The Role of Apoptosis within the Retina of Coronavirus-Infected Mice

Abstract

purpose. To evaluate the possible roles of apoptosis in the murine retinopathy induced by coronavirus.

methods. Mice were inoculated with virus intravitreally. Mouse eyes harvested at varying times after inoculation were evaluated for apoptotic and immunologic events by hematoxylin and eosin staining, immunohistochemical staining, in situ terminal deoxynucleotidyltransferase dUTP nick-end labeling (TUNEL) assay, and electron microscopy. Isolated retinas were analyzed for infectious virus and for expression of apoptosis-associated genes.

results. The number of apoptotic events was significantly elevated in infected eyes from BALB/c and CD-1 mouse strains, reaching a maximum at days 6 through 10, and returning to normal levels at day 20. The majority of apoptotic cells were observed in the outer nuclear layer of the infected retina. In contrast, few apoptotic cells were observed in normal or mock-injected mouse eyes. Apoptotic events within the retina were associated with the presence of viral antigen, infiltration of CD8⁺ T cells, and clearance of infectious virus. Reverse transcription–polymerase chain reaction (RT-PCR) analysis identified the upregulation of Fas ligand (FasL) and granzyme B mRNAs within the infected retinas. The development of apoptosis, regulative gene expression, and viral clearance were similar in both retinal degeneration–susceptible (BALB/c) and –resistant (CD-1) mice.

conclusions. Retinal apoptosis was associated with retinal inflammation, a decrease in infectious virus, and upregulation of genes associated with CTL killing. These studies indicate that retinal apoptosis may be one of the host mechanisms that contribute to limiting this retinal infection.

Apoptosis is a process of cell suicide, which is involved in the development and pathogenesis of a variety of diseases, including cancer, autoimmune disease, and viral infection. ^{1 2 3 4} Recent laboratory studies indicate that apoptosis is involved in many ocular diseases, such as glaucoma, retinitis pigmentosa, cataract formation, retinoblastoma, retinal ischemia, diabetic retinopathy, and ocular murine herpes simplex virus infection. ^{5 6 7}  

A number of studies have shown that apoptosis plays an important role in virus infections by direct virus induction of cell death and/or by cytotoxic T lymphocyte (CTL) killing of infected cells. ^{8 9} Cytotoxicity is usually associated with CD8⁺ T cells; however, CD4⁺ T cells may also present a cytotoxic response. ¹⁰ CTLs can kill their targets by two contact-dependent mechanisms: a secretory and membranolytic pathway involving perforin and granzymes and a nonsecretory receptor-mediated pathway involving Fas-FasL. ^{11 12 13} In the perforin-granzyme pathway, target-cell lysis is induced as a consequence of T-cell receptor activation and the release of lytic granules. These granules contain perforin and serine proteases known as granzymes. In the Fas/FasL pathway, target cell cytotoxicity is initiated when FasL expressed predominantly on activated T cells, binds Fas on the target cells. 

Coronaviruses cause a number of diseases in mammalian and avian species. ¹⁴ Recently, we reported that the murine coronavirus, a mouse hepatitis virus JHM strain, induces a biphasic retinal disease in adult BALB/c mice. The early phase is associated with retinal vasculitis and viral replication in selected cells, such as retinal pigment epithelial cells, ciliary body epithelial cells, and Müller-like cells. The late phase is associated with retinal degeneration, development of anti-retinal and anti-RPE cell autoantibodies, and the absence of infectious virus. ¹⁵ Furthermore, this model is characterized by a genetic predisposition. ¹⁶ The virus induces an inflammation in the retinas of CD-1 mice only in the early phase of disease. In these CD-1 mice, the retinal architecture returns to a normal appearance in the late phase of disease, and no autoantibodies are observed. 

The mechanisms responsible for retinal tissue damage during viral infections are poorly understood. In this study, we evaluated apoptosis in JHM virus-induced retinopathy. Moreover, we determined the relationship among the development of apoptosis, JHM viral clearance, and retinal degeneration in both retinal degeneration–susceptible and– resistant mice. 

