Retinal and Choroidal Pathologies in Aged BALB/c Mice Following Systemic Neonatal Murine Cytomegalovirus Infection

Abstract

Although pathologies associated with acute virus infections have been extensively studied, the effects of long-term latent virus infections are less well understood. Human cytomegalovirus, which infects 50% to 80% of humans, is usually acquired during early life and persists in a latent state for the lifetime. The purpose of this study was to determine whether systemic murine cytomegalovirus (MCMV) infection acquired early in life disseminates to and becomes latent in the eye and if ocular MCMV can trigger in situ inflammation and occurrence of ocular pathology. This study found that neonatal infection of BALB/c mice with MCMV resulted in dissemination of virus to the eye, where it localized principally to choroidal endothelia and pericytes and less frequently to the retinal pigment epithelium (RPE) cells. MCMV underwent ocular latency, which was associated with expression of multiple virus genes and from which MCMV could be reactivated by immunosuppression. Latent ocular infection was associated with significant up-regulation of several inflammatory/angiogenic factors. Retinal and choroidal pathologies developed in a progressive manner, with deposits appearing at both basal and apical aspects of the RPE, RPE/choroidal atrophy, photoreceptor degeneration, and neovascularization. The pathologies induced by long-term ocular MCMV latency share features of previously described human ocular diseases, such as age-related macular degeneration.

Latent virus infections are characterized by the absence of the production of infectious virus particles, although viral nuclei remain within the host cell in an essentially dormant form. While in the latent state, the virus remains hidden from immune surveillance, allowing it to maintain a presence within its host for a prolonged period, sometimes for the entire lifespan of the individual. The absence of infectious virus distinguishes latent infection from chronic infection in which a generally low-level of ongoing production of virus occurs for an extended period. Latency may be interrupted and virus may be reactivated by a variety of factors generally related to changes in the cellular or tissue microenvironment. The long-term impact of latency on metabolism of host and surrounding cells is not well understood, and how latency impacts the host remains an open question.

Human cytomegalovirus (HCMV) is a ubiquitous β-herpesvirus, which infects 40% to 80% of humans¹ and persists for the life of the host through cycles of latency and reactivation following primary infection.² CMV latency can occur at multiple sites and cell types in the host, including endothelial cells and hematopoietic cells.³ HCMV is usually acquired during early life, with the incidence of congenital infection ranging from 0.5% to 2.4% of all live births4, 5, 6 and >12% of 1-year–old US infants infected with HCMV.⁷ Because the innate and adaptive immune systems are not fully mature during early life,⁸ acquiring HCMV infection during this period can lead to widespread virus dissemination throughout the body. This results in viral latency at a number of sites including the eye, which is one of the major target organs of congenital HCMV infections, with the incidence of HCMV chorioretinitis reported to be 25% in infants with symptomatic congenital HCMV infection.^4,9, 10, 11 Although only approximately 1% of infants who are asymptomatic and congenitally infected have CMV chorioretinitis,^4,9, 10, 11 CMV could spread to and become latent in the eye in significantly more asymptomatic infants without chorioretinitis. Indeed, recent studies of ocular tissue from human cadavers reveal that HCMV DNA is present in 4 of 24 choroid/retinal pigment epithelium (RPE) samples, suggesting that choroid/RPE might be a site of HCMV latency.¹²

Age-related macular degeneration (AMD) is a complex, multifactorial, progressive disease, which is the leading cause of irreversible visual dysfunction in older individuals.13, 14, 15 Because there is a significant association between elevated anti-HCMV IgG titers and neovascular AMD compared with controls with dry AMD or no AMD,¹⁶ HCMV infection might be a novel risk factor for the progression of AMD. The CMVs are species-specific, and murine CMV (MCMV) infection of mice is widely used to mimic human diseases. Previous studies have found that MCMV exacerbates the development of choroidal neovascularization (CNV) in both a laser-induced CNV model¹⁷ and a vascular endothelial growth factor–overexpressing model.¹² The studies show, for the first time, that latent ocular infection alone is associated with the development of retinal and choroidal pathologies. They share some features with AMD, including deposits at both basal and apical aspects of the RPE, degeneration of choriocapillaris, and RPE and photoreceptors in aged, MCMV-infected BALB/c mice following systemic neonatal infection.

Materials and Methods

Cells and Virus

MCMV strain K181 was originally provided by Dr. Edward Morcarski (Emory University, Atlanta, GA). Virus was prepared from the salivary glands of MCMV-infected immunosuppressed BALB/c mice, and virus stocks were titered on monolayers of mouse embryo fibroblast cells as described previously.¹⁸ A fresh aliquot of virus stocks was thawed and diluted to the appropriate concentration for each experiment.

Mice

Breeding pairs of BALB/c mice were purchased from Jackson Laboratory (Bar Harbor, ME). All mice were given unrestricted access to food and water and were maintained on a 12-hour light cycle alternating with a 12-hour dark cycle. Anesthesia protocols have been described previously.¹⁹ The breeding and treatment of animals in this study adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Institutional Animal Care and Use Committee of Augusta University. The rd8 mutation was excluded by genotyping.

Antibodies

Monoclonal antibody to an MCMV early gene product, pp56, was purified from the supernatant of hybridoma cell line 25G11 (a gift of Dr. John Shanley, Department of Medicine, University of Connecticut Health Center, Farmington, CT)²⁰ and labeled with fluorescein isothiocyanate (Sigma-Aldrich, St. Louis, MO) as previously described.²¹ Anti-RPE65 was kindly provided by Dr. Michael Redmond (National Eye Institute, National Institutes of Health, Bethesda, MD). Rat anti-mouse CD31 was purchased from Dianova (Hamburg, Germany). Alexa 594–labeled anti-rat IgG and Alexa 594–labeled anti-rat IgG were purchased from Vector Laboratories (Burlingame, CA). Alexa Fluor 594–labeled isolectin GS-IB4 was purchased from Thermo Fisher Scientific (Asheville, NC).

Experimental Design

A total of 50 plaque-forming units (PFU) of MCMV or culture medium as control were intraperitoneally injected into BALB/c mice at <3 days after birth. To determine whether acquiring CMV infection early in life is associated with widespread virus dissemination and latency at sites, including the eye, mice were euthanized at days 14 and 90 post infection (p.i.), and eyes and extraocular tissues, including salivary glands, lungs, and brain, were collected for analysis by plaque assay to quantify replicating virus. The presence of viral DNA was detected by PCR, whereas MCMV early antigen (EA) expression was analyzed by immunofluorescence staining or by immunogold electron microscopy. To determine the location of ocular latent infection, some eyes were removed at 3 months p.i., and posterior eyecups (consisting of RPE, choroid, and sclera) were separated and cultured as described below. The RPE cells and choroid were separated from additional posterior eyecups of mice latently infected with MCMV according to a protocol described previously²² and prepared for PCR to detect MCMV DNA. To determine whether latent ocular infection was associated with ocular pathologic changes, latently infected and control mice were anesthetized and spectral domain optical coherence tomography (SD-OCT) was performed at 4, 8, and 18 months p.i. as described below. Eyes were collected from latently infected BALB/c and control mice at 8 and 18 months p.i., and ocular ultrastructure was studied by electron microscopy. To determine whether latent ocular infection was accompanied by expression of MCMV genes and inflammatory/angiogenetic factors, eyes were collected from latently infected BALB/c and control mice at 8 and 18 months p.i., and expression of 19 MCMV latency-related genes as well as several inflammatory/angiogenetic factors was analyzed by real-time RT-PCR. Protein expression of IL-6 and chemokine (C-C motif) ligand 5 was measured by enzyme-linked immunosorbent assay (ELISA).

