Up-Regulation of Fas Ligand Expression by Human Cytomegalovirus Immediate-Early Gene Product 2: A Novel Mechanism in Cytomegalovirus-Induced Apoptosis in Human Retina

Cells and HCMV infection

HRPE cell line ARPE-19 and HCMV strain AD169 were obtained from American Type Culture Collection (Manassas, VA). HRPE cells were maintained in DMEM and Ham’s F-12 medium at a 1:1 ratio and supplemented with 10% FCS. HEL-299 cells were maintained in DMEM plus 10% FCS. Jurkat cells were maintained in RPMI 1640 medium plus 10% FCS. All of these cells were grown in a 37°C incubator supplied with 5% CO₂. Cell-free virus stock was prepared from the supernatants of AD169-infected HEL-299 cells as described previously (13). One day before infection, subconfluent HRPE cells were switched to serum-free medium. HRPE cells were infected with AD169 at a multiplicity of infection (MOI) of 5. After virus absorption at 37°C for 2 h, cells were washed with PBS twice, fed with fresh culture medium, and grown in a 37°C incubator supplied with 5% CO₂.

RNA analysis

Total RNA was extracted using the RNAeasy kit (Qiagen, Chatsworth, CA). First-strand cDNA synthesis was conducted using oligo(dT) primer and Moloney murine leukemia virus-reverse transcriptase (Promega, Madison, WI). The derived cDNA products (5 μl) were PCR amplified in a 50-μl reaction mixture containing 0.2 mM dNTP, 2 mM MgCl₂,10 pmol of each primer, and 1 U of Taq DNA polymerase. The primer pairs used were: hFasL forward (5′-CAACTCAAGGTCCATGCCTC-3′) and hFasL reverse (5′-AGATTCCTCAAAATTGACCAG-3′); hFas forward (5′-CTGCAGGAACGCCCCGGGACAC-3′) and hFas reverse (5′-GGTTGTTGAGCAATCCTCCGAAGT-3′); HCMV IE forward (5′-CCAAGCGGCCTCTGATAACCAAGCC-3′) and HCMV IE reverse (5′-CAGCACCATCCTCCTCTTCCTCTGG-3′); β-actin forward (5′-TCCTGTGGCATCCACGAAACT-3′) and β-actin reverse (5′-GAAGCATTTGCGGTGGACGAT-3′). PCR amplification was performed for 35 cycles at 94°C for 1 min, 52°C for 1 min, and 72°C for 2 min. The expected sizes of PCR products for FasL, Fas, and β-actin are 350, 266, and 314 bp, respectively.

Western blot assay

HCMV AD169-infected HRPE cells were harvested at days 1, 2, 3, 5, and 7 postinfection. The cell lysates were prepared as described. Fifteen microliters of sample was boiled at 95°C for 5 min and separated on a 10% SDS-PAGE. The proteins were transferred to Hybond-ECL nitrocellulose paper (Amersham, Arlington Heights, IL) by wet-transfer system. The mAbs used to identify protein products were anti-FasL mAb (F37720/lot 3; BD Transduction Laboratories, Lexington, KY), anti-Fas mAb (F22120; Transduction Laboratories), anti-CMV IE mAb (MAB810; Chemicon International, Temecula, CA), and anti-β-actin mAb (MAB81501; Chemicon International). The reactive protein bands were detected by the ECL detection system (Amersham).

Transfection and dual-luciferase assay

The Lipofectamine 2000 system (Life Technologies, Grand Island, NY) was used for transfection of DNA into HRPE cells. The IE1 and IE2 plasmids had been described previously (14). The FasL promoter constructs (wild-type (wt) and deletion mutants) were kindly provided by Dr. C. Warrant (15). Briefly, 10⁶ cells/well were seeded onto 6-well plates 24 h before transfection. DNA solution (250 μl of OPTI-MEM containing 3 μg of pGL2-hFasL reporter construct, 3 μg of p1R (IE1), 3 μg of pSIE2-wt (IE2) or mutant effector plasmids (IE2CD80), and 30 ng of pRL-tk Renilla normalizing luciferase vector) and reagent solution (5 μl of LF200 and 250 μl of OPTI-MEM) were separately preincubated at 25°C for 5 min. The transfections were then followed according to the manufacturer’s instructions. After 48 h of incubation, the cell lysates were prepared using the dual-luciferase reporter assay system (Promega). Both firefly and Renilla luciferase activities were monitored by illuminator (TD20/20; Turner Designs, Sunnyvale, CA).

