Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Herpes Simplex Virus Type 1 Alters Transcript Levels of Tumor Necrosis Factor-a and Interleukin-6 in Retinal Glial Cells Kristen M. Drescher* and Judith A. Whittum-Hudsonf Purpose. Studies were performed to determine whether retinal Muller cells transcribe genes for the proinflammatory cytokines interleukin-6 (IL-6) and tumor necrosis factor-a (TNFa). Isolated murine retinas were used to test whether these cytokines were upregulated in the retina in vivo after anterior chamber inoculation of herpes simplex virus type 1 (HSV-1). The effects of exposure to HSV-1 or interferon-7 (IFNy) on transcript levels of these cytokines in cultured retinal glia also were examined. Methods. In situ hybridization (ISH) using digoxigenin (DIG)-labeled RNA probes was used to localize mRNA for IL-6 and TNFa in cultured retinal glial cells. Changes in IL-6 and TNFa relative transcript levels were assessed in cultured retinal glial cells using a semiquantitative approach comprised of reverse transcription-polymerase chain reaction (RT-PCR) assay at low amplification cycle number followed by slot blotting and hybridization with DIG-labeled internal sequence probes. In the murine model of herpetic retinitis, the same methods were used to compare temporal changes in relative cytokine transcript levels in retinas isolated from eyes 1 to 7 days after anterior chamber injection of live HSV-1 (KOS strain; 2 X 104 pfu/eye) or buffer with levels in retinas isolated from normal, uninjected eyes. Densitometry was used to quantify relative signal changes obtained with serial diluted samples in slot blot assays. Cytokine signal was normalized to hypoxanthine phosphoribosyl transferase signal obtained from the same cDNA samples. Results. Under baseline culture conditions, ISH and RT-PCR indicated that both IL-6 and TNFa were transcribed by cultured retinal glia. In vitro exposure to either viral (HSV-1) or inflammatory (IFN7) stimulants increased levels of these transcripts in a time-dependent manner. Peak TNFa mRNA levels were detected 4 hours after exposure to HSV, whereas IL6 peaked 4 hours later (increases of 10.3 and 8.7 times over baseline, respectively). Differential increases in TNFa and IL-6 transcript levels were detected in retinas isolated from BALB/c mice that received anterior chamber injections of either HSV-1 or Hanks' balanced salt solution (HBSS). By day 3 after HSV-1 injection, increases of 4.5-fold in TNFa and 17-fold in IL-6 were detected, whereas substantially smaller changes in TNFa and IL-6 (1.5-fold and 6.3-fold, respectively) were observed in HBSS-injected eyes. Virus-induced changes in TNFa mRNA levels occurred slighdy earlier than for IL-6 because maximal levels of TNFa were detected 2 to 3 days after infection, but IL-6 peaked at day 3. Conclusions. Cultured retinal glial cells exhibit upregulated TNFa and IL-6 transcript levels after exposure to virus or inflammatory mediators. HSV-1 infection of die anterior segment of the mouse eye markedly upregulates TNFa and IL-6 mRNA levels compared to smaller responses to nonspecific inflammation. Taken togedier, these results identify retinal Muller cells as an intraretinal source of TNFa and IL-6 and support die potential of diese resident cells to act as intraretinal modulators of immune and inflammatory responses. Invest Ophthalmol Vis Sci. 1996;37:2302-2312. .M.uller cells are a major nonneuronal component of the retina. These retinal cells respond rapidly to ocular injury or inflammation by the upregulation of glial fibrillary acidic protein (GFAP).' ~5 Although the func- 2302 Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 tional significance of GFAP expression is unknown, changes in levels of this protein suggest that Muller cell activation is an intraretinal response to ocular injury. Muller cells may participate in intraocular im- Investigative Ophthalmology & Visual Science, October 1996, Vol. 37, No. 11 Copyright © Association for Research in Vision and Ophthalmology Production of TNFa and IL-6 by Retinal Glia mune or inflammatory reactions through soluble mediator release. Central nervous system (CNS) astrocytes have been shown to upregulate GFAP'1 and to produce soluble mediators in vitro and in vivo in response to a variety of exogenous agents.7"8 However, the capacity of Miiller cells to produce immunologically important cytokines has not been investigated in detail.9 Cytokine production and release have been shown to play a vital role in the initiation, regulation, and subsequent resolution of immune and/or inflammatory responses to a variety of pathogens. These multifunctional soluble mediators are produced locally and act either in an autocrine or a paracrine manner. The cytokine balance during a host immune response to an infectious agent influences whether the pathogen is eliminated, with minimal damage to the host, or whether pathogen- or immune-mediated damage occurs.10"13 The interactions of multiple mediators may be antagonistic or synergistic and may induce the production of additional soluble mediators. Factors altering these balances of soluble mediators within the neural retina to permit uncontrolled inflammation or infection could have grave consequences, including loss of vision. Both tumor necrosis factor-a (TNFa) and interleukin (IL)-6 are proinflammatory cytokines synthesized in vitro and in vivo by a variety of cell types including T cells, macrophages, fibroblasts, and CNS astrocytes. These cytokines may be produced either constitutively or in response to a variety of exogenous stimuli ranging from endotoxin to infectious pathogens.1'1""' Antiviral properties of TNFa and IL-6 have been reported, although in some cases they were associated with increased viral replication.1718 Although intravitreal delivery of TNFa and several other soluble products promotes photoreceptor survival in an experimental rat model of retinal degeneration,19 intraretinal production of TNFa or IL-6 has not been demonstrated previously. Information regarding production of these and other immunologically relevant soluble mediators by resident cells of the neural retina is limited. In the eye, several resident cell populations From the * Department of Molecular Microbiology and Immunology, /ohns Hopkins University School of Hygiene, and Public Health, and fl'lie Wilmer Ophthalmological Institute, Johns Hopkins University School of Medicine, Baltimore. Maryland. Presented in part at the 1994 annual meeting of the Association for Research in Vision and. Ophthalmology, Sarasota, Florida, and. in the doctoral dissertation of HMD. Supported in part l/y Research to Prevent Blindness, Inc., and l/y grants from Fight for Sight Research Division of Prevent Blindness America (WH); Sigma Xi Granlin-Aid oj Research (KD); and Public. Health Service National Research Service Award 5T2 EYO7O47 (KD). Submitted for publication Sef>tember 15, 1995; revised May 21, 1996; accepted June 14, 1996. Proprietary interest category: N. Reprint requests: Judith A. Wliittum-Hudson, The Wilmer' Institute, Johns Hopkins University School of Medicine, 457 Wilmer-Woods, 600 North Wolfe Street, Baltimore, MD 21287-9142. Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 2303 have been shown to produce proinflammatory cytokines either constitutively or in response to inflammatory stimuli.20"2'1 The majority of previous studies have focussed on cells of the anterior segment, in particular the iris, ciliary body, and cornea.20"22'2'1 The retinal pigment epithelium has been shown to produce soluble mediators, including IL-6 and IL-8.2'"27 Recendy, we reported that murine retinal glia could be induced in vitro by IFNy to express or upregulate major histocompatibility complex class I and II antigens and intercellular adhesion molecule-1. Immunohistochemical staining indicated that TNFa and IL-6 were constitutively expressed and that secretion of both cytokines was induced by exposure to IFNy, lipopolysaccharide, or herpes simplex virus type l(HSV-l).28 We wanted to extend these studies to determine the temporal sequences of transcriptional changes for these cytokines in retinal glia that might correlate with in vivo responses by Miiller cells. In this study, we demonstrate that cultured retinal glial cells modulate their transcript levels of IL-6 and TNFa in vitro when stimulated by virus or exogenous cytokines. Further, results obtained from retinas isolated from HSV-infected eyes demonstrate that one or more neural retinal cell populations transcribe the genes encoding IL6 and TNFa. Anterior segment HSV-1 infection, with its associated prolonged inflammation, upregulates transcript levels of these genes more than observed during transient inflammation. Together, these results support the hypothesis that murine retinal glia act as participants in intraocular immune and inflammatory responses by soluble mediator production and release and that the duration and amplitude of these responses will be influenced by the nature of the stimulus. MATERIALS AND METHODS Animals Breeding pairs or pregnant BALB/c mice were obtained from Harlan Sprague-Dawley (Indianapolis, IN). Adult BALB/c female mice used for in vivo studies were obtained from the Charles River Breeding Facility (Wilmington, MA). Mice were maintained in the animal facilities of the Wilmer Institute in the Johns Hopkins Medical Institution. All studies complied with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and with the National Institutes of Health Guidelines on the Use of Animals in Research. Retinal Glial Cell Cultures Primary retinal glial cell cultures were established using our published techniques,21 which are a modification of the methods of Politi et al.30 Cells were main- 2304 Investigative Ophthalmology & Visual Science, October 1996, Vol. 37, No. 11 tained at 37°C in 5% CO2 until they became semiconfluent. Primary cultured glia exhibited flat cell morphology, and 100% were positive for GFAP by immunohistochemical staining.29 Cells were trypsinized and passaged to establish retinal glial cell lines, which were maintained in complete Dulbecco's minimal essential medium (Gibco, Gaithersburg, MD) containing 10% fetal calf serum (Gibco), 100 U/ml penicillin (Gibco), 100 /ig/ml streptomycin sulfate (Gibco), and 200 mM L-glutamine (Gibco). Cells from passage numbers less than 20 to more than 100 were compared. The cells replicated rapidly, were fibrous in appearance, and more than 99% expressed GFAP, which was verified periodically by avidin-biotin-peroxidase complex immunohistochemistry. In Vivo Herpes Simplex Virus Type 1 Ocular Infection Herpes simplex virus type 1 (KOS strain) was propagated in VERO cells, and 2 X 104 pfu/4 (A was inoculated into the anterior chambers of eyes of anesthetized (0.66 mg ketamine hydrochloride, Vetalar; Parke, Davis, Detroit, MI) BALB/c mice as described31; control eyes received 4 /il of HBSS. All virus dilutions were made in HBSS. Mice were killed by cervical dislocation, the eyes were removed, and the retinas were isolated immediately for RNA extraction as described below. Retinas from uninjected mice were used as baseline controls. Stimulation of Glial Cells and Isolation of RNA Long-term cultured glial cells were grown in T25 flasks in complete medium and exposed to rIFNy (50 U/ ml; Gibco) or HSV-1 (KOS) at a multiplicity of infection (moi) of 10 or 0.1. Virus (0.5 to 1 ml) was adsorbed to monolayers for 60 minutes at 37°C, after which 10 ml of complete medium was added; rIFNy was diluted directly in 10 ml of medium. Flasks were incubated with stimulant at 37°C in 5% CO2 for 1 to 24 hours; medium control flasks were cultured in parallel. Total RNA was extracted from cultured cells (one T25 flask per treatment and time point, ~2 X 10° cells) or ocular tissues (four retinas per time point for each treatment) with the Trizol Reagent (Gibco) using the manufacturer's instructions. The resultant RNA pellets were resuspended in 0.1% diethylpyrocarbonate (depc)-treated distilled water (Sigma). Samples were then DNase-treated (DNase I; Promega, Madison, WI), resuspended in depc-water to a concentration of 1 fjLg/fjt\, and stored at —80°C until use. Probe Preparation and In Situ Hybridization Plasmids containing cDNA specific for IL-6 and TNFa were gifts from Dr. Paula M. Pitha-Rowe (Johns Hopkins University, Baltimore, MD). RNA probes were digoxigenin (DIG)-labeled for in situ hybridization Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 (ISH) according to the manufacturer (Boehringer Mannheim, Indianapolis, IN) using either the SP6 (IL6) or the T7 (TNFa) promoter. Negative-control sense probes were prepared using the alternate promoter. Cultured retinal glia were obtained by trypsinization and 3 X 105 cells/ml cytocentrifuged onto poly-L-lysine-coated (Sigma) slides. Slides were fixed in freshly prepared 4% paraformaldehyde, dehydrated in 90% methanol, and stored at —80°C until use. At the time of ISH, slides were rehydrated in a methanol series (75%, 50%, 25%), acetylated (300 ml depc-treated water, 3.5 ml 0.1 M triethanolamine, 750 fi\ acetic anhydride), and prehybridized (1 hour at 48°C). Slides were hybridized with anti-sense or control, sense DIG-labeled probes overnight