To trigger MCMV reactivation, latently infected mice were divided into two groups at 3 months p.i. when replicating virus is no longer recoverable from any ocular or nonocular site. Mice in group 1 and age-matched uninfected control mice were injected with methylprednisolone acetate, (2 mg per mouse intramuscularly every 3 days) and T-cell–specific antibodies [0.45 mg of anti-CD4 (GK1.5) and 0.1 mg of anti-CD8 (2.43) intravenously 1 and 7 days after beginning treatment with methylprednisolone). Mice in group 2 were injected with phosphate-buffered saline only. Mice were euthanized 2 weeks after initiation of immunosuppression, and blood was collected by cardiac puncture using EDTA as an anticoagulant before euthanizing. Peripheral blood leukocytes were separated from blood using ACK Lysing Buffer (Thermo Fisher Scientific) according to the manufacturer's instructions, and viral gene expression was analyzed by RT-PCR. Eyes, salivary glands, and lungs were collected for analysis by plaque assay, RT-PCR, and immunofluorescence staining.

Posterior Eyecup Culture

The method of posterior eyecup culture has been previously described by our laboratory.²³ Eyes were collected from mice latently infected with MCMV, and the posterior eyecup, consisting of sclera, choroid, and a monolayer of RPE, was isolated. The posterior eyecup was attached to a sterile membrane filter (Schleicher & Schuell, Dassel, Germany) with the sclera side in contact with the filter and mounted on a coverslip (Nalge Nunc international, Rochester, NY) with Matrigel (BD Biosciences, Bedford, MA). The coverslip with attached eyecup was inserted into a culture tube in 1 mL of culture medium (Dulbecco’s modified Eagle’s medium, 10% FBS) and cultured in a roller incubator at 37°C with a rotation rate of 10 to 15 rpm. Culture medium was collected after 1 day and twice weekly thereafter and examined by plaque assay for replicating virus. Cultures were also harvested and stained for MCMV EA as described below.

SD-OCT Examinations and Measurement of Retinal Thickness

Mice were anesthetized, and SD-OCT was performed using the Bioptigen Spectral-Domain Ophthalmic Imaging System (Envisu R2200; Bioptigen, Morrisville, NC) as described previously.^12,19 Briefly, pupils were dilated with 1% tropicamide, and Systane Ultra lubricant eye drops were applied liberally to keep the eye moist during imaging. Images, including averaged single B scan and volume intensity scans, were acquired. A total of three scanning images were acquired from the center of the optic nerve head, the left sphere of the eye (from optic nerve head to left iris), and the right sphere of the eye (from optic nerve to right iris) in each mouse. The highly reflective CNV lesions located above the RPE layer were quantitated followed by measurement of total retinal thickness in the scanning image taken from the center of the optic nerve head by using InVivoVue Diver version 2.4 software (Bioptigen).

Immunofluorescence Staining

Eyes were embedded in OCT compound, frozen, and sectioned in a cryostat. Sections were then fixed with 4% paraformaldehyde for 15 minutes and stained for CD31, isolectin, F4/80. Iba-1, MCMV EA, and RPE65 as described previously.^12,21,24

Electron Microscopy and Immunogold Staining

As previously described by our laboratory,^23,25 eyes of MCMV-infected and control mice were fixed (4% paraformaldehyde, 0.5% glutaraldehyde in 0.1 mol/L cacodylate buffer) overnight at 4°C. After washing, dehydrating, and embedding, ultrathin sections were cut and stained with uranyl acetate and lead citrate and visualized in a JEM 1230 transmission electron microscope (JEOL USA Inc., Peabody, MA) at 110 kV and imaged with an UltraScan 4000 CCD camera & First Light Digital Camera Controller (Gatan Inc., Pleasanton, CA). For immunogold staining,²³ additional ultrathin sections were cut and collected on 200 mesh nickel grids. Following etching with 2% H₂O₂ for 20 minutes and treatment with 1 mol/L ammonium chloride for 1 hour, sections were blocked with 0.1% BSA in phosphate-buffered saline for 2 to 4 hours and floated in anti-MCMV EA antibody overnight at 4°C. After washing, anti-mouse IgG nanogold was added for 2 hours at room temperature, and nanogold particles were enhanced for 3 to 8 minutes with silver enhancement solution (HQ Silver Enhancement kit, Nanoprobes, Yaphank, NY). Following washing, grids were stained with 2% uranyl acetate in 70% ethanol and lead citrate and visualized in a JEM 1230 transmission electron microscope.

Nucleic Acid Purification

DNA was extracted from eyes, salivary glands, lungs, and peripheral blood leukocytes by overnight digestion with proteinase K at 56°C with continuous vigorous mixing, followed by centrifugation. The supernatant was removed and DNA precipitated with an equal volume of isopropanol before resuspension in water. Genomic DNA was diluted to 50 ng/μL before use. Total RNA was extracted from eyes, salivary glands, lungs, and peripheral blood leukocytes using the RNeasy mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Reverse transcription of mRNA was performed on 500 ng of total RNA using the genomic DNA Clear cDNA Synthesis Kit (Bio-Rad, Hercules, CA) according to the manufactures' instructions. DNA and cDNA were stored at −80°C until use.

PCR, RT-PCR, and Real-Time PCR

All primer sequences used for PCR are listed in Table 1. For genomic DNA, PCR, and RT-PCR, MCMV genes were amplified in 25 μL of reaction consisting of 0.1 μL of Taq DNA polymerase (Sigma-Aldrich), 0.2 μL of 20 pmol/μL primer mixture, 12.5 μL of a 2× PCR supermix (Thermo Fisher Scientific), and 50 ng of DNA or 1 μL of cDNA prepared by reverse transcription of mRNA using 1 μg of total RNA. Reaction conditions were as follows: 5 minutes at 95°C followed by 40 cycles of 95°C for 10 seconds, 60°C for 20 seconds, and 72°C for 30 seconds followed by a final elongation of 5 minutes at 72°C. Amplification products were separated on 1.5% agarose gels and stained with ethidium bromide and visualized with the ChemiDoc Imaging System (Bio-Rad). For real-time PCR, inflammatory cytokine genes were amplified in a 20-μL reaction consisting of 10 μL of 2× SYBR Mix (Bio-Rad), 0.2 μL of 20 pmol/μL primer mixture, and 1 μL cDNA, using the CFX96 Real Time PCR System (Bio-Rad). PCR conditions were as follows: 3 minutes at 94°C followed by 40 cycles of 94°C for 10 seconds, 60°C for 20 seconds, and 72°C for 30 seconds. All Ct values were analyzed and normalized to β-actin using the method of 2^−ΔΔCT.