Materials and Methods

Mice and Virus Inoculations

BALB/c mice were obtained from Harlan Sprague Dawley (Indianapolis, IN). CD-1 mice were obtained from Charles River (Raleigh, NC). All experimental procedures conformed with the ARVO Resolution for the Use of Animals in Ophthalmic and Vision Research. Ten- to 12-week-old male mice were inoculated by the intravitreal route with 5 μl of 10^5.5 median tissue culture infective dose (TCID₅₀) per milliliter of JHM virus or minimum essential medium (MEM; Gibco, Gaithersburg, MD). At days 1, 3, 4, 6, 7, 8, 10, and 20 after inoculation, mouse eyes and livers were removed and fixed in 10% formalin or 2.5% glutaraldehyde for hematoxylin and eosin staining, in situ terminal deoxynucleotidyltransferase dUTP nick-end labeling (TUNEL) assay, or electron microscopy. Tissues were also frozen with and without optimal cutting temperature compound (OCT; Miles, Elkhart, IN) at −20°C for immunohistochemical staining or for viral infectivity assay. 

Virus

JHM virus was obtained from the American Type Culture Collection (Rockville, MD). Stock virus was propagated in mouse L2 cells (a gift from Kathryn Holmes, University of Colorado, Denver), harvested by centrifugation at 15,000g for 2 hours, and the pellet resuspended in MEM. 

Virus titers were determined by 10% homogenization of infected eyes onto mouse L2 cells in a 96-well microtiter plate. Infectivity was recorded as the induction of cytopathic effect by serial 10-fold dilution of the samples as described. ¹⁷  

TUNEL Assay

TUNEL assay was performed with a kit (ApopTag Plus-In Situ Apoptosis Detection; Oncor, Gaithersburg, MD). Mouse eye sections (8μ m thick) were deparaffinized and digested in 20 μg/ml of proteinase K for 5 minutes. Endogenous peroxidase was blocked by pretreatment with 2% hydrogen peroxide in phosphate-buffered saline (PBS) for 5 minutes. Sections were preincubated with equilibration buffer for 10 minutes and then incubated with terminal transferase enzyme (TdT) and digoxigenin (DIG)-11 dUTP at 37°C for 60 minutes. The enzymatic reaction was terminated by incubating the sections in stopping buffer for 30 minutes. Anti-DIG peroxidase was applied for 30 minutes, and 0.05% 3,3-diaminobenzidine (DAB) substrate plus 0.3% H₂O₂ were added for color development. All steps were performed at room temperature, except when indicated otherwise. The sections were counterstained in 0.5% methyl green and observed by light microscopy. 

Antibodies and Immunohistochemistry

Rat IgG2b monoclonal antibody reacting with L3/T4 (Sera-Laboratory, Crawley Down, UK) was used to identify CD4⁺ T helper cells. Rat IgG2b monoclonal antibody reacting with Lyt-2 (Sera-Laboratory) was used to identify CD8⁺ suppressor-cytotoxic T cells. Rat IgG2b monoclonal antibody reacting with Mac-1 (Boehringer Mannheim, Indianapolis, IN) was used to identify both macrophages and natural killer (NK) cells. Mouse F88 monoclonal antibody (a gift from Julian Leibowitz, University of Texas, Houston), was used to identify JHM virus. Mouse IgG (Biodesign, Kennebunkport, ME) and rat IgG (Organon Teknika, Durham, NC) were used as negative controls. 

For double-fluorescence staining studies, frozen eye sections were air dried, digested by proteinase K, and incubated with primary antibodies for 30 minutes. After washing in PBS, the sections were incubated with TdT enzyme, DIG-labeled deoxynucleotides, and biotinylated nucleotides at 37°C for 1 hour. Goat anti-rat IgG or horse anti-mouse IgG conjugated with Texas red and streptavidin conjugated with FITC (Vector, Burlingame, CA) were applied for 30 minutes. All steps were performed at room temperature, except when indicated otherwise. The slides were evaluated by a laser scanning confocal fluorescence microscope (TCS-4D-DMIR-BE; Leica, Heidelberg, Germany). 