at 48°C, then washed in 50% formimide/2 X SSC, RNase treated (4 Mg/ml RNase A, 20 U/ml RNase Tl), blocked (2% blocking reagent; Boehringer Mannheim), and incubated with the alkaline phosphatase-conjugated antiDIG antibody (1:500). Slides were developed with nitroblue tetrazolium (Sigma) and 5-bromo-4-chloro-3indolylphosphate (Sigma) for 14 to 18 hours according to the manufacturer's instructions. Reverse Transcription-Polymerase Chain Reaction for Cytokines Total cellular RNA (1 //g) from each source was reverse transcribed with 0.5 U random hexamer oligonucleotides as primer (Boehringer Mannheim) and 200 U Moloney Murine Leukemia Virus (M-MLV) reverse transcriptase (Gibco) according to the manufacturer's instructions. Each resultant cDNA preparation was RNase treated (RNases H, Tl, A), extracted in phenolxhloroform, ethanol precipitated, and dissolved in 25 fil depc-water. Polymerase chain reaction amplifications of cDNA were performed under the following conditions: 50 pmol each of the appropriate upstream and downstream primers (obtained from previously published sequences for hypoxanthine phosphoribosyl transferase [HPRT] and TNFa,32 and IL-633), 10X PCR buffer (Promega, Madison WI), 1.5 mM MgCl2, and 200 fiM dNTP (Pharmacia, Piscataway, NJ) were prepared as a master mix that was aliquoted. cDNA (4 (A) and 2.5 U DNA Taq polymerase (Ampli-Taq; Perkin Elmer, Foster City, CA) were added to an aliquot of the master mix in a total reaction volume of 100 [A. Samples were overlaid with mineral oil (Sigma) and amplified in a DNA thermal cycler (BarnsteadThermolyne, Dubuque, IA). Conditions of the PCR reaction were initial denaturation at 94°C for 3 minutes followed by cycles consisting of annealing at 57°C for 1 minute, extension at 72°C for 2 minutes, and denaturation at 94°C for 30 seconds. For reverse transcription-polymerase chain reaction (RT-PCR) to be analyzed by Southern slot blotting, 27 cycles were performed with a final extension at 72°C for 7 minutes. Production of TNFa and 11^6 by Retinal Glia The housekeeping gene HPRT was amplified (25 cycles) from an additional aliquot of each cDNA to ensure that equal amounts of input RNA were added per reaction and to provide a means of comparison between samples. For screening RT-PCR, cDNA was amplified under the same conditions but for 35 to 40 cycles. Predicted sizes of the cDNA for each cytokine primer pair were IL-6, 426 bp 33 ; TNFa, 300 bp 32 ; and HPRT, 162 bp.32 The use of random hexamer priming in the reverse transcription step reduced intersample variability because aliquots of the same resultant cDNA were amplified with the different primer sets and compared to cytokine transcript signal for the cDNA obtained from normal retinal RNA. Samples from all time points of each independent time course experiment were reverse transcribed in the same RT step; PCR reactions for a given cytokine and HPRT were then performed with the various cDNA. The same normal retinal RNA was used for comparisons with all time points in each experiment. Control experiments demonstrated that amplifications for slot blotting were performed on the linear portion of the curve. cDNA products from screening RT-PCR were verified by electrophoresis of 10 //I (10%) of each sample on a 1% agarose gel and subsequent transfer of the DNA to a nylon membrane (Schleicher & Schuell, Keene, NH) using standard Southern blot procedures.34 Membranes were probed with DIG-labeled cytokine- and HPRT-specific oligonucleotides for published internal sequences32'33 using the Genius system (Boehringer Mannheim) to verify identities of the PCR products. Slot Blotting of Polymerase Chain Reaction Products After amplification, 15 y\ (IL-6) or 30 //I (TNFa) of the respective PCR products were denatured by heating at 100°C for 10 minutes in 500 (i\ of 6 X SSC (90 mM sodium citrate, 900 mM sodium chloride). Slot blot assays and analyses were performed by a modification of the method of Wynn et al.34 Samples were placed immediately on ice, and 2-fold serial dilutions (500 /il/dilution) were prepared in 6 X SSC buffer. Samples were transferred onto Nytran membrane by standard methods33'34 using a slot-blot apparatus (Schleicher & Schuell). DNA was cross-linked to the membrane using a Stratagene (La Jolla, CA) 2400 UVcross-linker. Blots were prehybridized in standard prehybridization solution (Boehringer Mannheim) overnight at 40°C. Hybridization to DIG-labeled oligonucleotides, and subsequent washes were performed according to manufacturer's instructions. After incubation with an alkaline phosphatase labeled anti-DIG antibody (1:500), hybridized products were visualized by development with nitroblue tetrazolium and 5bromo-4-chloro-3-indolylphosphate using the Genius Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 2305 1 Kit (Boehringer Mannheim). Blots were analyzed by densitometry, and signals were normalized to those obtained for HPRT. Values for each cytokine were expressed as the mean-fold increase in relative signal (±SEM) of two dilutions for duplicate slots (four densitometric values per time point) compared to unstimulated cells or normal retina. Cytokine transcript levels in the control samples arbitrarily were assigned a value of 1 to calculate fold-increase over baseline. Two independent time course experiments were performed, each with duplicate samples of extracted retinal tissue and cultured retinal glia. RESULTS In Situ Hybridization for Tumor Necrosis Factor-or and Interleukin-6 mRNA in Cultured Retinal Glia When cultured glia were exposed to either rlFNy or live HSV, increased hybridization signal was detected for both groups of cells from 1 to 12 hours after stimulation. Representative ISH results for glia exposed to HSV-1 for 8 hours are shown in Figure 1. Greater signal for both TNFa and IL-6 mRNA was observed in glia cultured with HSV-1 (Figs. 1A, 1C) than in unstimulated cells (Figs. IB, ID). Digoxigenin-labeled anti-sense probes specific for either TNFa or IL-6 hybridized weakly to 100% of uninfected retinal glia (Figs. IB, ID), confirming that there are low levels of mRNA specific for these genes under standard culture conditions. Co-culture with rIFNy also upregulated transcripts of both cytokines (not shown). Control cells that received the DIG-labeled sense probes demonstrated no detectable signal regardless of pretreatment (not shown). These results are consistent with our previous studies that demonstrated increased intensity of immunohistochemical staining for cytokine protein after exposure to IFNy or HSV-1.28 Stimulation of Tumor Necrosis Factor-** and Interleukin-6 Transcription in Murine