Table 1.

Specific Primers for Murine Cytomegalovirus Genes, rd8, and Inflammatory/Angiogenic Genes

Gene	Forward primer (5'->3′)	Reverse primer (5'->3′)	Gene description or function
m123 (IE1)	5′-GTTACACCAAGCCTTTCCTGGAT-3′	5′-TGTGTGGATACGCTCTCACTCTCTAT-3′	Immediate early gene
m122 (IE3)	5′-TGTGAGGCAGTAGTTATACC-3′	5′-CCTCGAGTCTGGAACCGAAA-3′
m18	5′-ACGAGGTAGGAGCAGGTGAT-3′	5′-TTCAACCGCCCGATCTCAAA-3′	Early gene
m135	5′-CACCCCGTTGTTCTGTACCA-3′	5′-CGCTATATTGGTGGTGGCGA-3′
m136	5′-CGCATCTCTCCGTCGTGAAT-3′	5′-ACATCGAGTGGTGCTTTCGT-3′
m37	5′-ACCGTCTCGGTGTGATTTCC-3′	5′-AGCTCAAGACGACGATGGAC-3′	Anti–programmed cell death
m36	5′-CAACAGCCCTCCATCATCGT-3′	5′-TGCAGGTATCGCGCATAGAC-3′
m41	5′-TCGACGGGAAACAGTTCGAG-3′	5′-AGACCTAATCTCCGTCGCCT-3′
m45	5′-GATGTGCAGGTCGCGATAGA-3′	5′-TCAACCTGGAGAACTGCGTC-3′
m38.5	5′-CTCTCCTGATGTCCCGCAAG-3′	5′-CAGTCAGAACTGCTCGGGAA-3′
m53	5′-GAGACATCCCGACCTCGAAC-3′	5′-GCGGTGCTTCATCACGTTAG-3′	DNA replication related
m54	5′-GTGAAGAGGTGGTTCTCGGG-3′	5′-TCGAAGAGCAGAGCAACTGG-3′
m72	5′-TCAGGTTGATGGTCTGCCAC-3′	5′-GTGATGGACATGGCGGGTTA-3′
m04	5′-TTCGTGGACCTTGGAATGGG-3′	5′-AGATTACCACCGTTGCCGTT-3′	Immune evasion or modulation
m138	5′-CGTTGACGTAGTAGACCCCG-3′	5′-GTGACGGCGCATCAATTACC-3′
m35	5′-TACGTGCGCCACATCTACTC-3′	5′-GAGTACATCGTCACGGCCTC-3′
m83	5′-TTTGGAGTCCGGATCGTTGG-3′	5′-AGTACCTGGGTTTCTGCGTG-3′
m80	5′-GATCCCTTCGGTTCGGACTC-3′	5′-GCGTCTAATACTCTGCGGCT-3′	Assembly protein, protease
m55 (GB)	5′-AGGGCTTGGAGAGGACCTACA-3′	5′-GCCCGTCGGCAGTCTAGTC-3′	Envelope glycoprotein B
rd8	F1:5′-GTGAAGACAGCTACAGTTCTGATC-3′ F2:5′-GCCCCTGTTTGCATGGAGGAAACTTGGAAGACAGCTACAGTTCTTCTG-3′	5′-ATCTGGTGCTCCTCAGATCAGCTAA-3′	Retinal degeneration 8
TGFB1	5′-GCTGCGCTTGCAGAGATTAAA-3′	5′-TTGCTGTACTGTGTGTCCAG-3′	Transforming growth factor β1
Ccl5	5′-GTGCTCCAATCTTGCAGTCG-3′	5′-AGAGCAAGCAATGACAGGGA-3′	Chemokine (C-C motif) ligand 5
Angpt1	5′-GGGGGAGGTTGGACAGTAA-3′	5′-CATCAGCTCAATCCTCAGC-3′	Angiopoietin 1
IL6	5′-GTCCTTCCTACCCCAATTTCCA-3′	5′-TAACGCACTAGGTTTGCCGA-3′	IL-6
Vegfa	5′-CACAGCAGATGTGAATGCAG-3′	5′-TTTACACGTCTGCGGATCTT-3′	Vascular endothelial growth factor a
ICAM1	5′-GTGATCCCTGGGCCTGGTG-3′	5′-GGAAACGAATACACGGTGATGG-3′	Intercellular adhesion molecule 1
Mcsf	5′-TGATTGGGAATGGACACCTG-3′	5′-AAAGGCAATCTGGCATGAAGT-3′	Macrophage colony-stimulating factor
Angtp2	5′-GATCTTCCTCCAGCCCCTAC-3′	5′-TTTGTGCTGCTGTCTGGTTC-3′	Angiopoietin 2
Pdgfb	5′-ACTCCATCCGCTCCTTTGAT-3′	5′-GTCTTGCACTCGGCGATTA-3′	Platelet-derived growth factor beta chain
Ccl7	5′-AAGTGGGTCGAGGAGGCTAT-3′	5′-CCATTCCTTAGGCGTGACCA-3′	Chemokine (C-C motif) ligand 7
Sdf1	5′CAGAGCCAACGTCAAGCA-3′	5′AGGTACTCTTGGATCCAC-3′	Stromal cell–derived factor 1

ELISA

Eyes were collected and homogenized in NP40 lysis buffer (Thermo Fisher Scientific) with 1 mmol/L phenylmethylsulfonyl fluoride and protease inhibitor cocktail (Roche, Mannheim, Germany). Protein levels of CCL5 and IL-6 in the tissue lysate were measured by using an ELISA assay kit (R&D Systems, Minneapolis, MN) according to the manufacturer's instructions.

Statistical Analysis

All data are expressed as means ± SEM, with n representing the number of mice used in each of the experimental groups. Statistical analyses were used to determine the significance of observed differences between treatment groups in all experiments. Statistical significance was calculated by means of a two-tailed, unpaired, noparametric, Mann-Whitney test using GraphPad Prism 8 software (GraphPad Software, San Diego, CA). P < 0.05 was considered to be significant.

Results

MCMV Disseminates to the Eye Following Systemic Infection of Neonatal Mice

A total of 50 PFU of MCMV was injected intraperitoneally into neonatal BALB/c mice at <3 days after birth. A total of 198 of the 305 mice survived viral infection and appeared healthy. Body weight was lower at 2 to 4 weeks of age in mice infected as newborns compared with uninfected controls of the same age. However, this weight difference diminished at later time points. At day 14 p.i., replicating virus was detected in eyes and extraocular tissues, including salivary glands and lungs, of all mice (Figure 1A). Most MCMV-infected cells were located in the choroid (Figure 1B and Table 1), whereas a few MCMV infected cells were also observed in the RPE layer (Figure 1B). In addition, virus-infected cells were also occasionally observed in anterior segments, including the iris and ciliary body (Table 2). No MCMV EA–positive cells were observed in the inner retina (Figure 1B). To investigate more closely the localization of MCMV in the eye, infected ocular tissue was examined using electron microscopy following immunogold staining with anti-MCMV EA. These experiments showed that MCMV EA was present in the nuclei of some vascular endothelial cells (Figure 1C) and pericytes (Figure 1D) in the choriocapillaris and also in sporadic RPE cells (Figure 1E).

