| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-Induced Indoleamine 2,3 Dioxygenase and Inducible Nitric Oxide Synthase in the Replication of Human Cytomegalovirus in Retinal Pigment Epithelial Cells1
*
Unité d’Immunologie Virale, Institut Pasteur, Paris, France; and
Institut National de la Santé et de la Recherche Médicale, U450, Developpement, Vieillissement et Pathologie de la Rétine, Paris, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IFN-ß were potent inhibitors of CMV replication in
RPE cells, while TNF-
, IL-1ß, or TGF-ß2 did not affect viral
replication. Inhibition by IFN-, and to a lesser extent IFN-ß, was
almost completely reversed by addition of L-tryptophan to
the culture medium, strongly implicating the indoleamine 2,3
dioxygenase (IDO) pathway. Polyadenylated IDO mRNA accumulation was
detected as early as 2 h after IFN stimulation. Furthermore, CMV
blocked the production of nitric oxide by the inducible form of nitric
oxide synthase. This inhibition depended on a functional viral genome.
However, exogenous nitric oxide significantly inhibited viral protein
expression in RPE cells. Thus, CMV infection blocks the inducible
nitric oxide synthase pathway activated by IFN- and IL-1ß, but
cannot counteract the IFN-induced IDO pathway, which ultimately
controls its replication in primary human RPE
cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced up-regulation of class II
Ags on human retinal pigment epithelial
(RPE)3 cells, thereby
altering their characteristics as APCs (5). Fas ligand, which is
constitutively expressed on the neurosensory retina and on RPE cells,
also plays an important role (6). Human retinal pigment epithelium is a
multifunctional and strategic component of the retina (7). It forms
part of the blood retinal barrier, the other part being at the
level of retinal blood vessels. The pigment epithelium allows
continuous renewal of the outer segments of photoreceptors via
specialized phagocytosis. CMV retinitis, one of the most common opportunistic infections in HIV-infected patients (8, 9), manifests itself when the patient’s CD4+ lymphocyte counts are less than 100/mm3. The pathophysiology of this retinal disease remains poorly understood. The onset of CMV retinitis presumably results from alterations in the ocular immune privilege status and/or in the antiviral mechanisms effective until retinal infection occurs. All 10 layers of the retina are sites of necrotic lesions. However, CMV seems to replicate efficiently only in RPE and glial cells (10, 11). It has been shown that both primary cell types are fully permissive for CMV infection in vitro, even if the cycle seems to be protracted compared with that in routinely used targets for CMV replication such as diploid fibroblasts (12, 13).
Issues concerning the influence of cytokines on viral disease
progression are still under debate. The occurrence of CMV disease in
the late phase of immunosuppression may be related to perturbation of
cytokine production and secretion, associated with the progressive loss
of the CD3+CD8+ cell subset (14). During ocular
inflammation, inflammatory mediators, such as IFN- and TNF-, are
detected in the vitreous humor and the retina of HIV-infected patients
presenting with CMV retinitis (15, 16).
The antiviral effects of IFNs have been principally attributed to 1)
the phosphorylation of eukaryotic protein synthesis initiation factor
eIF-2 by P1 kinase; 2) the activation of a latent endoribonuclease
by the end products of 2',5'-oligoadenylate synthetase; and finally 3)
expression of Mx proteins during infections with influenza and
vesicular stomatitis viruses (17, 18). Previous studies have shown that
IFN- and IFN-ß inhibit CMV replication in vitro and in vivo
(19, 20, 21, 22). However, the mechanisms of this inhibition are still unclear.
In human and bovine retina, IFN- can induce the expression of
indoleamine 2,3 dioxygenase (IDO) (23, 24) and, in combination with
IL-1ß, the inducible form of nitric oxide synthase (NOSII) (25). IDO
is responsible for conversion of tryptophan and other indole
derivatives to kynurenine (26), thus explaining the inhibitory effect
of IFN on many intracellular organisms, such as Toxoplasma
gondii and Chlamydia psittaci (27, 28). NOSII has
been shown to have antiviral effects in different models (29, 30, 31, 32, 33, 34, 35). Much
evidence has previously highlighted the role played by the inducible
form of NOSII in ocular inflammatory and infectious disorders (36).
This enzyme is responsible for the synthesis of nitric oxide (NO) from
L-arginine. At least two major retinal components express
NOSII. It has been shown that CMV-infected retinal glial cells from
HIV-infected patients express NOSII (37). Human RPE cells are another
major source of NOSII in vitro and possibly in vivo, even though enzyme
expression is easily masked by the large amounts of pigment present in
the cells (25, 26).
In the current study, we analyzed the influence on human CMV
replication of different cytokines known to be present in the eye under
normal conditions or during different inflammatory disorders and
infectious diseases, and which have been reported to affect CMV
replication in vivo or in vitro. It was possible to show that
inhibition of CMV replication by IFN-, and less markedly by IFN-ß,
involves the IDO pathway. Furthermore, exogenously added nitric oxide
(NO) showed a clear, antiviral effect in RPE cells, but CMV infection
inhibited endogenous NO production by NOSII.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The tenets of the Declaration of Helsinki were followed, and approval was obtained from the Office of Human Subjects Research.