Retinal Glia by Herpes Simplex Virus Type 1 and Interferon y Cultured retinal glial cells transcribed the genes for TNFa and IL-6 under normal culture conditions as assessed by RT-PCR. Similar results were obtained for long-term and unpassaged cells from BALB/c and C57B1/6 mice. A representative ethidium bromidestained gel is shown for cells exposed to stimuli for 8 hours in Figure 2 and demonstrates further that unstimulated glia transcribe TNFa and IL-6 genes. Because basal transcript levels for these cytokines were detected, we used a more quantitative assay to determine relative changes in TNFa and IL-6 tianscript levels in cultured retinal glia after exposure to IFNy 2306 Investigative Ophthalmology & Visual Science, October 1996, Vol. 37, No. 11 c .4 », I e FIGURE l. In situ hybridization (ISH) evidence that herpes simplex virus type 1 (HSV-1) upregulates tumor necrosis factor (TNF) a and interleukin {IL)-6 transcription in cultured retinal glia. ISH with digoxigenin (DIG)-labeled RNA antisense probes for TNFa and IL-6 was used to detect cytokine mRNA in cultured retinal glial cells. (A) Exposure of retinal glia (passage 122) to HSV-1 (moi = 0.1) results in intense positive hybridization to a DIGlabeled antisense probe for TNFa mRNA within 1 hour. Representative hybridization at 8 hours is shown. (B) Unstimulated glia (passage 122) hybridize more weakly with the same DIG-labeled antisense probe for TNF-a mRNA, (C) Similar results are obtained under the same conditions using a DIG-labeled antisense probe specific for IL-6 mRNA. (D) Unstimulated glia of the same passage number exhibit low levels of hybridization to the IL-6 probe. Sense probes for both cytokines resulted in totally negative hybridization, and noncounterstained cells could be visualized only under phase contrast (not shown). Similar results were obtained with cells of other passage number and with independentiy prepared probes. Original magnification, X250. or HSV-1. Levels of TNFa and ILr6 mRNA were upregulated over time by exposure to HSV-1 and IFNy. Upregulation was rapid because, within 1 hour of exposure to either IFNy (50 U/ml) or HSV-1 (moi = 10), increased transcript levels for TNFa (Fig. 3) and IL-6 (Fig. 4) were detected. Earlier times were not tested. HSV-1 and IFNy had similar effects on TNFor mRNA levels (Fig. 3). At 1 hour after exposure to any stimulant, modest (3- to 4-fold) increases over basal levels of TNFa mRNA were detected. mRNA levels of TNFa peaked within 4 hours of exposure to IFNy or HSV-1 (moi = 10). Maximal changes in mRNA were increased 5.9- and 10.7-fold by IFNy and HSV-1, respectively, over control levels. Transcript levels for both cytokines remained slightly elevated at 24 hours. The low dose of HSV-1 (moi = 0.1) caused TNFa transcript levels to reach maximal levels more slowly Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 (8 hours) with lower peak levels of 6.2-fold above baseline. Herpes simplex virus type 1 (moi = 10) induced the most rapid upregulation of IL-6 mRNA levels, with a 5.6-fold increase in signal detected within 1 hour of in vitro infection (Fig. 4). Maximum transcript levels were detected at 8 hours, with a 9-fold increase in IL6 mRNA message. Changes in transcript levels after exposure to a lower moi of 0.1 of HSV-1 or rIFNy also peaked 8 hours after stimulation, but with more modest increases (5.6- and 4.2-fold respectively; not shown). The levels of IL-6 transcripts detected in cultures remained elevated through 12 hours after exposure to the higher dose of HSV-1 (4.5-fold over baseline) . Cells exposed to IFNy or low-dose HSV-1 continued to transcribe IL-6 at levels approximately two to three times higher than controls through 12 hours. Transcript levels for ILr6 began to drop 8 to 12 hours Production of TNFa and IL-6 by Retinal Glia TNFa 300 bp IL-6 426 bp FIGURE 2. Baseline transcription of tumor necrosis factor-a (TNFa) and interleukin-6 (IL-6) in retinal glia in vitro. A representative ethidium bromide-stained gel containing amplified products from reverse transcription-polymerase chain reaction (RT-PCR) is shown for cultured glia. TNFa and IL-6 transcripts were detected in cultured retinal glia regardless of in vitro stimulation. Lanes marked ( + ) indicate RNA, which was reverse transcribed. Those marked (—) indicate omission of reverse transcriptase from the reactions. Ten percent of the reaction product was run on each gel. Pairs of lanes designated 1 to 5 contained cDNA generated from RNA extracted from unstimulated cells, cells infected with herpes simplex virus type 1 (moi = 10), cells co-cultured with IFNy (50 U/ml), cells co-cultured with lipopolysaccharide (10 ^tg/ml), or a known positive sample amplified in the same assay. All cDNA in this screening RT-PCR were amplified for 35 cycles. Identity of PCR products obtained with these primer pairs was confirmed by predicted size determined from DNA ladders and Southern blotting and hybridization with internal sequence probes for each cytokine.92"34 Cells were exposed to virus or other stimulants for 8 hours. after exposure to HSV-1. This shift presumably coincided with host cell shutdown to a viral infection,33 although at 24 hours exposure, IL-6 transcripts remained approximately two times higher than basal levels. Localized Intraocular Changes in Cytokine mRNA Levels Induced by Anterior Segment Inflammation Screening RT-PCR was used to determine whether IL-6 and TNFa mRNA could be detected in normal retinas from BALB/c eyes and to test whether in vivo ocular HSV-1 infection by anterior chamber injection altered mRNA levels of these cytokines in retina. To localize these cytokines better, the neural retina was isolated from both the retinal pigment epithelium and the anterior segment because the latter would be expected to yield positive cytokine signals under some conditions.24"27 Both IL-6 and TNFa were detected in retinas from uninjected, normal control eyes. Uninjected, normal eyes also exhibited IL-6 transcripts in the anterior segment, but no TNFa transcripts were detected in the absence of inflammation or infection Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 2307 (Fig. 5). In view of the constitutive expression of TNFa and IL-6 in normal retinas, we performed RT-PCR on RNA extracted from isolated retinas under several experimental conditions. After the injection of buffer or HSV-1 into the anterior chamber of the eye, increased TNFa (Fig. 6) and IL-6 (Fig. 7) transcript levels were detected compared to normal tissue. Comparisons of HSV-1- and HBSS-injected eyes allowed us to distinguish between changes in retinal cytokine