Figure 1.

Systemic neonatal murine cytomegalovirus (MCMV) infection disseminates to the eye. A: Titer of MCMV at day 14 post-infection (p.i.) in eyes, salivary glands, and lungs of BALB/c mice. B: Merged photomicrographs of ocular staining for retinal pigment epithelium (RPE) 65 (red), MCMV early antigen (EA) (green), and DAPI (blue) at 14 days p.i following intraperitoneal infection with MCMV at <3 days after birth. Most MCMV EA–positive cells were located in the choroid. A few MCMV-infected cells were also observed in the RPE layer (arrow). Some RPE65-, MCMV EA–infiltrating cells were observed in the photoreceptor layer (arrowheads). C–F: Immunogold staining with anti–MCMV EA showing immunogold-labeled MCMV EA in nuclei of some vascular endothelial cells (C, arrows) and pericytes (D, arrow) in the choriocapillaris and in sporadic RPE cells (E, arrow). F: Negative control stained with secondary antibody alone. G–I: Compared with eyes from uninfected control mice (G), the outer blood retinal barrier and Bruch's membrane had no remarkable disorganization or disruption, whereas most RPE cells appeared indistinguishable from those of infected mice (I). However, large vesicles were noted in occasional RPE cells (I, circle), whereas a few infiltrating cells (I, arrow) and increased phagocytosis of outer segments (I, arrowhead) were observed in the photoreceptor layer. No remarkable pathologic changes were found in the inner retina of MCMV-infected mice (H). Scale bar = 2 μm. INL, inner nuclear layer; IS, inner segments; ONL, outer nuclear layer; OS, outer segments.

Table 2.

Frequency of Murine Cytomegalovirus Early Antigen Detection in Various Ocular Locations

Location	Acute infection	Reactivation
Choroid	7/7	6/6
Retinal pigment epithelium	3/7	2/6
Sclera	5/7	6/6
Iris/ciliary body	1/7	6/6

Electron microscopical analysis was used to determine whether acute MCMV infection in the choroid and RPE was associated with a compromised blood-retina barrier and/or pathological changes in the inner retina. As shown in Figure 1I, the outer blood-retina barrier appeared to be intact, and Bruch's membrane had no remarkable disorganization or disruption when viewed through electron microscope. Although the majority of RPE cells in acute infected mice were indistinguishable from RPE cells in uninfected control mice (Figure 1G), large vesicles were noted in occasional RPE cells (Figure 1I), whereas increased phagocytosis of outer segments was also observed (Figure 1I). Neither cell death nor any remarkable infiltration was found in the inner retina of MCMV-infected mice (Figure 1H), although a few infiltrating cells were observed in the photoreceptor layer (Figure 1I).

Ocular MCMV Infection Becomes Latent in Choroid and RPE

To determine whether MCMV infection undergoes latency in the eye and extraocular organs/tissues following systemic infection, MCMV-infected mice were sacrificed at 3 months p.i. and tissue samples were collected. No replication-competent MCMV could be recovered from eyes, lungs, or salivary glands at this time, whereas MCMV DNA was detected in nine out of nine eyes (Table 3) and eight out of eight lungs. Because MCMV-infected cells were observed primarily in the choroid (Figure 1B), posterior eyecup cultures were used to determine whether latent ocular MCMV could be reactivated in vitro. Eyes were collected at 4 months p.i., and posterior eye cup, consisting of sclera, choroid, and RPE, were separated and cultured at 37°C. Culture medium was collected biweekly and assayed by plaque assay for replicating virus. Posterior eyecup cultures started producing replicating virus beginning at day 7 of culture (2 of 12 positive), whereas at day 14 of culture. 8 of 12 eyecup cultures produced replicating virus. Following 3 weeks in culture, the virus was detected in 11 of 12 samples. To identify the location of the reactivated virus in vitro, eyecup cultures were recovered and stained for MCMV EA antigen following 2 weeks in culture. The findings indicate that MCMV EA was present in sclera and choroid and co-localized with some RPE65-positive RPE cells (Figure 2A). In addition, RPE cells and choroid were separated from posterior eyecups of mice latently infected with MCMV according to a protocol described previously²² and PCR was performed to detect MCMV DNA. These experiments indicate that MCMV DNA was present in both RPE cells (four of four mice) and choroid (four of four mice) (Table 3).

Table 3.

Frequency of MCMV DNA and Plaque Assay in Ocular Tissue from Latently Infected Mice

Tissue	MCMV DNA	Plaque assay
RPE cells from eyes of infected mice at 4 months p.i.	4/4	N/A
Choroids from eyes of infected mice at 4 months p.i.	4/4	NA
Eyes of infected mice at 3 months p.i.	9/9	0/4
Eyes of infected mice at 8 months p.i.	6/6	0/4
Eyes of infected mice at 18 months p.i.	9/9	0/4

MCMV, murine cytomegalovirus; NA, not applicable; p.i., post infection.

Figure 2.

Ocular latent murine cytomegalovirus (MCMV) can be reactivated in vitro and in vivo following systemic immunosuppression. A: Merged photomicrographs of ocular staining for retinal pigment epithelium (RPE) 65 (red), MCMV early antigen (EA) (green), and DAPI (blue) in eyecup cultures prepared from BALB/c mice at 4 months post intraperitoneal neonatal infection. MCMV EA was observed in choroid and RPE (arrows). B: Titer of MCMV in eyes, salivary glands, and lungs of MCMV latently infected BALB/c mice following systemic immunosuppression for 2 weeks. C: Merged photomicrographs of ocular staining for RPE65 (red), MCMV EA (green), and DAPI (blue) in eyes of mice latently infected with MPMV following systemic immunosuppression for 2 weeks. Most MCMV EA-staining cells were observed in the choroid (arrows). D: Merged photomicrographs of ocular staining for MCMV EA (green) and DAPI (blue) in eyes of mice latently infected with MCMV following systemic immunosuppression for 2 weeks. MCMV EA staining cells were observed in the iris (arrows).

Virus Reactivation in Latently Infected Eyes Following Systemic Immunosuppression

Latent virus can be reactivated from infected eyes by deep systemic immunosuppression several months after MCMV intraocular inoculation.²⁴ To determine whether latent MCMV could be reactivated in eyes of neonatally infected mice by immunosuppression in vivo, newborn BALB/c mice were inoculated intraperitoneally with 50 PFU of MCMV. Four months later, some were deeply immunosuppressed with methylprednisolone plus anti–T-cell antibodies, which depleted >99% of CD4 and CD8 T cells, as previously described.^24,26 After 2 weeks, the eyes were collected and analyzed by plaque assay. No replicating virus was present in the eyes or extraocular tissues, including salivary glands and lungs, of any latently infected immunocompetent mice. In contrast, replicating MCMV was recovered from 6 of 8 eyes of immunosuppressed mice and from most extraocular tissues (Figure 2B). Although significantly more replicating virus was recovered from lungs and salivary glands compared with eyes during acute infection (Figure 1A), a surprisingly similar amount of replicating virus was recovered from eyes, salivary glands, and lungs during reactivation by immunosuppression (Figure 2B).