Cell cultures
Primary human RPE cell lines were established from donor eyes. Human eyes were obtained from the Bristol Eye Bank (a generous gift from Dr. M. Nash, Oxford, U.K.) after removal of corneas for transplantation. RPE cells were isolated by trypsinization, resuspended in HAM-F10 (Life Technologies, Cergy Pontoise, France) with 10% heat-inactivated FCS, and transferred to a 25-cm2 flask. RPE cells form monolayers like epithelial cells and at confluency are typically hexagonal. Homogeneity was confirmed by positive immunostaining with mAbs to cytokeratins (38). Cultures from three different primary cell lines were used at passages 3–11 for all experiments. All cells were routinely tested for mycoplasma and found to be negative.
Virus
Human CMV strain AD169 was used throughout. Stocks of CMV were generated in human foreskin fibroblasts, and titers were determined by plaque formation under 0.64% carboxymethyl cellulose, as previously described (39). Titration was performed for all stocks using both passive particle adsorption and centrifugal enhancement (40); centrifugation usually resulted in a 0.5–1 log enhancement of virus titers compared with passive adsorption. UV inactivation of CMV (6 x 105 joules) resulted in a reduction of the viral titer from 2 x 106 to less than 50 plaque-forming units/ml. Inactivated virus was tested for its capacity to induce IE proteins and found to be negative 24 h after virus adsorption. Heat-inactivated virus (56°C, 30 min) and UV-irradiated CMV were used at the same multiplicities of infection (MOI), based on preinactivation titers of the viral stock. For all experiments, virus was adsorbed to cells for 2 h at 37°C. Cells were then washed twice with PBS and refed complete medium. Infection times are indicated in Results. All virus stocks were negative for the presence of mycoplasma.
Chemicals
Human rIFN- and rIL-1ß were purchased from Peprotech, TEBU
(France). Human rTNF- was provided by the MRC ADP reagent
program. Human rIFN-ß was purchased from Biosource (Rungis, France).
Human rTGF-ß2 was purchased from R&D Systems (Abington, U.K.).
A stock solution of 3 µg/ml was prepared in 4 mM HCl containing 0.1%
BSA. NG-monomethyl L-arginine
(L-NMMA) and diethylamine dinitric oxide complex (DEANO)
were obtained from Alexis Laboratories (COGER, Paris, France).
L-tryptophan was purchased from Sigma (St.
Quentin-Fallavier, France).
Protocol of infection
Confluent RPE cells were infected in eight-well glass chamber slides (Labteks, NUNC, Naperville, IL) for immunofluorescence, in 24-well plates for Western blot analysis, and in 25-cm3 culture flasks for RT-PCR analysis. High MOI (1, 2, 3, 4, 5) was used to study the effects of cytokines, whereas low MOI (0.1–0.01) was used to study the antiviral effects of organic NO donors. After a 2-h viral adsorption at 37°C in HAM-F10 without FCS, cells were washed twice with PBS and refed HAM-F10 supplemented with 10% FCS. Cytopathic effects were observed daily by light microscopy. At 5, 10, and 15 days postinfection, supernatants were collected for viral titration. Labtek slides were fixed at 2, 3, and 5 days for immunofluorescence. Proteins were extracted at the same times for Western blot analysis.
Effects of cytokines on CMV replication and viral reactivation
Confluent RPE cultures were treated the day before or
simultaneously to virus adsorption with various cytokines: IFN-,
10–100 U/ml; IFN-ß, 100–500 U/ml; TGF-ß2, 3 ng/ml; IL-1ß, 100
U/ml; TNF-, 5–20 ng/ml or the combinations of cytokines: IFN-,
100 U/ml + IL-1ß, 100 U/ml; IFN-, 100 U/ml + TNF-, 10–20
ng/ml. Controls were run in the presence of solvents used for drug or
cytokine solutions. Cytokines were removed after viral adsorption in
all experiments, except for IFN- + IL-1ß used to induce and
maintain the continuous production of NO. After infection, cells were
harvested at different times (5, 10, and 15 days) for analysis of CMV
protein expression.
To study the reversion of viral replication inhibition by IFNs, we used
the classic approach of adding L-tryptophan to cells
treated with either IFN- or IFN-ß (26, 41, 42, 43).
L-tryptophan (5–100 µg/ml) was added to the medium
during cytokine stimulation and again at day 1 and day 2 postinfection.
After a 2-h viral adsorption in the presence or absence of cytokines
and/or tryptophan, cells were washed twice with PBS and refed. At 5,
10, and 15 days postinfection, supernatants were collected for virus
titration and proteins were extracted for analysis of viral protein
expression. The same protocol was used when L-NMMA (final
concentration: 0.5 mM) was added to the cultures. L-NMMA
was added simultaneously with IFN- + IL-1ß stimulation. Nitrite
levels were undetectable under these conditions (see below).