transcript levels caused by nonspecific and infectioninduced inflammation. Injection of live HSV-1 into the anterior chamber of the eyes of BALB/c mice resulted in larger relative increases in the mRNA levels of TNFa and IL-6 at all time points studied when compared to buffer-injected eyes. Although only minimal changes in TNFa and IL-6 were seen at day 1 after injection of either HBSS or HSV-1, by day 3, transcript levels of both cytokines were increased 5.1- and 2.1fold, respectively, over the levels observed in HBSSinjected eyes (Figs. 6, 7). Buffer alone increased transcript levels of TNFa and IL-6 above those seen in 15 n IFNy HSV FIGURE 3. Tumor necrosis factor-a (TNFa) transcript levels in retinal glia increase after in vitro exposure to herpes simplex virus type 1 (HSV-1) or rIFNy. Reverse transcription-polymerase chain reaction for TNFa upregulation in cultured retinal glia at several time points after exposure to HSV-1 (shaded bars) or IFN7 (open bars). Glial cells (passage 16) were stimulated and the total cellular RNA extracted at the times indicated. After random hexamer priming in the reverse transcription, cDNA was amplified for 27 cycles with TNFa-specific primers and subjected to Southern slot blot analysis with digoxigenin-labeled oligonucleotide probes.'12"34 cDNA levels in the samples were standardized to serial dilutions of the housekeeping gene, HPRT, as described in Materials and Methods. Results are from one of two independent experiments performed with duplicate slots of each of three serial dilutions prepared from the same cDNA.34 Results for each time point are expressed as the mean-fold increase in signal over unstimulated controls as measured by densitometry of two serial dilutions of each cDNA duplicate (±SEM). Investigative Ophthalmology & Visual Science, October 1996, Vol. 37, No. 11 2308 associated immune deviation for many years." l h " C However, the potential of the retinitis model for the if) study of retinal cell responses to anterior segment in-H '— 8 fection and associated inflammation has not been exIB IFNy ploited fully. The injection of HSV-1 into the anterior HSV i 6chamber of the eye of a BALB/c mouse results in large amounts of virus in the anterior segment of injected eyes but only the transient, focal appearance of % ro viral antigens in the retinas of virus-injected eyes.'" " Despite this transient presence of virus, the retinas of 33 HSV-1-injected eyes do not develop retinal necrosis, o and they remain intact indefinitely. Within 1 day of anterior chamber injection of either buffer or live vi8 24 12 Hours rus, Muller cells, a major nonneural population within the retina, upregulate their expression of GFAP. HSVFIGURE 4. lnterleukin-6 (IL-6) transcript levels are upregu1 injection induces more intense and prolonged exlated in cultured retinal glia by herpes simplex virus type pression of GFAP than is observed after the injection 1 (HSV-1) infection or stimulation by rlFNy. Quantitative of buffer,' suggesting that Muller cells respond differreverse transcription-polymerase chain reaction (RT-PCR) entially to limited nonspecific inflammation versus demonstrates HSV- and IFNy-mediated upregulation of ILprogressive and persistent virus-induced injury and in6 mRNA levels in cultured retinal glia. Retinal glia were exposed to rlFNy (50 U/ml; open bars) or HSV (moi = 10; flammation. The significance of GFAP upregulation shaded ban), and total cellular RNA was isolated at 1, 4, 8, for other Muller cell functions is unknown. However, 12, and 24 hours before RT-PCR assay, followed by slotbased on the rapid GFAP upregulation in the retinitis blotting and densitometric analyses. See Figure 3 and the model, we hypothesize an intraretinal contribution by Materials and Methods section for experimental details. Muller cells to retinal preservation by their soluble mediator production. _ 10 n X 1 normal, uninjccted retinas. However, only IL-6 transcript levels showed large increases (2.4- and 6.3-fold, respectively). By day 7 after injection with HSV-1, TNFa transcript levels returned toward those seen in control mice, whereas IL-6 remained elevated. Transcript levels at later times were not assessed. DISCUSSION Our studies demonstrate that relative transcript levels of IL-6 and TNFa are upregulated rapidly in intact retina and cultured retinal glial cells after exposure to HSV-1. Transcripts for both cytokine genes were detected in the retinas of normal BALB/c mice, indicating that one or more resident retinal cell populations produces these mediators. The combined ISH and RT-PCR results with cultured cells support the potential of Muller cells as one source of these intraretinal soluble mediators. To our knowledge, this is the first report of proinflammatory cytokine transcription by cells in the normal murine retina and of increased cytokine transcripts in retinas from murine eyes experiencing anterior segment HSV-1 infection or nonspecific inflammation. These results with cultured retinal glia suggest that our previous observations—that these cells secrete TNFa and IL-6 after exposure to several stimuli2*—were a result of active transcriptional and translational responses. Anterior chamber injection of HSV-1 has been used for studies of virus-induced anterior chamber Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 Glial fibrillary acidic protein is not upregulated in TNFa 300 bp IL-6 426 bp 1 i 2 i 3 I 4 '5 FIGURE 5. Reverse transcription-polymerase chain reaction (RT-PCR) to detect tumor necrosis factor-a and interleukin-6 transcripts in murine eyes. Total RNA was extracted from isolated anterior segments (first lane in lane pairs 1 to 5) or retinas (second lane in lane pairs 1 to 5) of eyes obtained from normal mice (lane pair 1), eyes receiving anterior chamber injections of 4 //I of Hank's balanced salt solution (lane pair 2), eyes receiving 2 X 10' pfu of UVinactivated herpes simplex virus type 1 (HSV-1) (lane pair 3) or live HSV-1 (lane pair 4), or known positive RNA for each cytokine. For the ethidium bromide-stained gel shown, samples represent both eyes from four mice per treatment injected and removed on the same day, 24 hours after injection. cDNA from all samples was amplified for 35 cycles. Nonreverse transcribed RNA from each group of mice were run in parallel for each experiment to confirm the absence of DNA (not shown). Identities of PCR products obtained with tissue extracts were confirmed by molecular weight and Southern blotting with digoxigenin-labeled internal sequence probes1-' ™ as described in the Materials and Methods section. 