To assay for the presence of virus in leukocytes, blood was collected by cardiac puncture from two groups of mice, and DNA and total RNA were isolated from peripheral blood leukocytes. Expression of the MCMV IE1 gene was analyzed by real time RT-PCR, whereas PCR was used to test for the presence of virus DNA. Although no MCMV DNA was detectable in the blood of nonimmunosuppressed, latently infected mice (0 of 6 mice), MCMV DNA was detected in leukocytes in 5 of 8 immunosuppressed mice. No MCMV IE1 transcripts were detectable in any blood samples from immunosuppressed or nonimmunosuppressed latently infected mice.

To identify the ocular location of MCMV reactivation, antibody staining was performed with both anti–MCMV EA and anti-RPE65. As previously observed during acute infection (Table 2), MCMV-infected cells were present in the choroid of all eyes (6 of 6 mice) and in RPE cells of some eyes (3 of 6 mice) (Table 2 and Figure 2C). No virus was detected in the inner retina, although virus-infected cells were observed in anterior segments, including the iris and ciliary bodies of 6 of 6 mice (Figure 2D), compared with only 1 of 7 mice during acute infection (Table 2). This finding suggests that reactivated MCMV might spread from the choroid to the ciliary body and iris via the uveal tract.

Retinal and Choroidal Pathologies Detected by SD-OCT

SD-OCT is a noninvasive imaging technique that provides high-resolution, cross-sectional images of the retinal microstructure in vivo.^27,28 SD-OCT imaging was performed with the Envisu R2210 system of Leica Microsystems (Wetzlar, Germany), and retinal thickness was measured in the eyes of mice latently infected with MCMV and age-matched controls at 4, 8, and 18 months p.i. (Figure 3A). Mean retinal thickness was significantly reduced in the eyes of latently infected mice compared with age-matched control eyes (Figure 3B). In control mice, mean retinal thickness deteriorated with age and was lower in the eyes of control BALB/c mice at 18 months of age compared with control mice at 8 months of age. However, apart from changes in retinal thickness, no other remarkable pathologic changes were detected by SD-OCT in the eyes of mice latently infected with MCMV at 4 or 8 months p.i., or in control eyes at all ages. In contrast, besides significantly lower retinal thickness in all 40 eyes examined (Figure 3A), severe photoreceptor degeneration, including disappearance of the entire outer nuclear layer (ONL) in some areas, was observed in 21 eyes of mice latently infected mice at 18 months p.i. Other pathological changes observed in these 21 eyes included CNV-like lesions (mean of two per eye) (Figure 3C) and retinal detachment (not shown), which were noted in six and four eyes, respectively. Eyes with CNV-like lesions were removed, sectioned, and stained with anti-CD31 and isolectin. The results confirmed the presence of vascular endothelial cells in CNV-like lesions (Figure 3, D and E).

Figure 3.

Spectral domain optical coherence tomography (SD-OCT) and choroidal neovascularization (CNV) lesions. A: Representative images of SD-OCT in eyes of latently infected mice at 4, 8, and 18 months post infection (p.i.) together with age-matched control eyes. Green line indicates positions of the SD-OCT scans. B: Retinal thickness by SD-OCT. C: CNV-like lesions detected by SD-OCT (circles). D: Merged photomicrograph of ocular staining for CD31 (green) and DAPI (blue) at 18 months p.i. CD31⁺ vascular endothelial cells were observed in a CNV-like lesion (circle). E: Merged photomicrograph of ocular staining for isolectin (red) and DAPI (blue) at 18 months p.i. Isolectin and vascular endothelial cells were observed in a CNV-like lesion (circle). ∗P < 0.05, ∗∗P < 0.0001 (U-test). INL, Inner nuclear layer.

Retinal and Choroidal Pathologies Observed by Electron Microscopy

To more precisely define the types of pathologic changes that are associated with latent MCMV infection of the choroid and RPE, eyes were collected from latently infected BALB/c mice at 8 and 18 months p.i. and from age-matched, uninfected controls, and processed for examination by electron microscopy. This process revealed the presence of various pathologies in all eight eyes from mice latently infected with MCMV at 8 (three eyes) and 18 (five eyes) months p.i.

Specifically, at 8 months p.i., the lumen of choroidal capillaries of infected mice was typically narrowed (Figure 4B) compared with that of age-matched control eyes (Figure 4A). Likewise, the capillary wall was thickened, and some endothelial cells appeared hypertrophic in infected mice. In some areas, increased numbers of cells (Figure 4B) and collagen fibers (Figure 4B) were observed in the choroid, with the lumens of some choroidal capillaries mechanically compressed against Bruch's membrane, possibly because of the increased number of cells and fibers. Capillary lumens were narrowed to the point of collapse in some instances (Figure 4B). Cell death was also noted in some areas of the choroid, including in capillary endothelia (Figure 4B). Many large lipid vesicles were present inside some RPE cells (Figure 4D), whereas only occasional lipid vesicles were noted in RPE cells of control eyes (Figure 4A). In some areas, basal lamina deposits were present between the plasma membrane and basal lamina of the RPE (Figure 4, C, F and G) in three of three eyes from mice latently infected with MCMV. As indicated in Figure 4C, disruption of tight junctions occurred between some RPE cells located below basal lamina deposits with evacuated areas observed between these RPE cells. In addition, some RPE cells appeared to exhibit an atypical, marked vacuolization (Figure 4, B and E) with swollen mitochondria and loss of mitochondrial cristae (Figure 4B). Large vesicular structures were observed in the subretinal space adjacent to atypical RPE cells, which were separated by an intact septum or membrane (Figure 4E) and which contained heterogeneous material, including dark-staining fragments (Figure 4E). The morphology of nearby photoreceptors was disturbed, with shortening and loss of outer segments (Figure 4E). Although the identity of these vesicular structures is uncertain, they may be related to phagolysosomes or contain subretinal deposits. A few infiltrating cells were observed in the subretinal space of infected (Figure 4, G and I) and control mice (not shown) at 8 months p.i., whereas dark-staining outer segment–like fragments were observed inside some infiltrating cells of infected mice. As shown in Figure 4H, discs were occasionally noted in these fragments.

Figure 4.

Representative electron micrographs of ocular ultrastructure in the 8 months post neonatal intraperitoneally infected or age-matched uninfected control BALB/c mice. A: Only occasional lipid vesicles were noted in retinal pigment epithelium (RPE) cells of control mice. B: Thicker wall, hypertrophic endothelia, and narrowed lumens of choroidal capillaries (B, asterisks) compared with the control eye (A, asterisks). More cells and collagen fibers (white arrows) were observed in the choroid. Cell death was also noted in some areas of the choroid, including capillary endothelia (black arrows). RPE cells exhibited marked vacuolization and swollen mitochondria with loss of the cristae (arrowheads). C: Loss of tight junctions between RPE cells located below one large diffuse basal lamina deposits with evacuated areas created between these RPE cells (arrows). D: Large lipid vesicles inside RPE cells (arrows). E: One large structure (asterisks) that contained heterogeneous materials, including dark-staining fragments (arrowheads) in the subretinal space nearing these atypical RPE cells; nearby photoreceptors exhibited shortening and loss of outer segments (OSs) (arrows). F: One diffuse lightly stained basal lamina deposit. G–I: Some infiltrating cells were observed in the subretinal space of infected mice (arrows). Dark-staining OS-like fragments were also observed inside some of these cells (G and I), and discs were occasionally noted among these fragments (H). Scale bar = 5 μm. BM, Bruch's membrane; IS, inner segment.