Electron Microscopy

Mouse eyes were fixed in 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer (PB) for 2 hours at 4°C, washed in PB, postfixed with 1% osmium tetroxide in PB for 1 hour at 4°C, dehydrated in graded ethanol, and embedded in Spurr’s compound. Ultrathin sections were stained with uranyl acetate and lead citrate and examined with a transmission electron microscope (Carl Zeiss, Thornwood, NY). 

Reverse Transcriptase–Polymerase Chain Reaction

For reverse transcriptase–polymerase chain reaction (RT-PCR), each retinal RNA sample was dissected and pooled from eight mouse eyes. RNA was isolated by a commercially available protocol (RNA Stat-60; Tel-Test, Friendswood, TX). cDNA was synthesized from the same 0.5 μg of RNA and was amplified by using a kit (GeneAmp RT- PCR; Perkin Elmer–Roche Molecular Systems, Branchburg, NJ) and specific primers as follows: Fas primers ¹⁸ : 5′-CAGACATGCTGTGGATCTGG-3′ (sense), 5′-CACAGTGTTCACAGCCAGGA-3′ (antisense), yielding a 422-bp product; FasL primers ¹⁹ : 5′-CAGCTCTTCCACCTGCAGAAGG-3′ (sense), 5′-AGATTCCTCAAAATTGATCAGAGAGAG-3′ (antisense), yielding a 509-bp product; perforin primers ²⁰ : 5′-GTCACGTCGA AGTACTTGGTG-3′ (sense), 5′-ATGGCTGATAGCCTGT CTCAG-3′ (antisense), yielding a 202-bp product; granzyme B primers ²¹ : 5′-CCCAGGCGCAATGCTAAT-3′ (sense), 5′-CCAGGATAAGAAACTCGA-3′ (antisense), yielding a 330-bp product; Bcl-2 primers ²² : 5′-AGAGGGGCTACGAGTGGGAT-3′ (sense), 5′-CTCAGTCATCCACAGGGCGA-3′ (antisense), yielding a 454-bp product; interleukin-1β converting enzyme (ICE) primers ²³ : 5′-GGTACCCATACTGATAGTGG-3′ (sense), 5′-GATTCATGGTTGGCAGCTAC-3′ (antisense), yielding a 511-bp product; and p53 primers, c-fos primers, and β-actin primers (Clontech, Palo Alto, CA). PCR products were electrophoresed in 4% agarose gels. After it was denatured and neutralized, DNA was transferred to a nylon membrane, hybridized with specific internal DIG-labeled oligonucleotide or cDNA probes made by using kits (Genius; Boehringer Mannheim). 

Results

Induction of Apoptosis in Infected Mouse Eyes

In this study, as in our previous studies, ¹⁶ ocular infection of BALB/c mice with JHM virus resulted in retinal inflammation at postinoculation (PI) days 1 to 8 and retinal degeneration after PI day 10. In contrast, JHM virus only induced the early inflammatory phase in CD-1 mice. After PI day 10, the CD-1 mouse retinas appeared normal. 

The kinetics of the development of apoptosis within the eyes of virus-infected BALB/c and CD-1 mice are shown in Figure 1 . Between PI day 1 and day 20, 17 BALB/c and 18 CD-1 eyes from untreated and mock-injected mice contained zero to eight TUNEL-positive cells (mean number per section). In contrast, 24 BALB/c and 15 CD-1 eyes from virus-infected mice contained significantly more TUNEL-positive cells. The number of apoptotic cells seen in the virus-infected eyes was low at PI days 1 and 3, reached a maximum at PI days 6 to 10, and returned slowly to normal levels. The TUNEL-positive cells identified from BALB/c and CD-1 mice at PI days 6 through 10 were 35 and 75, respectively. The difference between the apoptotic events in the virus-infected eyes compared with the control eyes (untreated and mock-injected eyes) was statistically significant in both strains of mouse eyes (P < 0.01, student t-test). At PI day 20, more apoptotic cells could still be detected in both strains of virus-infected mice in comparison with the uninfected mice. However, these differences seen at PI day 20 were not significant. Moreover, a comparison of apoptotic events within infected retinas (PI days 6 through 10) from these two strains of mice did not show a significant difference. 

