Download Herpes simplex virus type 1 alters transcript levels of tumor

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.