At 18 months p.i., five eyes from infected mice, including one eye with severe photoreceptor degeneration and CNV-like lesions identified by SD-OCT, were analyzed by electron microscopy, together with age-matched controls. Large lipid vesicles were present in some RPE cells of control mice (not shown), whereas a few infiltrating cells were observed in the subretinal space, with loss or shortening of outer segments (Figure 5A). However, no other pathologies, such as basal lamina deposits, were observed in any of the three control eyes examined. The following retinal and choroidal pathologies were observed in the eyes of MCMV-infected mice at 18 months p.i.:

• i)Loss of choroidal capillaries. Extensive choroidal platelet infiltration both within choroidal capillaries and in perivascular tissue (Figure 5F) was observed. One particularly striking abnormality was the presence of relatively large arterioles in some areas of the choroid and lack of choroidal capillaries. An example is shown in Figure 5E, with a large arteriole located between the sclera and Bruch's membrane. In addition, lumens of choroidal capillaries were also narrowed, and hypertrophic endothelia were noted (Figure 5, D, E and G) compared with age-matched control eyes (Figure 5A).
• ii)Infiltrating cells were sometimes observed in the subretina, and dark-staining outer segment–like fragments were observed inside some of these cells (Figure 5C).
• iii)Deposits at the basal and apical aspects of the RPE. As shown in Figure 5G and Table 4, basal lamina deposits were observed in four of five eyes examined. Some subretinal deposits (Figure 5, B, D, and G) and some outer segment fragments (Figure 5D) were observed on the apical side of RPE. The morphology of nearby photoreceptors was disturbed, with shortening and loss of outer segments (Figure 5D), whereas outer segments were completely absent in some areas (Figure 5D). Some subretinal deposits exhibited a finely granular texture (Figure 5, B and D), a feature described in human subretinal drusenoid deposits,^29,30 which have recently been recognized as another type of extracellular lesion located in the subretinal space of AMD eyes.29, 30, 31, 32, 33
• iv)Degeneration of choroidal endothelia, RPE, and photoreceptors. A thinned, disrupted RPE structure was observed adjacent to areas of subretinal deposits (Figure 5B), and some RPE cells exhibited marked vacuolization (Figure 6, B and D). Swollen mitochondria with loss of cristae were also noted in these RPE cells (Figure 6B) and in the choroid (Figure 6A). Some photoreceptors appeared apoptotic, and exhibited nuclear shrinkage and strong chromatin condensation (Figure 6D). In addition, a few ectopic photoreceptor nuclei were noted in the inner segment and outer segment layers (Figure 6C).
• v)Presence of CNV-like lesions and severe retinal degeneration. As shown in Figure 6, F and G, new blood vessels were noted in the sub-RPE space. Bruch's membrane appeared to extend into the sub-RPE space adjacent to the new vessels (Figure 6G), suggesting a choroidal rather than a retinal origin. The RPE was located adjacent to the inner nuclear layer (INL) because of the loss of the entire ONL (Figure 6F). Severe retinal degeneration was not confined to areas of CNV but was also observed in non-CNV areas of the retina (Figure 6E). In this region, the ONL was absent, whereas the INL was reduced in size and composed of only three to four layers of cells.

Figure 5.

Representative electron micrographs of ocular ultrastructure in 18 months post neonatal intraperitoneally infected or age-matched uninfected control BALB/c mice. A: A few large lipid vesicles were present in some retinal pigment epithelium (RPE) cells of control eyes. A few infiltrating cells were also observed in the subretinal space of control eyes (arrow). Loss or shortening of the outer segment (OS) was noted surrounding the infiltrating cells. Asterisks indicate choroidal capillaries. B–G: Thinned, disrupted RPE structure (B). Choroidal capillaries with thicker wall and smaller lumens (D, E, G; asterisks). Lack of choroidal capillaries in some areas of choroid; one arteriole is located between the sclera and Bruch’s membrane (BM) (E). Basal lamina deposits were observed (G). Some subretinal deposits (SDs) with a finely granular texture (B, D, and G) and some OS fragments (D; arrows) were observed on the apical of RPE. The morphology of nearby photoreceptors was disturbed, with shortening and loss of OS (D, arrowheads); OSs were completely absent in some areas (D, star). Extensive choroidal platelet infiltration in perivascular tissue (F, circle). Infiltrating cells were observed in the subretina with dark-staining fragments (C, arrow). Scale bar = 5 μm. IS, inner segment; ONL, outer nuclear layer.

Table 4.

Frequency of Retinal and Choroidal Pathologies in Latently Infected Mice

Pathology	8 Months old	18 Months old
Total ONL loss in some areas	0/22	21/40
Retinal detachment	0/22	4/40
CNV-like lesions	0/22	6/40
Choriocapillaris degeneration	3/3	5/5
RPE degeneration	3/3	5/5
Basal lamina deposits	3/3	4/5
Subretinal deposit	0/3	5/5

CNV, choroidal neovascularization; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

Figure 6.

Representative electron micrographs of ocular ultrastructure in 18 months post neonatal intraperitoneally infected mice. A: Swollen mitochondria in the choroid with loss of cristae (arrows). B: An ectopic photoreceptor nucleus was noted in the outer segment (OS) layer (arrow). C: Retinal pigment epithelium (RPE) exhibited marked vacuolization and swollen mitochondria with loss of the cristae (arrows). D: Absence of entire OS, inner segment (IS), and death of photoreceptor cells (arrow), whereas RPE exhibited marked vacuolization. E: Severe retinal degeneration, with complete loss of the outer nuclear layer (ONL), whereas the inner nuclear layer (INL) was diminished with only three to four layers of cells. F and G: A choroidal neovascularization (CNV)–like lesion in the sub-RPE space and RPE cells adjacent to INL attributable to the absence of the entire ONL. Dark staining Bruch’s membrane (BM) extended along new vessels into the sub-RPE space (G, arrows). Scale bar = 2 μm.

Expression of MCMV Genes

Cytomegalovirus latent infection can result in the expression of a number of virus-encoded proteins with the potential to significantly alter homeostasis of both latently-infected cells and the surrounding cellular environment.³⁴ Thus, expression of 19 genes was analyzed by real-time RT-PCR using RNA isolated from eyes of latently infected BALB/c mice at 8 and 18 months p.i. As indicated in Table 5, expression of the MCMV genes IE1 and IE3 and other virus genes that function in either an antiapoptotic capacity (m36, m37, m38.5, m45) or immune modulation (m04, m138) was detected in eyes of latently infected mice at 8 and 18 months p.i. Expression of the m18 and m80 virus genes was detected only at 18 months p.i., whereas expression of the m83 and m54 virus genes was not detected at 8 or 18 months p.i. No replicating virus was detected by plaque assay in any eyes of latently infected mice at 8 or 18 months p.i.