TUNEL-positive cells were detected in the eyes from BALB/c and CD-1 mice, showing nuclear chromatin condensation and nuclear fragmentation. As seen in Figure 2 , the majority of TUNEL-positive cells were observed in the outer nuclear layer of the infected retina. Additionally, many TUNEL-positive infiltrating cells were noted in the vitreous and around the iris and ciliary body. Positive reactions were less frequently seen in the inner nuclear layer, ganglion cell layer, layers of outer and inner segments, and retinal pigment epithelium layer. Although TUNEL-positive cells were also seen in the choroid, there was no difference observed among untreated, mock injected, and virus-infected animals. 

Eyes were also evaluated by electron microscopy to further characterize apoptosis. Cells within the retina showed features that are consistent with apoptosis. As is seen in Figure 3A , these apoptotic cells were characterized by the presence of cell shrinkage, and condensation of chromatin and cytoplasm. A later stage of apoptosis may be seen in Figure 3B . The cell surface appeared smoother, the cell shrinkage was more compact, and the nuclear chromatin displayed a homogenous condensation. 

Because apoptotic events were observed in infiltrating cells within the vitreous and around the iris and ciliary body after infection, we compared the number of apoptotic events in the vitreous and the retina. As is seen in Figure 4 , the majority of apoptotic cells were observed in the retina when compared with the vitreous (P < 0.01, student t-test). In addition, apoptosis was more frequently observed in virus-infected CD-1 mice in comparison with virus-infected BALB/c mice. However, this difference was not significant (P = 0.07, student t-test). The data in Figure 4 compare apoptotic events in the retina and vitreous and do not include apoptotic events in the ciliary body. Occasionally apoptosis could be seen in the ciliary body region of the eye (Fig. 2B) ; however, this consisted of approximately 1% of the total apoptotic events observed. 

Comparison of Virus Replication and Apoptosis within Mouse Eyes

To determine the relationship between virus replication and apoptotic events within the retina, viral infectivity titers within the retinas from BALB/c and CD-1 mice at PI days 1, 3, 8, and 20 were evaluated. JHM virus was detected in the infected mouse eyes at PI days 1, 3, and 8. Peak levels of infectious virus were seen at PI day 3. In contrast, virus was not detected after day PI 8 (Fig. 5) . Eyes from untreated mice and mock-injected mice did not contain any detectable virus (data not shown). There was also no significant difference in levels of virus detected in retinas removed from BALB/c and CD-1 mice. 

Virus-Infected Retinal Cells Associated with Apoptosis

To determine whether apoptosis occurred in virus-infected cells, a double-labeled immunofluorescent staining assay was used to evaluate frozen retinal sections from BALB/c and CD-1 mice at PI day 6. Apoptotic cells were identified by FITC-labeled TdT, whereas viral antigens were identified by Texas red–labeled anti-JHM virus monoclonal antibody. Double-labeled immunofluorescent staining revealed that there were a few cells in the infected retinas that displayed a positive reaction for both viral protein and apoptosis. However, the majority of apoptotic cells within the retina did not display double labeling with viral protein. 

Retinal Cell Type Associated with Apoptosis

Maximum levels of apoptosis were observed between PI days 6 through 10 in both strains of mice. This is also the time for maximal infiltration of macrophages and CD8⁺ T cells (Authors’ unpublished data). To determine whether apoptosis is associated with infiltrating cells and/or within resident retinal cells, the double-labeled immunofluorescent staining assay was used. 