Indirect immunofluorescence assay
To follow viral protein expression, we performed indirect immunofluorescence using the following mAbs: E13 (44) (a gift from Dr. M.-C. Mazeron, Laboratoire de Virologie-Bactériologie, Hôpital Lariboisière, Paris, France) to detect immediate early (IE) proteins, F6b (45) to detect the early tegument protein pp65 (ppUL83) (46), and mAbs F4a and 87-55/02/2 (Behring, Germany, a generous gift from Dr. Walter), which detect, respectively, the late capsid protein p149 (S.M., unpublished results) and the late tegument protein pp150 (UL32) (47). Cells were seeded onto eight-well Labtek glass slides (Nunc, Naperville, IL) at 60,000 cells/well. At confluency (5 to 7 days), they were incubated with 5, 1, or 0.1 PFU of CMV per cell for 2 h, rinsed twice in PBS, and refed medium. Forty-eight hours to 5 days postinfection, depending on the MOI, cells were rinsed with PBS, air dried, and fixed in ice-cold acetone or in ethanol/acetone (V/V) for 10 min at -20°C. Cells were air dried before incubation with the first Ab. Following a 1-h incubation with the first Ab, slides were washed in PBS and incubated with a 1/80 dilution of fluorescein-conjugated anti-mouse IgG (Dako, Trappes, France). After final washing, slides were fixed with 10% formol in PBS, mounted in Vectashield, Vectalabs (Biosys SA., Compiegne, France), and examined and photographed in a Leitz Dialux microscope.
Western blot analysis
At the times indicated under Results, cells were washed twice with PBS and were lysed by addition of Lamelli’s electrophoresis buffer (0.1 ml/well of 24-well plates). Lysates were heated at 95°C for 10 min. Following SDS-PAGE electrophoresis, proteins were transferred to nylon-reinforced nitrocellulose (Sartorius, France) by semidry electrophoresis in transfer buffer (20 mM Tris-base, 150 mM glycine, 20% methanol) overnight at 30 mA. Blots were stained with 0.2% Ponceau red in 3% TCA to control the quality of transfer and saturated with 5% fat-free milk in PBS/0.1% Tween-20 (PBST) for at least 30 min at room temperature. Incubation with 5 µg/ml of primary Abs (see mAbs listed above; note that mAb E13 recognizes two proteins, IE1 and IE2) was conducted for 1 h at room temperature in saturation buffer (5% fat-free milk, PBS, 0.1% Tween-20). Following washing in PBST, blots were incubated with peroxidase-labeled anti-mouse IgG (1/400; Amersham, les Ulis, France) for 1 h at room temperature in saturation buffer. After final washing, blots were developed by chemoluminescence using ECL, according to the manufacturers’ specifications (Amersham).
Extracellular virus production
Supernatants of infected cells were sampled at the times indicated under Results and clarified of cell debris by centrifugation at 2000 rpm for 5 min in an Eppendorf centrifuge. Viral content was assessed by plaque formation under carboxymethyl cellulose, as previously described (39). Each point was run in triplicate.
Nitrite assay and treatment with exogenous NO donors
RPE cells were grown to confluency in 24-well plates and treated
with IFN- (100 U/ml) + IL-1ß (100 U/ml) to stimulate expression of
NOSII (25). Cells were infected at the same time (MOI = 5) for
2 h, then washed and incubated further with the cytokines at the
above concentrations. Infection with UV or heat-inactivated viruses was
performed at the same MOI based on the viral titer before inactivation.
At 72 h postinfection, supernatants were collected and nitrite
levels were determined by a colorimetric assay based on the Griess
reaction, using sodium nitrite standards. Briefly, 100 µl of
cell-free supernatant was mixed with 200 µl of Griess reagent (1%
sulfanilamide, 0.1% naphthyl-ethylenediamine), and, after 10 min,
adsorbance was read at 540 nm. The exogenous NO donor DEANO was used at
different concentrations (0.01, 0.05, 0.1, 0.25, 0.5, and 1 mM) during
virus adsorption. After removal of the inoculum, cells were further
incubated in the presence of DEANO and media were changed every day
with freshly prepared DEANO. The t1/2 of the
molecule was estimated to be 12 h. Complete inactivation of the
DEANO solution was obtained after incubation in medium (pH 7.4) at room
temperature for 24 h. At the end of the incubation period, cell
viability was determined by trypan blue exclusion. Cells were fixed for
immunofluorescence or harvested for Western blot analysis, as described
above.
RNA isolation and RT-PCR-IDO mRNA expression
RPE cultures grown to confluence in 25-cm3 flasks
were treated with medium alone or medium containing IFN- (100 U/ml)
or IFN-ß (500 U/ml) in the presence or absence of CMV (MOI = 5).