Table 5.

Frequency of Expression of the Murine Cytomegalovirus Genes in the Eyes of BALB/C Mice at 8 and 18 months p.i. Analyzed by Real-Time RT-PCR

Gene	Function	8 Months p.i.	18 Months p.i.
m123 (IE1)	Immediate early	4/6	9/11
m122 (IE3)	Immediate early	4/6	9/11
m18	Early	0/6	3/7
m135	Early	4/6	2/7
m136	Early	1/6	4/7
m37	Anti–cell death	5/6	5/7
m36	Anti–cell death	4/6	0/7
m41	Anti–cell death	3/6	2/7
m45	Anti–cell death	3/6	5/7
m38.5	Anti–cell death	3/6	5/7
m53	DNA replication	1/6	3/6
m54	DNA replication	0/6	0/6
m72	DNA replication	3/6	3/6
m80	Assembly protein, protease	0/6	3/6
m04	Immune modulating	2/6	3/6
m138	Immune modulating	3/6	4/7
m35	Immune modulating	0/6	1/6
m83	Virion associated	0/6	0/6
m55 (GB)	Envelope glycoprotein B	1/6	2/11

p.i., post infection.

Expression of Host Inflammatory/Angiogenic Factors

CMVs induce angiogenesis via the production of various angiogenic factors,³⁵ including cytokines/chemokines, such as IL-6^36,37 and CCL5,³⁸ and growth factors, such as angiopoietin,³⁹ vascular endothelial growth factor,⁴⁰ transforming growth factor β,³⁹ and macrophage colony-stimulating factor.³⁹ Real time RT-PCR was used to test for the expression of these genes in latently infected and control eyes. Transcript levels of some cytokines and growth factors, particularly CCL5, were highly up-regulated in latently infected eyes compared with age-matched control eyes at both time points (Figure 7A). At 8 months p.i., levels of CCL5 gene transcripts were elevated, although mRNA levels of other inflammatory factors and growth factors were similar to those of controls. At 18 months p.i., IL6 and CCL7 transcripts were 27-fold and fourfold higher, respectively, compared with controls, in addition to persisting elevated CCL5 transcripts.

To determine whether elevated transcript levels were associated with increased protein expression, relative protein levels of CCL5 and IL6 were measured by ELISA. At both 8 and 18 months p.i., protein levels of CCL5 were significantly elevated in the eyes of latently infected mice compared with age-matched control eyes (8 months p.i.: 7.787 ± 2.404 pg/mg protein compared with 1.110 ± 0.064, n = 3, P < 0.05; 18 months p.i: 49.030 ± 8.354 compared with 14.130 ± 1.913, n = 6, P < 0.01 by the Mann-Whitney test). IL-6 protein levels were also significantly elevated in 18-month–old latently infected eyes compared with age-matched controls (10.060 ± 2.705 compared with 4.408 ± 0.997, n = 6, P < 0.01 by the Mann-Whitney test). Interestingly, protein levels of CCL5 were also significantly elevated in 18-month compared with 8-month control eyes (P < 0.001 by the Mann-Whitney test).

Mononuclear Phagocytes

Protracted accumulation of mononuclear phagocytes, including macrophages, microglia, and monocytes in the subretinal space and photoreceptor cell layer, are a common feature observed in human AMD.41, 42, 43 To determine whether MCMV ocular latency is associated with infiltration of mononuclear phagocytes, eye sections from latently infected BALB/c mice at 4, 8, and 18 months p.i. as well as from age-matched, uninfected controls were stained for Iba1 or F4/80. F4/80-positive mononuclear phagocytes were observed in the subretinal space and ONL in a progressive manner starting at 4 months p.i. (Figure 7B).

Figure 7.

Host inflammatory/angiogenic factors and mononuclear phagocytes. A: Relative expression levels of inflammatory/angiogenic genes in eyes of murine cytomegalovirus (MCMV)–infected mice at 8 and 18 months post infection (p.i.) compared with age-matched control eyes. B: Representative merged photomicrographs of staining for F4/80 (red) and DAPI (blue) in eyes prepared from BALB/c mice at 4, 8, and 18 months post intraperitoneal neonatal infection and from 18 months in uninfected control mice. Merged photomicrographs of staining for secondary antibody (Ab) only (red) and DAPI (blue) in adjacent sections are used as staining controls. A few F4/80⁺ cells are present in the subretinal space of control eyes. Only a few Iba⁺ cells were observed in the outer nuclear layer (ONL) at 4 months p.i., whereas many Iba1⁺ cells were noted in the ONL at 8 months p.i. Iba1⁺ cells were observed in the subretinal space at 18 months p.i. RPE cells were in contact with the inner nuclear layer (INL) because of the loss of the entire ONL. Data are expressed as means ± SEM. n = 6 (A, 8 months p.i.); n = 4 (A, 18 months p.i.). ∗P < 0.05, ∗∗P < 0.0001 (U-test).

Discussion

MCMV can initiate latent ocular infection following intraocular inoculation.^24,44, 45, 46 Although MCMV DNA is not detected within eye tissues of immunocompetent adult C57BL/6 mice several months post-systemic MCMV infection,¹⁷ systemic MCMV can still spread to, and become latent in, the eye in MCMV-susceptible mouse strains. A study by Bale et al⁴⁷ reported that latent ocular MCMV was detected in 10% of Swiss-Webster mice infected with MCMV via the intraperitoneal route. On the other hand, a recent study from Voigt et al⁴⁸ found that MCMV spreads to the eye following i.p. inoculation of adult BALB/c mice with 1 × 10⁴ PFU, resulting in latent infection in both the iris (26%) and choroid (22%). No latent MCMV or pathologic changes were detected in eyes of MCMV-resistant C57BL/6 mice several months post-systemic neonatal infection (our unpublished preliminary studies). However, the results presented here demonstrate that MCMV undergoes ocular latency in BALB/c mice infected very early in life and that latent virus can be reactivated both in vitro and in vivo, the former from posterior eye cups, which include the choroid and RPE, and the latter following systemic immunosuppression.

Virus-induced disruption of an immature outer blood-retina barrier and weak antivirus immunity⁴⁹ in neonatal mice can contribute to the establishment of MCMV latency in the choroid/RPE of mice infected early in life. Systemic MCMV disseminated not only to the choroidal endothelia, but also passed through the outer blood-retina barrier, and infected pericytes and RPE cells in most neonatally infected mice, although the barrier appeared intact when viewed through an electron microscope. Relevant to this observation is the fact that the eye is an immune-privileged site in which inflammatory responses are limited to minimize the risk to vision integrity.50, 51, 52, 53, 54 RPE cells play an important role in this phenomenon by producing immunosuppressive factors, such as transforming growth factor β, α-melanocyte-stimulating hormone, and vasoactive intestinal peptide50, 51, 52, 53 as well as by inhibiting immune T cells and converting T cells to T regulatory cells.⁵⁴ Therefore, the immunosuppressive ocular microenvironment around the choroid and RPE may facilitate the establishment of MCMV latency in this tissue.