Apoptotic cells were identified from BALB/c and CD-1 mouse eyes at PI days 4, 7, and 10 by FITC-labeled TdT, and infiltrating cells were identified by Texas red–labeled monoclonal anti MAC-1, anti-CD4, and anti-CD8. The majority of infiltrating cells were MAC-1 positive. These were detected in 100% of the virus-infected eyes. There was no Mac-1–positive reaction associated with apoptotic staining. Infrequently, CD4⁺ T cells were observed; however, these were not associated with apoptotic markers. CD8⁺ T-cell staining was less intense than MAC-1 staining and was observed in 60% of the virus-infected eyes. Nevertheless, we detected CD8⁺ cells in the retinas that were associated with apoptotic markers (Fig. 6) . The majority of apoptotic staining was not associated with the presence of infiltrating cells in the infected retinas. 

When apoptotic cells were observed in association with infiltrating cells, those cells were CD8⁺ T cells in the retina. Examination of infected retinas by electron microscopy identified most of the apoptotic cells as retinal cells. As shown in Figure 3A , the apoptotic cells were similar in shape, size, and intracellular structure to resident retinal cells in the outer nuclear layer. 

Gene Expression of Fas, FasL, Perforin, and Granzyme B in Infected Mouse Retinas

There are a variety of cellular genes associated with various apoptotic events, such as Fas, FasL, perforin, granzyme B, p53, ICE, bcl-2, and c-fos. Pooled retinal mRNAs from untreated, mock-injected, and virus-infected BALB/c and CD-1 mice were evaluated by RT-PCR for the expression of these genes. Data (Table 1) evaluated from four experiments revealed that FasL and granzyme B mRNAs were not frequently detected in retinas from uninfected or mock-injected animals. However, FasL and granzyme B mRNAs were more frequently detected in the infected retinas. This was especially noteworthy at day 8 PI, when 100% of the retinal samples from BALB/c mice contained mRNAs specific for FasL and granzyme B. In total, FasL and granzyme B mRNAs were detected in 67% of virus-infected retinas and 17% of mock-injected retinas from BALB/c mice. Likewise, in CD-1 mice, these mRNAs were detected in 50% of virus-infected retinas and 9% of mock-injected retinas. In contrast, Fas, perforin, p53, ICE, bcl-2, and c-fos mRNAs were expressed within retinas obtained from untreated, mock-injected, and virus-infected BALB/c and CD-1 mice (data not shown). Together, these studies show an association between the upregulation of gene expression for Fas L and granzyme B and apoptosis in the infected retinas. 

Relationship of Viral Proteins and Apoptosis in Infected Mouse Livers

Intravitreal injection of JHM virus in BALB/c mice resulted in acute hepatitis in which infectious virus and viral protein were detectable within the liver at PI days 3 through 6. ²⁴ TUNEL-positive cells were also detected in JHM virus–infected livers, but were not detected in the livers from normal and mock-injected mice. After virus infection, TUNEL-positive cells were observed at day 3, increased at day 6, and decreased at day 10. At PI day 20, TUNEL-positive cells were detected infrequently. These data demonstrate that the temporal development of apoptosis in the infected liver was very similar to the apoptosis observed in the infected retinas. Thus, infectious virus clearance within the liver and the retina was associated with apoptosis. 

Discussion

Previous studies have shown that apoptosis within the retina was not frequently observed in the normal adult mouse but was observed during retinal cell development in the fetal and newborn mouse. ^{25 26} In addition, apoptosis has been observed in retinal degeneration induced by photic injury and experimental retinal detachment and in inherited retinal dystrophy in mice and rats. ^{27 28 29 30} Recently, Jen et al. ³¹ have used the retina to evaluate neurotoxicity of the amyloid-β peptide (Aβ), a peptide which may play a central role in the pathologic course of Alzheimer’s disease. They found that intravitreal injection of Aβ into rats resulted in photoreceptor apoptosis and neurotoxicity. These studies clearly identify that retinal apoptosis can lead to retinal cell damage. 