Stimulation was stopped after 2 or 24 h. Total RNA was always
prepared from cultures at 24 h using a Qiagen RNA preparation kit
RNeasy (Qiagen S.A., Courteboeuf, France). Total extracted RNA was
treated with RNase-free DNase and quantified spectrophotometrically.
Polyadenylated mRNA was then obtained after purification with a
Oligotex mRNA kit (Qiagen S.A.). Extraction and purification were
performed in accordance with the manufacturers’ instructions.
Polyadenylated mRNA was retrotranscribed in PCR buffer (1.5 mM
MgCl2) containing 100 pmol random primers, 1 mM each
deoxynucleoside triphosphate, and 200 U Superscript reverse
transcriptase (Life Technologies) by incubation at 25°C for 10 min,
followed by incubation at 42°C for 60 min in a final volume of 20
µl. The resulting cDNAs were amplified using 30 pmol of the primers
reported by Nagineni et al. (24) for PCR amplification of IDO and 1 U
of Taq polymerase. Amplification was performed with a
thermocycler 9600 (Perkin-Elmer, Norwalk, CT) at 94°C for 30 s,
57°C for 30 s, and 72°C for 1 min, for a total of 30 cycles.
Samples were subjected to 1% agarose gel electrophoresis, stained with
ethidium bromide, and photographed under UV light. As a positive
control, human ß-actin RNA was retrotranscribed and amplified in
parallel for each sample, and genomic DNA was used as a positive
control for the PCR. As a negative control, an identical amount of RNA
for each sample was amplified without being retrotranscribed.
Data analysis
Results are expressed as mean ± SD and were analyzed statistically by using Mann Whitney test. Statistical analysis was performed with the Statview 4.5 software package (Abacus Concepts, Berkeley, CA). Probability values of <0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Expression of viral proteins corresponding to the three phases of viral replication was analyzed by immunofluorescence and immunoblot. Two different isoforms of IE proteins (72 and 86 kDa) are the first proteins to be expressed (48). The early protein pp65 (UL83) is a matrix phosphoprotein that plays a major role in the clearance of infected cells by the immune system (49). Finally, appearance of either the late tegument protein pp150 (47) or the p149 capsid protein occurs after viral DNA synthesis and corresponds to the terminal phase of viral replication.
The initial phase of infection, monitored by immunofluorescence for IE
protein expression, was not affected by any cytokine treatment, as
shown in Fig. 1
, A–C.
Expression of the viral late protein pp150 was significantly inhibited
after IFN- treatment (Fig. 1, D and F). This
treatment was ineffective when applied more than 24 h after viral
adsorption (Fig. 1E). In contrast, treatment of cells with
TNF-, IL-1ß, or TGF-ß2 had no significant effect on CMV
replication, as assessed by titration of extracellular virus production
(Fig. 2). At 5 days postinfection (using
a MOI of 5), extracellular virus production was 6.4 x
103 PFU/ml in the absence of cytokine treatment (Fig. 2).
However, virus production was reduced significantly in the presence of
IFN- or IFN-ß, decreasing from 6.4 x 103 to
3 x 101 PFU/ml (p < 0.001)
(Fig. 2). Cytopathic effects, observed 72 h after infection in
unstimulated RPE cells, were completely abolished by IFN- (10 U/ml)
or IFN-ß (100 U/ml) treatment (data not shown). Western blot analysis
confirmed these observations by showing a strong inhibition of early
and late protein expression after IFN- treatment (Fig. 3). The apparent inhibition of IE protein
expression, as shown by Western blot analysis, was due to inhibition of
the replication and propagation of progeny virus by IFN- treatment.
A pattern similar to that obtained with IFN- was observed with
IFN-ß (Fig. 3). These effects were dose dependent starting at 10 U/ml
(IFN-) and 50 U/ml (IFN-ß), and were maximal at 100 U/ml (IFN-)
and 500 U/ml (IFN-ß). We observed no synergistic inhibition of viral
protein expression upon treatment with different cytokine combinations
such as IFN- + IL-1ß or IFN- + TGF-ß2 (data not shown).
|
|
|
Slight cytopathic effects were seen on day 3 after infection (MOI:
5) in IFN--stimulated cells in the presence of 20 µg/ml of
L-tryptophan. This effect was greatly enhanced in the
presence of 50 µg/ml L-tryptophan and presented the same
pattern as that seen in untreated, infected cells (data not shown).
Increasing the concentration of L-tryptophan did not
enhance its protective effect, since at high concentrations (100
µg/ml), L-tryptophan was toxic. A similar effect of
tryptophan was obtained for infected cells treated with IFN-ß (data
not shown). Western blot analysis of CMV protein expression confirmed
the restoration of early and late viral protein synthesis upon
L-tryptophan treatment, in a dose-dependent manner (Fig. 3
). Patterns of early and late viral protein expression were similar to
those in untreated, infected cells 5 days after infection (Fig. 3). No
reversion was observed with different L-tryptophan
concentrations after IFN-ß treatment at 500 U/ml. Reversion was more
efficient in cells treated with IFN- (100 U/ml) compared with those
infected with IFN-ß (100 U/ml).