2309 Production of TNFa and IL-6 by Retinal Glia ter signal transduction events.'" It will be interesting to determine whether TNFa or IL-6 is dysregulated +(. 12 H in models of retinal degeneration in which increased apoptosis has been reported." ''' These cytokines and TO 10 _L <n neurotrophic factors have documented, complex inI HBSS £ 8^ I HSV-1 terrelationships of positive and negative feedback I loops that most likely occur at the level of receptordependent signal transduction (e.g., ref. 50). Events such as infection, which alter the balance of this circuitry, would be expected to influence multiple tissue responses, including the production of additional soluble mediators and surface antigens. 1 2 3 4 5 6 7 TNFa and II.-6 are multifunctional cytokines and Days Post-injection are among several soluble mediators associated with ocular inflammatory conditions, including uveitis in FIGURE 6. Anterior chamber injection of herpes simplex vihumans and in experimental models.'"'-'2™'-'" Studies rus type 1 (HSV-1) and Hank's balanced salt solution upregby Planck et al2' and Yoshida et al'7 demonstrated that ulates tumor necrosis factor-a (TNFa) mRNA levels in intact TNFa and IL-6 were transcribed in the iris and ciliary nuirine retina. Quantitative reverse transcription-polymerbody of rats. Transcription of both cytokines was inase chain reaction with TNFa-specific primers was performed on RNA isolated at the times indicated from whole duced in the rat neuroretina after endotoxin injecretinas after anterior chamber inoculations of nonspecific tion, but the cellular source(s) was not identified.27 (buffer; open bars) or viral stimuli (HSV-1, 2 X W pfu/4 //I; TNFa has antiviral capacity alone* and in synergy with shaded bars). Results are expressed as mean-fold increase IFN/9 or IFNy,'""' either directly by killing of infected in TNFa signal over uninjected controls as measured by cells or indirectly by making cells refractory to infecdensitometry (±SEM) and as described in Figure 3. Similar tion. TNFa can be induced or upregulated on exporesults were obtained in two independent experiments, each sure to viruses, including human immunodeficiency of which was performed in duplicate using four pooled reuvirus and measles.17'12 Though antiviral functions are nas per time point (two mice). more speculative for IL-6 than for TNFa, the ability „ LJJ W 14 1 Miiller cells of contralatcral, uninjected eyes of the retinitis model except where there is retinal necrosis.' This suggests that a signal(s) from the inflamed anterior segment may activate Miiller cells before entry of HSV-1 into ipsilateral retinas to provide a component of retinal protective responses. For example, a small amount of IFNy from the inflamed anterior chamber may reach the retina to synergize with locally produced cytokines. Protein levels of IL-6 and TNFa are influenced by IFNy, a product of natural killer cells and T cells.1'" The latter cells are abundant in the anterior segment of HSV-1-injected eyes, and natural killer cells are present in the normal or infected retina and may modulate Miiller cell responses.4'" "' The precise roles of IL-6 and TNFa in the retina have not been investigated as extensively as in the CNS,1718'4'"'" but these pleiotropic cytokines may have roles in retinal homeostasis as well as during ocular inflammation. Several in vitro studies showed that TNFa and IL-6 induce secretion of nerve growth factor by CNS Conversely, nerve growth factor inastrocytes. duces IL-6 transcription in astrocytes.1' Nerve growth factor and ciliary neurotrophic factor are neurotrophic factors produced in the retina by Miiller cells.>l '' These reports are compatible with indirect roles for Miiller cells in retinal repair mechanisms through production of these cytokines, which may al- Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 15 -i LU 1 ra I 10 HBSS HSV-1 c I T JL 5- T T3 O 1 2 3 4 5 6 n 7 Days Post-injection FIGURE 7. Evidence for in vivo increases in interleukin-6 (IL-6) transcript levels in retinas after anterior chamber inoculation of herpes simplex virus type 1 (HSV-1) or Hank's balanced salt solution. Quantitative reverse transcription polymerase chain reaction with 11^6-specific primers for amplification of cDNA was performed as described in Figures 3 and 4. Nonspecific inflammation induced by buffer alone (open bars) resulted in substantial increases in Ilv6 transcript levels, which were not observed for tumor necrosis factor a. Temporal changes in IL-6 transcript levels in retinas from HSV-1 injected eyes are shown (shaded ban). Similar results were obtained in two independent experiments. See Materials and Methods section for experimental details. 2310 Investigative Ophthalmology & Visual Science, October 1996, Vol. 37, No. 11 of IL-6 and TNFa to synergize7>b3'b4 might promote antiviral responses. We recently determined that type 1 interferons also are induced in Miiller cells, and these mediators also limit viral replication 65 (also Drescher and Whittum-Hudson, manuscript submitted). Though not revealing absolute transcript amounts in tissue extracts, the RT-PCR assay, combined with Southern slot blot analysis, demonstrates relative increases in TNFa and IL-6 transcript levels over time after the induction of nonspecific or virus-induced anterior segment inflammation compared to normal BALB/c retinal tissue. It is important to note that this assay allows analysis of transcript changes in an individual cytokine relative to controls, but no conclusions can be made regarding the mRNA levels for the two cytokine genes relative to each other. Although these studies do not discriminate between increased rate of transcript production and increased stability of existing transcripts, the former seems the most likely explanation for the results obtained. Coordinate regulation of the two cytokines occurs because it has been shown that TNFa upregulates IL-6, which can then downregulate TNFa production. 