ELISA test and flow cytometry

The concentration of soluble FasL (sFasL) in vitreous samples and in supernatants from infected HRPE cells was determined by ELISA kit (MBL, Nagova, Japan). The developed reaction was quantified by reading at A₄₉₀ nm (MRX; Dynatech Laboratories, Chantilly, VA). Each individual sample was analyzed in triplicate. Apoptotic cells were determined by flow cytometry using propidium iodide and FITC-labeled annexin V (Immunotech, Luminy, France). Cells (10⁶ cells/ml) were incubated with relevant Abs in PBS containing 1% BSA. The reaction was conducted at 4°C for 30 min in the dark. After incubation, the cells were washed twice with cold PBS, resuspended to 10⁶ cells/ml in 250 μl of PBS, and analyzed by flow cytometry using a FACSCalibur apparatus (BD Biosciences, San Diego, CA). For blocking assays, the culture supernatants were pretreated with anti-FasL mAb (clone 4H9; MBL, Nagova, Japan) for 60 min at 37°C before the addition of Jurkat cells. Alternatively, Jurkat cells were pretreated with anti-Fas mAb (clone ZB4; MBL) for 60 min at 37°C before the addition of supernatants from AD169-infected HRPE cells.

Immunohistochemistry and in situ hybridization

The research followed tenets of the Declaration of Helsinki. All samples were obtained after informed consent from the patients. Twenty-four eyeballs enucleated from dead AIDS patients were obtained at autopsy and immediately fixed with 4% paraformaldehyde. The 4-μm paraffin sections were deparaffinized in xylene, rehydrated in a series of graded alcohols, and immunostained with Abs against CMV IE Ag (MAB810; Chemicon), FasL (Transduction Laboratories), and HIV-p24 (clone Kal-1; DAKO, Glostrup, Denmark). Immunoreactive signals were detected with a mixture of biotinylated IgG Ab and peroxidase-conjugated streptavidin (LSAB2 system; DAKO). Apoptotic cells in tissue sections were identified by the TUNEL method (In Situ Cell Death Detection kit, POD; Boehringer Mannheim, Mannheim, Germany) as described previously (16). As a negative control, the TdT was omitted from the TdT buffer. Chromogenic development was then applied with 3-amino-9-ethyl-carbazole and slides were counterstained with methyl green. Positive staining was identified under the light microscope as red granules. The RNA in situ hybridization procedure was performed as described previously (17). The 4-μm deparaffinized slides were predigested with nuclease-free proteinase K (0.1 mg/ml; Boehringer Mannheim). One hundred microliters of FITC-conjugated antisense and sense probes was added onto each slide. The sequences of FasL probes used were: 5′-GAGCTTGAAACATCCCCAGGCCCAATCCTACCAAGGCAACC-3′ (antisense primer, 461–500 nt) and 5′-GGGATGTTTCAGCTCTTCCACCTACAG-3′ (sense primer, 486–512 nt). Hybridization products were visualized by alkaline phosphatase conjugated secondary Abs and chromogen nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (Sigma, St. Louis, MO). Positive staining was identified as purple granules under the microscope.

Up-regulated expression of FasL in HCMV-infected HRPE cells

Since constitutive expression of FasL is critical for maintaining the immune privilege of the eye, we hypothesized that dysregulated FasL expression by HCMV might be responsible for the pathogenesis of HCMV retinitis. In the eye, HRPE cells are located at the blood-retina barriers (13, 18). Infection of HRPE cells has been shown to result in serious ocular manifestations and suspected to be critical for the development of HCMV retinitis (13). In this study, we first infected a HRPE cell line, ARPE-19, with HCMV AD169 and monitored the expression of FasL. We found that the synthesis of FasL protein in HCMV-infected HRPE cells was enhanced early after infection. The up-regulation of FasL was initiated at day 1 after infection (Fig. 1⇓A). In contrast, the expression level of Fas remained unchanged (Fig. 1⇓A). Consistent to the Western blot data, we observed an increase of FasL mRNA (Fig. 1⇓B). These experiments suggest that HCMV induction of FasL is specific and is determined at the transcriptional level. The early induction of FasL in infected cells further implicates that viral IE genes might play a role in this event. This idea was supported by the evidence that the mRNA level of IE genes was also induced at day 1 after infection (Fig. 1⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Analysis of FasL expression in HCMV-infected HRPE cells. A, Up-regulated FasL protein expression in HCMV-infected HRPE cells. HRPE cells were infected with HCMV AD169 at a MOI of 5 PFU/cell. The infected cell lysates were collected at days 1 and 2 after infection and were analyzed by Western blot assay. B, HCMV infection induces the mRNA level of FasL in HRPE cells. Infection was performed as described above. Total RNA was collected at day 1 after infection and subjected to RT-PCR with oligonucleotide primers specifically complementary to the coding region of FasL, HCMV IE, and β-actin, respectively. The amplified PCR products were visualized with ethidium bromide on agarose gel.