In our study, apoptosis was observed in JHM virus–infected mouse eyes, whereas very low levels of apoptosis were observed in untreated and mock-injected eyes. Maximal levels of retinal apoptosis occurred at days 6 through 10 after intravitreal inoculations, when retinal inflammation was present and infectious virus was eliminated. Extensive apoptosis was identified in infected eyes from both BALB/c and CD-1 mice. 

In this report, apoptosis within the retina was identified by the TUNEL method that labels the cut DNA ends and identifies DNA fragmentation. Electron microscopic analysis was used to support the TUNEL analysis. The major electron microscopic features of apoptosis observed in the virus-infected retinas were cell shrinkage and condensation of chromatin and cytoplasm. This is in contrast to necrosis, in which cells typically swell and lyse after damage to the cell membrane and entry of calcium. This analysis of retinal apoptosis is consistent with studies by Chang et al. ^{32 33} They observed that membrane blebbing is not a major feature of apoptosis within the mouse retina. When TUNEL analysis was performed at the electron microscopic level, permitting a correlation of DNA fragmentation with chromatin condensation, they found that 90% of the nuclei showing chromatin condensation were also TUNEL positive. 

Several studies have shown that many viruses, such as lymphocytic choriomeningitis virus, ¹¹ hepatitis virus, ¹² herpesviruses, ⁶ alphavirus, ³⁴ and retroviruses, ³⁵ result in activation of a programmed cell death pathway in a variety of cells. In addition to being associated with inflammation in virus infection, apoptosis has also been linked with reduction of viral replication and activation of cytokine responses. ^{36 37 38} Barac–Latas et al. ³⁹ have found that JHM virus–induced intracerebral infection of rats is associated with apoptosis. However, viral antigen is almost completely cleared from lesions at the time of apoptosis. In spite of the absence of detectable virus, ongoing demyelination and tissue destruction occurred with destruction of oligodendrocytes by apoptosis. 

The majority of apoptotic events identified within the retina were not directly coreactive with viral protein, indicating that alternative mechanisms of apoptosis might exist. Thus, the retinal apoptosis observed may be, at least in part, an indirect response to the virus infection. One possible mechanism of apoptosis induction could be alteration of levels of cytokines within the infected retina, similar to that reported in HIV-infected brain cells of dementia patients. ⁴⁰ In fact, preliminary studies on cytokine gene expression demonstrated that tumor necrosis factor (TNF), interleukin (IL)-1α, and IL-6 were activated within the infected retina (Authors’ unpublished data). 

A number of investigations have demonstrated the importance of CTLs in coronavirus clearance. ^{41 42 43} It is known that the CTLs induce cell death of their target cells primarily either by the surface interaction between Fas and FasL or by the release of perforin and granzyme, and those two pathways may account for all short-term cytolysis. ^{44 45 46 47} We demonstrated that apoptotic events within the retina occurred where viral antigens and CD8⁺ T cells were present. Moreover, FasL and granzyme B genes were upregulated within the retina after JHM virus infection. The presence of granzyme B gene expression at the time of CD8⁺ T cells infiltration suggests that these CD8⁺ T cells may be CTLs. Granzyme B is a major component of CTLs and triggers the cell death cascade by activating key target cell caspases and several downstream caspase substrates. In this way, granzyme B is thought to directly contribute to apoptotic nuclear morphology. ^{42 48}  

Experimental coronavirus retinopathy is characterized by an initial viral replication within selected target cells. Between days 1 and 3, virus can be detected in the RPE and ciliary body epithelial cells. Between days 4 and 6, virus is more frequently detected in Müller-like cells and some photoreceptors. ⁴⁹ The murine host responds in a variety of ways to initiate virus elimination within the retina. In this model, the host responds to this virus trigger by producing cytokines, inducing major histocompatibility complex (MHC) molecules on RPE cells, producing anti-virus neutralizing antibodies, and infiltrating the retina with macrophages and CD8 T cells. ^{50 51} In this study we observed an upregulation in apoptosis coincident with the presence of CD8 T cells and gene expression of granzyme B and FasL. Taken together, these studies identify the varied mechanisms of virus elimination that are activated within the infected retina. 