14 " 10 In the current studies, peak levels of TNFa message were detected in cultured retinal glia after 4 hours of exposure to HSV-1 or IFNy; peak IL-6 transcript levels occurred later, at 8 hours. This temporal sequence is consistent with the known interdependence of these two mediators.21 '42>b4 There appear to be additional controls over TNFa and IL-6 transcription in retinas of HSV-injected eyes because transcript levels of both mediators clearly decline within 7 days in vivo. In addition to negative feedback inhibition, the control may be passive in that the stimulus is gone (e.g., inactivation of virus, apoptosis, or lysis of infected cells). Both IL-6 and TNFa are regulated at the transcriptional and the posttranslational levels. Observed increases in cytokine transcript levels do not of themselves indicate increased protein synthesis and secretion by retinal glia. However, the increases are consistent with our previous studies showing that protein synthesis and secretion of both TNFa and IL-6 are upregulated two to three times within 24 hours of in vitro exposure to either HSV-1 or IFNy.28 Responses in these cytokines to HSV-1 and IFNy were similar under the experimental conditions used. IL-6 secretion was slightly greater than TNFa before and after stimulation. Secretion of both cytokines was most dramatically increased by co-culture with lipopolysaccharide (10-fold).28 The presence of detectable basal transcript levels of IL-6 and TNFa in intact, normal retinas and in cultured retinal glial cells suggests that protein levels could be upregulated rapidly. Our data strongly support the participation of resident retinal cells by soluble mediator production during responses to an- Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 terior segment injury or inflammation. It remains unclear whether the changes in TNFa and IL-6 in neuroretina relate to protective or pathologic responses by the host. Studies are planned to investigate these questions further. Key Words cytokines, glia, herpes simplex virus type 1 (HSV-1), interleukin-6, mouse, Muller cells, retina, tumor necrosis factor-a Acknowledgments The authors thank Drs. Robert A. Prendergast and Alan P. Hudson for their helpful reading of this manuscript, Mr. Donnell Berry and Mr. Timothy Conn for technical assistance, and Mrs. Irene Skop for assistance in proofreading this manuscript. References 1. Whittum-Hudson JA. HSV-induced stimulation of retinal Muller cells in vivo. ARVO Abstracts. Invest Ophthalmol Vis Sri. 1992;33(suppl):786. 2. Bignami A, Dahl D. The radial glia of Muller in the rat retina and their response to injury: An immunofluorescence study with antibodies to the GFAP. Exp Eye Res. 1979; 28:63-69. 3. Sarthy PV, Fu M. Transcriptional activation of an intermediate filament protein gene in mice with retinal dystrophy. DNA. 1989; 8:437-446. 4. Ekstrom P, Sanyal S, Narfstrom K, et al. Accumulation of glial fibrillary acidic protein in Muller radial glia during retinal degeneration. Invest Ophthalmol Vis Sri. 1988;29:1363-1371. 5. Erickson PA, Fisher SK, Guerin CJ, et al. Glial acidic fibrillary protein increases in Muller cells after retinal detachment. Exp Eye Res. 1987;44:37-46. 6. Eng LF. Glial fibrillary acidic protein (GFAP): The major protein of glial intermediate filaments in differentiated astrocytes. / Neuroimmunol. 1985; 8:203-214. 7. Lieberman AP, Pitha PM, Shin HS, Shin ML. Production of TNF and other cytokines by astrocytes stimulated with lipopolysaccharide or a neurotropic virus. Proc Natl Acad Sri USA. 1989; 86:6348-6352. 8. Chung IY, Benveniste EN. Tumor necrosis factor-a production by astrocytes: Induction by lipopolysaccharide, IFN-y, and IL-10. J Immunol. 1990; 144:29993007. 9. Roberge FG, Caspi RR, Chan C, et al. Long-term culture of Muller cells from adult rats in the presence of activated lymphocytes/monocytes products. Exp Eye Res. 1985; 4:975-982. 10. Sher A, Coffman RL, Hieny S, et al. Interleukin-5 is required for the blood and tissue eosinophilia but not granuloma formation induced by infection with Schistosoma mansoni. Proc Natl Acad Sri USA. 1990; 87:61-65. 11. Yamamura M, Uyemura K, Deans RJ, et al. Defining protective responses to pathogens: Cytokine profiles in leprosy lesions. Srience. 1991; 254:277-279. 12. Haanen JB, deWaal Malefijt R, Res PC, et al. Selection Production of TNFa and IL-6 by Retinal Glia 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. of a human T helper type 1-like T cell subset by mycobacteria./£x/> Med. 1991; 174:583-592. Kemp M, Hey AS, Kurtzhals JA, et al. Dichotomy of the human T cell response to Leishmania antigens: I: Th 1-like response to Leishmania major promastigote antigens in individuals recovered from cutaneous leishmaniasis. Clin Exp Immunol. 1994;96:410-415. Beutler B, Cerami A. The biology of cachectin/TNF— a primary mediator of the host response. Ann Rev Immunol. 1989; 7:625-655. Van SnickJ. Interleukin-6: An overview. Ann Rev Immunol. 1990; 8:253-278. Kishimoto T, Taga T, Akira S. Cytokine signal transduction. Cell. 1994; 76:253-262. Genis P, Jett M, Bernton EW, et al. Cytokines and arachidonic metabolites produced during human immunodeficiency virus (HIV)-infected macrophageastroglia interactions: Implications for the neuropathogenesis of HIV disease. J Exp Med. 1992; 176:17031718. Vitkovic L, Wood GP, Major EO, Fauci AS. Human astrocytes stimulate HIV-1 expression in a chronically infected promonocyte clone via interleukin-6. AIDS Res Hum Retroviruses. 1991; 7:723-727. LaVail MM, Unoki K, Yasumura D, et al. Multiple growth factors, cytokines, and neurotrophins rescue photoreceptors from the damaging effects of constant light. ProcNatlAcad Sd USA. 1992;89:11249-11253. Streilein JW, Bradley D. Analysis of immunosuppressive properties of iris and ciliary body cells and their secretory products. Invest Ophthalmol Vis Sd. 1991; 32:2700-2710. DeVos AF, VanHaren MAC, Verhagen C, et al. Tumour necrosis factor-induced uveitis in the Lewis rat is associated with intraocular interleukin-6 production. Exp Eye Res. 1995; 60:199-207. Streilein JW, Cousins SW. Aqueous humor factors and their effect on the immune response in the anterior chamber. Curr Eye Res. 1990;9:175-182. Roberge FG, Caspi RR, Nussenblatt RR. Glial retinal Miiller cells produce IL-1 activity and have a dual effect on autoimmune T helper lymphocytes: Antigen presentation manifested after removal of suppressive activity. / Immunol. 1988; 140:2193-2196. Pasquale LR, Dorman-Pease ME, Lutty GA, et al. Immunolocalization of TGF/51, TGF/32 and TGF/33 in the anterior segment of the human eye. Invest Ophthalmol Vis Sd. 1993; 34:23-30. Kuppner MC, McKillop-Smith S, Forrester JV. TGFP and IL-1/3 act in synergy to enhance IL-6 and IL-8 mRNA levels and IL-6 production by human retinal pigment epithelial cells. Immunology. 1995; 84:265271. Planck SR, Dang TT, Graves D, et al. Retinal pigment epithelial cells secrete interleukin-6 in response to interleukin-1. Invest Ophthalmol Vis Sd. 1992; 33:78-82. Elner VM, Strieter RM, Elner SG, et al. Neutrophil chemotactic factor (IL-8) gene expression by cytokinetreated retinal pigment epithelial cells. Am J Pathol. 1990; 136:745-750. Drescher KM, Whittum-Hudson JA. Modulation of Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 2311 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. immune-associated surface markers and cytokine production by murine retinal glial cells. / Neuroimmunol. 