Induction of FasL promoter by HCMV IE2

Since IE2 is a potent transactivator (19, 20), we set up to examine whether HCMV IE2 directly regulates FasL expression. We found that IE2 induced the activity of FasL promoter efficiently in HRPE cells. The activation was specific for IE2 because, first, another HCMV IE protein, IE1, had no effect on the same promoter (Fig. 2⇓A). Note that the inability of IE1 in inducing FasL was not due to a low level of protein being made. In fact, the IE1 protein was more abundant than IE2, based on the Western blot results using a mAb (mAb810) against the common region (exon 3) shared by IE1 and IE2 proteins. Second, removal of IE2 C-terminal 80 residues (IE2CD80), which are required for repressing the transactivation activity of the tumor suppressor protein p53 (14), greatly reduced the protein’s activation activity on FasL promoter (Fig. 2⇓B). Third, IE2 no longer activated a truncated version of FasL promoter without its 5′ region from −1147 to −370 nt (Fig. 2⇓C). Thus, we conclude that HCMV IE2 is able to specifically regulate the FasL promoter, providing a molecular mechanism linked to the pathogenesis of HCMV retinitis.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Specific activation of FasL promoter by HCMV IE2. A, HCMV IE2, but not IE1, activates the FasL promoter in transient transfection assays. FasL promoter-luciferase constructs were transfected into HRPE cells along with plasmids encoding IE1 or IE2. After harvesting the cells, lysates were assayed for luciferase activity. All relative luciferase activity was normalized with control Renilla luciferase expression. To ensure the expression of the effectors, Western blottings with mAb (MAB810; Chemicon) specifically recognizing the third exon, which is shared by HCMV IE1 and IE2 proteins, are shown. The expression level of β-actin was used as an internal control. B, IE2, without its C-terminal 80 residues, is unable to activate the FasL promoter. Transfections were done as described in A except that a truncated form of IE2 which loses its C-terminal 80 residues (IE2CD80) was used as indicated. C, 5′ deletion mutants of FasL promoter fail to respond to IE2 activation. Transfections were applied as described in A. The transcriptional activity of wt and two mutants of FasL promoter with serial 5′ deletion (−370 and −339) were assayed in the presence of IE2. The relative fold of luciferase activity has been normalized with the luciferase expression in the absence of IE2. Data shown here are the mean ± SE of two independent experiments, each with triplicate luciferase results.

Enhanced sFasL from HCMV-infected HRPE cells caused Fas-dependent apoptosis of Jurkat cells in vitro

Identification of the correlation between IE2 and FasL up-regulation in cultured HRPE cells prompted us to examine whether the increased FasL upon HCMV infection induces apoptosis of uninfected cells as observed in HCMV-infected human retina (see below). Consistent with previous data, the levels of sFasL in the cell-free supernatants of infected HRPE cells were increased since day 1 after infection (Fig. 3⇓A). To further investigate whether secreted FasL from HCMV-infected HRPE cells causes Fas-dependent apoptosis in vitro, we cultivated Jurkat cells in medium supplemented with 50% of supernatants from mock or HCMV-infected cell cultures, respectively. After 48 h of incubation, the percentage of apoptotic cells in the cultures was determined by flow cytometry. We found that cells exposed to HCMV-infected supernatants exhibited a higher percentage of apoptosis as compared with those exposed to mock supernatants (Fig. 3⇓B). Furthermore, functional blocking of Fas or FasL with respective mAbs prevented the HCMV-induced apoptosis of uninfected Jurkat cells (Fig. 3⇓B). Interestingly, we found that neither anti-Fas nor anti-FasL could block the apoptosis of ∼8% of Jurkat cells when cultured with mock-infected supernatants (Fig. 3⇓B), suggesting that other FasL/Fas-independent cell death pathways were also involved. Our results provide evidences that HCMV-infected HRPE cells are able to release large amounts of FasL, which, in turn, can induce apoptosis of nearby cells, presumably through a Fas-mediated pathway.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Secretion of FasL from HCMV-infected HRPE cells and its effects in mediating apoptosis of Jurkat cells. A, Induction of sFasL in the supernatants of HCMV-infected HRPE cells. HRPE cells were infected with HCMV AD169 at a MOI of 5 PFU/cell. The levels of sFasL in the supernatants of infected cells at days 1, 3, 5, and 7 after infection were determined by ELISA (clone 4H9; MBL). Data shown here are the mean ± SE of three experiments. B, Induced apoptosis of Jurkat cells by HCMV-infected supernatants was blocked by anti-Fas and anti-FasL Abs. Jurkat cells were incubated in medium supplemented at a 50% concentration with either HCMV-infected supernatants of HRPE cells from days 1, 3, 5, and 7 after infection or supernatants from mock-infected cells (Mock). The percentage of apoptotic cells was determined by annexin V staining and flow cytometry. To further examine whether observed apoptosis is due to Fas-FasL interaction, blocking assays were conducted using either anti-Fas mAb (200–500 ng/ml, clone ZB4; MBL) or anti-FasL mAb (500 ng/ml, clone 4H9; MBL). The anti-Fas Ab and anti-FasL Ab were added to Jurkat cells or supernatants before incubation, respectively. Data shown here are the mean ± SE of three experiments.