Extracellular virus production in cells stimulated with IFN- or
IFN-ß showed a similar profile in the presence of
L-tryptophan (Fig. 4). CMV
titers evaluated 5 days after IFN- (100 U/ml) or IFN-ß (100 U/ml)
stimulation in the presence of L-tryptophan (50 µg/ml)
were, respectively, 4.3 x 103 PFU/ml and 7 x
102 PFU/ml, close to the titer observed at the same time in
unstimulated infected cell supernatants (6.4 x 103
PFU/ml). To be effective, tryptophan had to be added no more than
12 h after cytokine stimulation (data not shown). In cells
stimulated with a combination of IFN- (100 U/ml) and TNF- (20
ng/ml), the restoration profile upon L-tryptophan treatment
was similar to that observed with IFN- alone (data not shown). In
all cases, reversion of IFN- inhibition started at a minimal
concentration of 5 µg/ml, progressed in a dose-dependent manner, and
was maximal at 50 µg/ml. The toxic effect observed on uninfected
cells at 100 µg/ml of tryptophan explains the low viral titers
obtained in cells treated with 100 µg/ml of L-tryptophan
(Fig. 4).
|
inhibition upon
L-tryptophan treatment strongly implicated involvement of
the IDO pathway induced by IFNs (26, 41, 42, 43). It has been shown
previously that this pathway can be activated at the level of mRNA
transcription by IFN- and, less efficiently, by IFN-ß stimulation
of RPE cells (24). To confirm these results in our RPE cells, we
performed RT-PCR to detect IDO mRNA. Polyadenylated mRNA corresponding
to the enzyme was detected after a 2-h stimulation with either IFN-ß
or IFN- (Fig. 5). The same pattern was
observed when cells were stimulated for 24 h with IFN-.
However, poly(A) mRNA was no longer detectable when cells were
continuously stimulated with IFN-ß for 24 h, suggesting less
sustained induction compared with that by IFN-. CMV infection had no
effect on IDO transcription in unstimulated cells, nor on mRNA
accumulation after IFN- treatment.
|
To evaluate the influence of endogenous NO production on viral
replication, we stimulated RPE cells with a combination of IFN- +
IL-1ß, which is required to induce NOSII in these cells (25). The
level of NO produced after cytokine stimulation of uninfected cells was
14.4 ± 0.8 µM. It was reduced significantly to 2.7 ± 0.3
µM after CMV infection (p < 0.001) (Fig. 6). This inhibition was not observed when
cells were infected with UV- or heat-inactivated virus (respectively,
14.9 ± 0.5 µM and 14.6 ± 0.8 µM), suggesting a
requirement for a functional CMV genome.
|
A). At this MOI, the
efficiency of DEANO on cytopathic effects was comparable with that of
IFN- treatment. In terms of viral protein expression, there was a
significant inhibition of early and late CMV protein expression
starting at 100 µM of DEANO with a maximum at 250 µM, as assessed
at 3 to 5 days postinfection by immunofluorescence (Fig. 7B). Cell toxicity started at 500 µM and was marked at 1
mM, which precluded testing the effect of DEANO at these
concentrations. Using a low MOI (0.1), DEANO at a concentration of 50
µM strongly reduced CMV pp65 and pp150 expression at 5 days after
infection (Fig. 7C). An inactive solution of DEANO that had
previously released all available NO had no cytopathic effect on RPE
cell, confirming the specific effect of NO. No toxic effect was
observed on uninfected cells at the highest concentration of DEANO
used, as determined by trypan blue exclusion.
|
To determine the respective roles played by the IDO and NOSII
pathways in the inhibition of CMV replication, we analyzed
extracellular virus production in the presence of L-NMMA
and/or L-tryptophan in cells stimulated with cytokines
(IFN-, 100 U/ml + IL-1ß, 100 U/ml), which induces the
expression of both enzymes (Fig. 8).
Levels of nitrites produced in culture supernatants after IFN- +
IL-1ß stimulation were negligible in the presence of
L-NMMA (data not shown). However, extracellular virus
production was still inhibited in the presence of L-NMMA,
which was not surprising since CMV itself can block the NO production
(reported above). Treatment of cells with IL-1ß alone had no effect
on viral replication (reported above). It was clear in our model that
treatment with IFN-, or at lower efficiency with IFN-ß, sufficed
to block CMV replication (reported above), therefore showing that
inhibition of viral replication occurred predominantly through
induction of the IDO, and not the NOSII, pathway.
|
In cells treated with IFN-, replication persisted at a very low
level as long as IFN- was maintained in the media (15 days).
Pretreatment of cells with IFN- or IFN-ß 1 day before infection
resulted in a 5-day delay (Fig. 9) before
detection of extracellular virus as compared with unstimulated infected
cells. At 15 days after infection (end of the observation period),
virus titers in supernatants of IFN-treated cells after the removal of
IFN were below the level of virus production by unstimulated cells.