1996;64:71-81. Merges MJ, Whittum-Hudson JA. In vitro susceptibility of murine retinal cells to herpes simplex virus type 1. Invest Ophthalmol Vis Sd. 1990;31:1223-1230. Politi L, Lehar M, Adler R. Development of neonatal mouse retinal neurons and photoreceptors in low density cell culture. Invest Ophthalmol Vis Sd. 1988; 29:534-543. WhittumJA, McCulleyJP, NiederkornJY, Streilein JW. Ocular disease induced in mice by anterior chamber inoculation of herpes simplex virus. Invest Ophthalmol VisSd. 1984;25:1065-1073. Murray L, Martens C. Abnormal T cells from lpr mice down-regulate transcription of interferon-y and tumor necrosis factor-a in vitro. Cell Immunol. 1990; 126:367-376. Montgomery RA, Dallman MJ. Analysis of cytokine gene expression during fetal thymic ontogeny using the polymerase chain reaction. / Immunol. 1991; 147:554-560. Wynn TA, Eltoum I, Cheever AW, et al. Analysis of cytokine mRNA expression during primary granuloma formation induced by eggs of Schistosoma mansoni. J Immunol. 1993; 151:1430-1440. Ginsberg HS. Herpesviruses. In: Dulbecco R, Ginsberg HS, eds. Virology. Philadelphia: JB Lippincott; 1988: 161-178. WhittumJA, NiederkornJY, McCulleyJP, Streilein JW. Intracameral inoculation of HSV type 1 induces anterior chamber associated immune deviation. Curr Eye Res. 1983;2:691-697. Whittum-Hudson J, Merges MJ. Variations in herpes simplex-induced systemic immunity after intracameral injection of DNA polymerase mutants. ARVO Abstracts. Invest Ophthalmol Vis Sd. 1987; 28:241. Whittum-Hudson JA, Merges MJ, Field HJ. Immunogenicity versus pathogeniciry after anterior chamber inoculation of an acyclovir-induced double mutant of HSV-1. Curr Eye Res. 1987;6:1459-1470. Cousins SW, Gonzalez A, Atherton SS. Herpes simplex retinitis in the mouse: Clinicopathologic correlations. Invest Ophthalmol Vis Sd. 1989; 30:1485-1494. Atherton SS, Altman NH, Streilein JW. Two waves of virus following anterior chamber inoculation of HSV1. Invest Ophthalmol Vis Sd. 1987; 28:571-579. Whittum-Hudson JA, Pepose JS. Immunologic modulation of virus-induced pathology in a murine model of acute herpetic retinal necrosis. Invest Ophthalmol Vis Sd. 1987;28:1541-1548. Gillis S. T-cell-derived lymphokines. In: Paul WE, ed. Fundamental Immunology. New York: Raven Press; 1989:621-638. Tamesis RR, Messmer EM, Rice BA, et al. The role of natural killer cells in the development of herpes simplex virus type 1 induced stromal keratitis in mice. Eye. 1994;8(pt3):298-306. Hendricks RL, Weber PC, Taylor JL, et al. Endogenously produced interferon-a protects mice from her- Investigative Ophthalmology 8c Visual Science, October 1996, Vol. 37, No. 11 2312 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. pes simplex type 1 corneal disease. J Gen Virol. 1991;72:1601-1610. El-Asrar AM, Maimone D, Morse PH, Lascola C, Reder AT. Interferon-gamma and tumour necrosis factor induce expression of major histocompatibility complex antigen on rat retinal astrocytes. BrJ Ophthalmol. 1991;75:473-475. Brandt CR, Salkowski CA. Activation of NK cells in mice following corneal infection with herpes simplex virus type 1. Invest Ophthalmol Vis Sci. 1992; 33:113120. Lotan M, Schwartz M. Cross talk between the immune system and the nervous system in response to injury: Implications for regeneration. FASEBJ. 1994;8:10261033. Benveniste EN. Inflammatory cytokines within the central nervous system: Sources, function, and mechanism of action. AmJPhysiol. 1992;263:C1-C16. Merrill JE, Jonakait GM. Interactions of the nervous and immune systems in development, normal brain homeostasis, and disease. FASEBJ. 1995;9:611-618. Benveniste EN, Benos DJ. TNF-a- and IFN-y-mediated signal transduction pathways: Effects on glial cell gene expression and function. FASEBJ. 1995;9:1577-1584. Chakrabarti S, Sima AA, Lee J, et al. Nerve growth factor (NGF), proNGF and NGF receptor-like immunoreactivity in BB rat retina. Brain Res. 1990;523:ll15. Selmaj K, Raine CS, Cannella B, Brosnan CF. Identification of lymphotoxin and tumor necrosis factor in multiple sclerosis lesions./ Clin Invest. 1991;87:949954. Hattori A, Tanaka E, Murase K, et al. Tumor necrosis factor stimulates the synthesis and secretion of biologically active nerve growth factor in non-neuronal cells. JBiolChem. 1993; 268:2577-2582. Wen R, Song Y, Cheng T, et al. Injury-induced upregulation of bFGF and CNTF mRNAs in the rat retina. / Neurosd. 1995; 15:7377-7385. Portera-Cailliau C, Sung OH, Nathans J, Adler R. Apoptou'c photoreceptor cell death in mouse models Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. of retinitis pigmentosa. Proc Natl Acad Sci USA. 1994;91:974-978. Lolley RN, Rong H, Craft CM. Linkage of photoreceptor degeneration by apoptosis with inherited defect in phototransduction. Invest Ophthalmol Vis Sci. 1994; 35:358-362. Yoshida M, Yoshimura N, Hangai M, et al. Interleukinla, interleukin-1/?, and tumor necrosis factor gene expression in endotoxin-induced uveitis. Invest Ophthalmol Vis Sci. 1994;35:1107-1113. Rossol-Voth R, Rossol S, Schutt KH, et al. In vivo protective effect of tumor necrosis factor-a against experimental infection with herpes simplex virus type l.JGen Virol. 1991; 72:143-147. Wietzerbin J, Gaudelet C, Catinot L, et al. Synergistic effect of interferon-gamma and tumor necrosis factoralpha on antiviral activity and (2'-5') oligo (A) synthetase induction in myelomonocytic cell line. JLeukoc Biol. 1990; 48:149-155. Chen S, Oakes JE, Lausch RN. Synergistic anti-HSV effect of tumor necrosis factor-a and interferon gamma in human corneal fibroblasts is associated with interferon-/? induction. Antiviral Res. 1993;22:15-29. Schijns VECJ, Van der Neut R, Haagmans BL, et al. Tumor necrosis factor-a, interferon-gamma and interferon-/? exert antiviral activity in nervous tissue cells. J Gen Virol. 1991;72:809-815. Schneider-Schaulies J, Schneider-Schaulies S, ter Meulen V. Differential induction of cytokines by primary and persistent measles virus infections in human glial cells. Virobgy. 1993; 195:219-228. Frei K, Malipiero UV, Leist TP, et al. On the cellular source and function of interleukin-6 in the central nervous system in viral disease. Eur J Immunol. 1989; 19:689-694. Maimone D, Cioni C, Rosa S, et al. Norepinephrine and vasoactive intestinal peptide induce IL-6 secretion by astrocytes: Synergy with IL-1/? and TNF~a.JNeuroimmunol. 1993; 47:73-82. Drescher KM. Glial cells of Miiller as intraretinal immunomodulators. Baltimore: Johns Hopkins University; 1995. Thesis.