Increased expression of FasL in vitreous and retina of HCMV retinitis patients

To examine whether our previous observation is physiologically relevant, we investigated whether sFasL could be released from HCMV-infected retina into a vitreous body. Diagnosis of HCMV retinitis was based on the clinical results of indirect ophthalmoscopy (Fig. 4⇓A) and fluorescein angiography (data not shown). As presented in Table I⇓, before anti-HCMV treatment, the averaged sFasL level of vitreous fluids among the eight AIDS patients with HCMV retinitis was 440 pg/ml, with a SD of 36 pg/ml. In contrast, after complete ganciclovir treatment and the appearance of regressed lesions in HCMV-infected retina, the sFasL levels of vitreous fluids in these eight patients declined dramatically (Table I⇓). To exclude the potential effects of HIV infection on the release of sFasL, seven AIDS patients without HCMV retinitis were also examined. We found that the sFasL levels in these patients were all below the detection limit (Table I⇓).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Enhanced expression of FasL and apoptotic cells in retina of HCMV retinitis patients. A, A fundus photograph of HCMV retinitis by ophthalmoscopic examination. The large areas of retinal hemorrhage with necrosis, diffuse edema, and yellowish white exudation extended along the lower branch of retinal vessels. B–D and H, Representative retina sections from AIDS patients without HCMV retinitis. INL, inner nuclear layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium. E–G, I, and J, Retina sections from AIDS patients with HCMV retinitis. B and E, Detection of HCMV IE by immunohistochemistry (arrows). C and F, Detection of apoptotic cells by TUNEL assay. D, G, and J, Detection of FasL by in situ hybridization (arrows). H and I, Detection of the colocalization of FasL and HCMV IE Ags by dual immunohistochemistry staining. FasL signals were identified by chromogen nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate staining as purple granules (black arrows). HCMV IE signals were identified in the nucleus by chromogen 3-amino-9-ethyl-carbazole as red granules (green arrowheads). Original magnification: B–G, ×200; H, ×400; I, ×480; J, ×560.

To test further, the postmortem enucleation of the eyeball was performed on patients with or without HCMV retinitis. The specificity of HCMV infection in these areas was confirmed by detection of HCMV IE Ags in all layers of retina within cytomegalic cells, a typical sign of HCMV infection, and in many nonmegalic cells (Fig. 4⇑E) as compared with that obtained from AIDS patients without HCMV retinitis (Fig. 4⇑B). To examine whether apoptosis occurs in these regions, TUNEL assay was performed. We observed many apoptotic cells in the foci of HCMV-infected retina (Fig. 4⇑F). In contrast, no apoptotic cells were identified in HCMV IE-negative samples (Fig. 4⇑C). Consistent with our observation that HCMV IE2 up-regulates the FasL expression, we detected higher levels of FasL in HCMV IE-positive specimens (Fig. 4⇑, G and J) as compared with the basal level of FasL in AIDS patients without HCMV retinitis (Fig. 4⇑D). Dual immunohistostaining for FasL and HCMV IE Ag was also conducted to determine whether increased expression of FasL does correlate with HCMV infection (Fig. 4⇑, H and I). We found that the FasL and HCMV IE Ags were colocalized in infected neuroretina and HRPE cells of HCMV-infected retina (Fig. 4⇑I). A number of HRPE cells with intranuclear and typically cytoplasmic inclusions in the lesions of HCMV-infected retina were also observed under the microscope (Fig. 4⇑I). Arguing against the involvement of HIV in the lesions of HCMV-infected retina, immunostaining with HIV Ag-specific Abs was performed on samples collected from AIDS patients with or without HCMV retinitis. We found that no HIV-1 p24 Ag was present in HRPE cells among all specimens examined (data not shown). The rarely detected HIV-1 p24 Ag was found to localize at few neuroretinal cells (Müll cells; data not shown). Therefore, HIV might not have substantial influence in this instance if there is any.