Recovery of virus production occurred more rapidly after removal of
IFN-ß than removal of IFN-. As expected, in infected cells
maintained in the presence of either type of IFN, early and late viral
proteins were undetectable, although virus particles were observed by
electron microscopy (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have reported on the influence of different cytokines
on CMV replication in vitro (53) and reviewed previously (54) and in
animal models, even though results may seem contradictory to those
obtained in vivo (20, 21, 55, 56, 57). In a study of a small number of
patients, IFN- and TNF- were detected in the vitreous or the
retina of HIV-infected patients at different stages of the disease
(15). TGF-ß2 is constitutively present in normal eyes, actively
participating in the maintenance of the ocular immune privilege. At
this level, TGF-ß2 is predominantly secreted by nontransformed,
cultured RPE cells (58). It can modulate expression of HLA class II Ags
in IFN--treated RPE cells, thereby decreasing their capacity as APCs
(5).
Since CMV presents strict host specificity, studies in animal models are not wholly relevant to human CMV. For this reason, we chose to study human CMV-retinal cell interactions and the effects of cytokines on CMV expression using an in vitro model based on human RPE cells. These cells are infected in vivo and have been shown to be permissive for CMV replication in vitro (10, 13). As part of the blood retinal barrier, RPE cells play an important regulatory function in infectious and inflammatory disorders.
While TNF-, IL-1ß, and TGF-ß2 had no effect on CMV replication
in RPE cells, viral protein expression and ultimately viral replication
were blocked by IFN- in a dose-dependent manner. IFN-ß was less
inhibitory since the same level of inhibition required higher doses of
this cytokine. CMV protein expression was shown by FACS analysis to be
markedly inhibited after IFN- treatment in a dose-dependent manner
in infected fibroblasts (59). In RPE cells, we used indirect
immunofluorescence to analyze viral protein expression, which is more
sensitive than FACS analysis; Detrick et al. (13) previously reported
very low efficiency of infection in these cells. Efficiency of viral
replication was highly dependent on the MOI used.
To our knowledge, we are the first to investigate the mechanisms of
IFN-induced inhibition of CMV replication. Others have shown that
inhibition of T. gondii and C. psittaci
replication by IFN- involves the IDO pathway due to consumption of
tryptophan (26, 27, 28), which is transformed into kynurenine and its
metabolites. Nagineni et al. (24) implicated this mechanism
specifically in human RPE cells infected with T. gondii. In
our system, accumulation of polyadenylated IDO mRNA upon IFN-
stimulation suggested that a similar mechanism might be operable in RPE
cells infected with CMV. We therefore tested whether inhibition of CMV
replication could be reversed by the addition of
L-tryptophan. Tryptophan reversion of IFN- inhibition
was efficient and dose dependent. RT-PCR did not show any influence of
CMV infection on IDO mRNA transcription under our experimental
conditions. To our knowledge, this is the first study to demonstrate
that the IDO pathway induced by IFN- is involved in the inhibition
of CMV replication.
It has been shown in a model of human embryonic lung fibroblasts that
CMV replication is blocked by natural human IFN-ß (53). In this
model, inhibition was reportedly due to 2',5'-oligoadenylate (2–5A)
synthetase, although involvement of the IDO pathway was not analyzed.
It is possible that IFN-ß can induce 2',5'-oligoadenylate (2–5A)
synthetase in human RPE cells, therefore explaining the incomplete and
lower efficiency of reversion by L-tryptophan compared with
reversion of IFN- inhibition.
NO has antiviral effects in vitro, but can also play a substantial role
in the pathogenesis of viral infections in vivo (30, 31, 32, 33, 34, 35, 60, 61). We
showed previously that human RPE cells express the inducible form of
NOS upon IFN- + IL-1ß treatment, leading to production of NO (25).
In this study, we addressed the question of the influence of CMV
infection on NO production induced by these cytokines and, conversely,
the effect of NO on CMV replication. CMV significantly down-regulated
the production of NO. This was dependent on the presence of infectious
virus, since no effect was observed with UV- or heat-inactivated virus.
Preliminary experiments not reported herein demonstrate that NOSII mRNA
accumulation after IFN- + IL-1ß stimulation is down-regulated by
CMV (O.G., unpublished results). We showed previously that CMV induces
the synthesis and secretion of TGF-ß1 in fibroblasts (62). TGF-ß is
known to inhibit NOSII expression, especially in RPE cells (25, 63).
Thus, TGF-ß induction by CMV could be contributing to the CMV-induced
inhibition of NOSII. More recently, Miller et al. (64) reported
inhibition of IFN--induced MHC class II expression in CMV-infected
endothelial cells. Induction of the transcription factor IFN-regulatory
factor-1 was blocked in endothelial cells due to disruption of the
JAK/STAT pathway. We have shown previously in bovine RPE cells that
induction of NOSII depends on the expression of IFN-regulatory factor-1
(65). We are currently investigating whether CMV-induced inhibition of
NOSII in human RPE cells also involves inhibition of IFN-regulatory
factor-1.