Although no replicating virus was recovered from eyes of latently infected mice, expression of multiple virus genes was detected in the eyes of aged mice following systemic neonatal infection. Many of these genes function in immune modulation or inhibition of cell death, which could facilitate the establishment of ocular virus latency. MCMV attempts to counteract the immune response via expression of multiple virus genes, including m04 and m138,⁵⁵ which was detected in the latently infected eyes. The 34-kDa glycoprotein encoded by m04 modulates expression of class I major histocompatibility complex and natural killer group 2 member D ligands, thereby silencing natural killer cells and avoiding cell-mediated immunity.56, 57, 58 m138 not only down-modulates expression of mouse natural killer group 2 member D ligands, including MULT-1, RAE-1, and H60 R,^59,60 but also restricts inducible T cell costimulator ligand expression on antigen-presenting cells, thereby inhibiting T-cell responses.⁶¹ Likewise, MCMV blocks induction of programmed cell death in virus-infected cells by encoding a number of proteins that interfere with apoptosis and necroptosis to keep infected cells alive.⁶² The MCMV genes m36, m38.5, m41, and m45 were expressed in the eyes of latently infected mice and interact with caspase-8,⁶³ Bax,⁶⁴ Bak⁶⁵ for apoptosis, and RIP1/RIP3⁶⁶ for necroptosis, respectively. Additional studies are needed to determine whether expression of these genes is essential for the establishment of ocular virus latency and whether expression of anti–cell-death genes contributes to the angiogenetic pathology in latently infected eyes.

Some pathologies and abnormalities observed following prolonged MCMV ocular latency have similar features to those observed in AMD, a progressive degenerative disease and the leading cause of severe, permanent vision loss in people >60 years of age.13, 14, 15 Although the exact events that contribute to the development of AMD remain uncertain, studies have implicated immunologic and inflammatory mechanisms67, 68, 69 with various clinical and genetic data supporting a tight association between chronic low-grade inflammation and the pathogenesis of AMD.^70,71 Although AMD models in mice, rats, rabbits, pigs, and nonhuman primates have recreated many of the histologic features of the disease and provided much insight into its underlying pathologic mechanisms, there is currently no single animal model in which all of the characteristics of AMD develop in a progressive manner.^67,68,72

Only a few RPE cells were infected with MCMV and no CMV was detected in the inner retina both during acute infection and reactivation. However, retinal pathologies, including RPE disruption and dysfunction (indicated by deposits at both basal and apical aspects of the RPE), and severe degeneration of the ONL were observed in the aged infected mice. Overall, the data suggest an indirect rather than a direct effect of MCMV infection on ocular architecture. Bystander ocular cell death is an important feature of both HCMV retinitis73, 74, 75 and ocular MCMV infection.^76,77 Previous studies^23,78 have found that spread of systemic MCMV infection to the choroid of the inner section of mice is associated with death of photoreceptors in the overlying retina as indicated by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling staining. Death of photoreceptors is probably related to inflammation subsequent to choroidal MCMV infection²³ but not related to infection, infiltration, or injection trauma, because the blood retinal barrier remains intact and MCMV does not spread to the RPE layer or to the inner retina.^78,79

IE genes were expressed in most eyes of aged infected mice. Major immediate-early enhancer activity alone does not guarantee full virus reactivation and production of progeny virus,⁸⁰ because ectopic expression of the immediate-early proteins is not sufficient to drive viral genome synthesis or infectious progeny production in infected cells.⁸¹ Nevertheless, immediate-early expression modulates the cellular environment as well as the transactivation of early virus genes.⁸⁰ For instance, HCMV IE1 can trigger a proinflammatory host transcriptional response via a STAT1-dependent mechanism.^82,83 In the current experiments, increased transcription and protein production of several inflammatory/angiogenic factors, including CCL5, was detected at both time points studied. CCL5 is produced by platelets and other cells, including macrophages, eosinophils, fibroblasts, endothelium, epithelium, and endometrial cells,⁸⁴ and participates in multiple biological processes, from pathogen control to enhancement of inflammation⁸⁴ and angiogenesis.⁸⁵ Previous studies have suggested that CCL5, which is produced by human RPE cells following chronic inflammatory stimulation, could play an important role in the development of AMD via interactions with CCR3.⁸⁶ Recent studies have found that patients with geographic atrophy have higher plasma levels of CCL-5 and higher expression of CCR5 in peripheral blood mononuclear cells than healthy controls.⁸⁷ Therefore, CCL5 production during MCMV ocular latency might contribute to the development of AMD-like pathology.

Aging probably plays a significant role in the development of retinal and choroidal pathologies in infected aged BALB/c mice. The frequency of HCMV reactivation increases with age,88, 89, 90 which is attributed to the immune senescence associated with aging or an increased inflammatory response due to age-related inflammation.^91,92 Oxidative stress and its downstream signaling have been reported in aging eyes.⁹³ Coincidently, oxidative stress may also mediate the initial activation of viral gene expression during CMV latency.^94,95 In this study, virus genes m80 and m18 were expressed in eyes of 18-month–old infected mice. But not in eyes of 8-month–old infected mice. m80 encodes a MCMV assembly protein protease,⁹⁶ whereas MCMV early gene m18 encodes a protein that is an antigenic peptide recognized by CD8 T cells⁹⁷ and drives the expression of the RAE-1 family of natural killer group 2 member D ligands, leading to subsequent activation of natural killer cells.⁹⁸ Therefore, activation of these two genes could stimulate an immune response and contribute to the progression of AMD-like pathology in aged infected mice.

Although a number of virus genes were expressed in infected ocular tissue, MCMV most likely remained latent in the eyes of aged mice because the replicating virus was never recovered. Additionally, expression of virus genes m83, which encodes a virion protein expressed late in the viral replication cycle,⁹⁹ and m54, which encodes a virus DNA polymerase,¹⁰⁰ was not detected in any eyes of infected mice at 8 or 18 months p.i. However, we cannot exclude the possibility that latent MCMV is reactivated intermittently in some aged eyes, particularly because MCMV genes encoding assembly protein (m80) and virus envelope (m55) were expressed in some eyes of 18-month–old infected mice.

In summary, the results presented herein indicate that systemic MCMV infection of BALB/c mice can spread to the eye with subsequent establishment of latency at the choroid and RPE when infection is acquired early in life. Furthermore, MCMV latency in the choroid/RPE of BALB/c mice was associated with the up-regulation of inflammatory/angiogenic factors and, most importantly, the development of retinal and choroidal pathologies, which share some features with human AMD, including deposits at basal and apical aspects of the RPE, severe photoreceptor degeneration, and CNV-like lesions in later life. The electron microscopy results suggest a possible choroidal origin of CNV-like lesions (in the sub-RPE space with Bruch's membrane appearing to extend into the sub-RPE space adjacent to the new vessels). However, we cannot exclude the possibility that these CNV-like lesions originated from the deep plexus of the retinal vasculature because the RPE in CNV-like lesions was located adjacent to INL because of the loss of the entire ONL, and RPE disruption was also observed. HCMV DNA was recently found in the choroid/RPE in approximately 17% of a group of human cadavers,¹² consistent with latent infection; therefore, it is possible that CMV latency could be a risk factor for the occurrence and development of AMD.