK: The in situ cellular immune response in acute herpes simplex encephalitis. AmJPathol 125:332, Rock DL and Fraser NW: Detection of HSV-1 genome in central nervous system of latently infected mice. Nature 302:523, Rock DL, Nesburn AB, Ghiasi H, Ong J, Lewis TL, Lokensgard JR, and Wechsler SL: Detection of latency-related viral RNAs in trigeminal ganglia of rabbits latently infected with herpes simplex virus type 1. J Virol 61:3820, Caudill JW, Romanowski E, Raullo-Cruz T, and Gordon YJ: Recovery of a latent HSV-1 thymidine kinase negative strain following iontophoresis and cocultivation in the ocularly-infected rabbit model. Curr Eye Res 5:41, 1986.