We also studied the influence of NO on viral replication. Indeed, it
has been shown, in vitro and in animal models, that NO can have
significant and potent antiviral effects (30, 31, 61). To induce NO
production in human RPE, cells have to be treated with IFN-/IL-1ß
(25). Treatment of cells with L-NMMA failed to increase CMV
replication in IFN-/IL-ß1-treated cells. However, due to the
induction of IDO by IFN-, we could not study the isolated effects of
NO induction in this system. We therefore added organic exogenous NO
donors to infected cells. There was a significant, dose-dependent block
of CMV protein expression and replication, even at high MOI and in the
absence of a toxic effect.
The observation that treatment of cells with L-NMMA failed
to increase CMV replication in IFN-/IL-1ß-treated cells suggests
that CMV infection acts like L-NMMA in RPE cells upon
induction of NO synthesis. Furthermore, the combined use of
L-NMMA and L-tryptophan did not enhance viral
replication above that seen with L-tryptophan alone,
strongly suggesting that the inhibitory effect of the IFN-/IL-1ß
is mainly due to the induction of the IDO pathway by IFN-. We
propose a model (Fig. 10) of the
mechanisms operating in IFN- and/or IL-1ß-stimulated RPE cells
relative to CMV replication.
|
).
IFN- inhibited de novo expression of the early viral protein, pp65,
as shown by Western blot analysis and immunofluorescence. However, IE
protein expression was not inhibited, meaning that the viral genome was
effectively transported into the host cell nucleus and that cells
became latently (or persistently) infected. The early viral protein
pp65 is the major, immunodominant target for specific CD8+
CTLs, which confer protection from CMV disease when adoptively
transferred to bone marrow transplant recipients (49, 66). Thus,
CMV-infected cells in the presence of IFN- might escape immune
surveillance due to their lack of pp65 expression. Such infected cells
could potentially represent sources of virus once IFN- is no longer
present. In addition, previous study showed that progressive loss of
cytotoxic T cells is associated with an increased risk of CMV retinitis
in HIV-infected patients (14). Finally, we show that CMV replication
recovers upon removal of IFN- and IFN-ß, indicating that
functional viral genomes are conserved in cells in the presence of IFN
and the absence of viral replication. Thus, cytokines such as IFN-
or IFN-ß may merely drive the virus underground before viral
proteins, which are targets of the immune system, are expressed,
thereby creating viral reservoirs that escape immune
surveillance.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. B. Bodaghi, Unité d’Immunologie Virale, 28 rue du Dr Roux, 75724 Paris, France. E-mail address:
3 Abbreviations used in this paper: RPE, retinal pigment epithelial; DEANO, diethylamine dinitric oxide complex; IDO, indoleamine 2,3 dioxygenase; IE, immediate early; L-NMMA, NG-monomethyl L-arginine; MOI, multiplicity of infection; NO, nitric oxide; NOSII, inducible nitric oxide synthase; PFU, plaque-forming unit.
Received for publication June 17, 1998. Accepted for publication October 2, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced HLA-DR expression in human retinal pigment epithelial cells. Invest. Ophthalmol. Vis. Sci. 35:4253. in the retina in acquired immune deficiency syndrome. Invest. Ophthalmol. Vis. Sci. 33:1829.. Annu. Rev. Immunol. 15:749.[Medline]
production during cytomegalovirus infection in immunosuppressed rats. J. Gen. Virol. 75:779. and tumor necrosis factor. J. Gen. Virol. 75:101. against human cytomegalovirus. Arch. Virol. 94:323.[Medline]
, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J. 5:2516.[Abstract]
blocks the growth of Toxoplasma gondii in human fibroblasts by inducing the host cells to degrade tryptophan. Proc. Natl. Acad. Sci. USA 81:908.-induced nitric oxide synthase. Science 261:1445. interferon-mediated chlamydial persistence. Infect. Immun. 62:3705. action: characterization of indoleamine 2,3-dioxygenase in cultured human cells induced by interferon- and evaluation of the enzyme-mediated tryptophan degradation in its anticellular activity. J. Biol. Chem. 263:2041. promotes replication and pathogenicity of rat cytomegalovirus. J. Virol. 68:2297. in host resistance to cytomegalovirus infection. J. Gen. Virol. 74:2215. interferon. J. Virol. 68:2305. and ß in bovine retinal pigmented epithelial cells. J. Biol. Chem. 272:32169.This article has been cited by other articles:
|
|
 |
|
  S Barry, G Clarke, P Scully, and T. Dinan Kynurenine pathway in psychosis: evidence of increased tryptophan degradation J Psychopharmacol, May 1, 2009; 23(3): 287 - 294. [Abstract] [PDF]