 Submitted for publication October 18, 1999; revised February 4 and April 14, 2000; accepted May 10, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Development of apoptotic cells within the eyes of BALB/c and CD-1 mice. Mean number of apoptotic cells per section observed in the eyes of (A) BALB/c mice and (B) CD-1 mice. The difference between the apoptotic events in virus-infected eyes compared with control eyes (untreated and mock-injected eyes) was statistically significant in both mouse strains (P < 0.01, student t-test). However, there was no significant difference in the development of apoptotic cells in the two strains of mice.

Figure 2.

 

Photomicrographs of apoptotic cells in the BALB/c mouse eyes detected by TUNEL assay. (A) Iris and ciliary body from normal eye and (B) JHM virus–infected eye (PI day 8); (C) retina from normal eye and (D) JHM virus–infected eye (PI day 8). Positive reaction is indicated by brown-stained nuclei (filled arrows) and apoptotic bodies (open arrows). Magnification, ×500.

Figure 3.

 

Electron micrograph showing the outer nuclear layer of infected BALB/c mouse retina (day 4 PI). (A) Three apoptotic cells and (B) another apoptotic cell. Magnification, (A) ×6,500; (B) ×10,000.

Figure 4.

 

Comparison of apoptotic cells in the vitreous and retina of (A) BALB/c and (B) CD-1 mice. The majority of apoptotic cells were observed in the retina in comparison with the vitreous. At days PI 6 through 10, apoptotic cells in the retina versus apoptotic cells in the vitreous from both strains of animals was P < 0.01, (Student t-test). The development of retinal apoptotic cells in CD-1 mice was temporally similar to the development of apoptotic cells in BALB/c mice.

Figure 5.

 

Comparison of infectious JHM virus from BALB/c and CD-1 mouse eyes (PI days 1 through 20). Infectious virus was detected only in the early phase of disease in both strains of infected mice.

Figure 6.

 

Laser confocal photomicrographs of frozen eye sections showing double-labeled immunofluorescent staining in infected retina of BALB/c mice. An apoptotic cell (arrowhead) was double labeled in the retina with (A) TUNEL-positive nucleus (green) and (B) with anti-CD8 antibody (red). (C) Overlaying images (orange–green) of (A) and (B) verified that an apoptotic cell was associated with a CD8⁺ T cell (PI day 7). Magnification, ×800.

Table 1.

 

Detection of FasL and Granzyme B mRNA in Isolated BALB/c and CD-1 Mouse Retinas by RT-PCR

	Treatment	Day (PI)					Total	Detection Rate (%)
				4	8	20
FasL
BALB/c	;l>None	0/3	0/4	0/1	0/8	0
	Mock	1/4	1/4	0/4	2/12	17
	Virus	2/4	4/4	2/4	8/12	67
CD-1	None	1/3	0/4	0/1	1/8	13
	Mock	0/4	0/4	0/4	0/12	0
	Virus	1/4	3/4	2/4	6/12	50
Granzyme B
BALB/c	None	0/3	0/4	0/1	0/8	0
	Mock	2/4	0/4	0/4	2/12	17
	Virus	3/4	4/4	1/4	8/12	67
CD-1	None	0/3	0/4	NT	0/7	0
	Mock	0/4	1/4	0/3	1/11	9
	Virus	2/4	2/4	2/4	6/12	50

 Data are number of detected mRNA/number of tested retina pools. In each of four experiments, eight retinas at each time point were isolated and pooled. After mRNA extraction and reverse transcription, FasL and granzyme B cDNAs were amplified. The PCR products were analyzed by southern hybridization with their special internal probes.

 NT, not tested.