|
|
|
|
 |
|
 E. Delair, C. Creuzet, J. Dupouy-Camet, and M.-P. Roisin In Vitro Effect of TNF-{alpha} and IFN-{gamma} in Retinal Cell Infection with Toxoplasma gondii Invest. Ophthalmol. Vis. Sci., April 1, 2009; 50(4): 1754 - 1760. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Becerra, R. V. Warke, K. Xhaja, B. Evans, J. Evans, K. Martin, N. de Bosch, A. L. Rothman, and I. Bosch Increased activity of indoleamine 2,3-dioxygenase in serum from acutely infected dengue patients linked to gamma interferon antiviral function J. Gen. Virol., April 1, 2009; 90(4): 810 - 817. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Yang, H. Li, P. W. Chen, H. Alizadeh, Y. He, R. N. Hogan, and J. Y. Niederkorn PD-L1 Expression on Human Ocular Cells and Its Possible Role in Regulating Immune-Mediated Ocular Inflammation Invest. Ophthalmol. Vis. Sci., January 1, 2009; 50(1): 273 - 280. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mailankot, D. Smith, S. Howell, B. Wang, J. W. Jacobberger, T. Stefan, and R. H. Nagaraj Cell Cycle Arrest by Kynurenine in Lens Epithelial Cells Invest. Ophthalmol. Vis. Sci., December 1, 2008; 49(12): 5466 - 5475. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Baron and J.-L. Davignon Inhibition of IFN-{gamma}-Induced STAT1 Tyrosine Phosphorylation by Human CMV Is Mediated by SHP2 J. Immunol., October 15, 2008; 181(8): 5530 - 5536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wang, Q.-C. Yu, J. Schroer, E. Murphy, and T. Shenk Human cytomegalovirus uses two distinct pathways to enter retinal pigmented epithelial cells PNAS, December 11, 2007; 104(50): 20037 - 20042. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Cuffy, A. M. Silverio, L. Qin, Y. Wang, R. Eid, G. Brandacher, F. G. Lakkis, D. Fuchs, J. S. Pober, and G. Tellides Induction of Indoleamine 2,3-Dioxygenase in Vascular Smooth Muscle Cells by Interferon-{gamma} Contributes to Medial Immunoprivilege J. Immunol., October 15, 2007; 179(8): 5246 - 5254. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-S. Suh, M.-L. Zhao, M. Rivieccio, S. Choi, E. Connolly, Y. Zhao, O. Takikawa, C. F. Brosnan, and S. C. Lee Astrocyte Indoleamine 2,3-Dioxygenase Is Induced by the TLR3 Ligand Poly(I:C): Mechanism of Induction and Role in Antiviral Response J. Virol., September 15, 2007; 81(18): 9838 - 9850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Moraes, T. de los Santos, M. Koster, T. Turecek, H. Wang, V. G. Andreyev, and M. J. Grubman Enhanced Antiviral Activity against Foot-and-Mouth Disease Virus by a Combination of Type I and II Porcine Interferons J. Virol., July 1, 2007; 81(13): 7124 - 7135. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zhang, J. Zhou, B. Marshall, H. Xin, and S. S. Atherton Lack of iNOS Facilitates MCMV Spread in the Retina Invest. Ophthalmol. Vis. Sci., January 1, 2007; 48(1): 285 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Seymour, V. Ganapathy, A. L. Mellor, and D. H. Munn A high-affinity, tryptophan-selective amino acid transport system in human macrophages J. Leukoc. Biol., December 1, 2006; 80(6): 1320 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Agaugue, L. Perrin-Cocon, F. Coutant, P. Andre, and V. Lotteau 1-Methyl-Tryptophan Can Interfere with TLR Signaling in Dendritic Cells Independently of IDO Activity J. Immunol., August 15, 2006; 177(4): 2061 - 2071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Dobbs, J. E. Strasser, C.-F. Chu, C. Chalk, and G. N. Milligan Clearance of Herpes Simplex Virus Type 2 by CD8+ T Cells Requires Gamma Interferon and either Perforin- or Fas-Mediated Cytolytic Mechanisms J. Virol., December 1, 2005; 79(23): 14546 - 14554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Obojes, O. Andres, K. S. Kim, W. Daubener, and J. Schneider-Schaulies Indoleamine 2,3-Dioxygenase Mediates Cell Type-Specific Anti-Measles Virus Activity of Gamma Interferon J. Virol., June 15, 2005; 79(12): 7768 - 7776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Halary, V. Pitard, D. Dlubek, R. Krzysiek, H. de la Salle, P. Merville, C. Dromer, D. Emilie, J.-F. Moreau, and J. Dechanet-Merville Shared reactivity of V{delta}2neg {gamma}{delta} T cells against cytomegalovirus-infected cells and tumor intestinal epithelial cells J. Exp. Med., May 16, 2005; 201(10): 1567 - 1578. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Terness, J.-J. Chuang, T. Bauer, L. Jiga, and G. Opelz Regulation of human auto- and alloreactive T cells by indoleamine 2,3-dioxygenase (IDO)-producing dendritic cells: too much ado about IDO? Blood, March 15, 2005; 105(6): 2480 - 2486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Eickhoff and M. Cotten NF-{kappa}B activation can mediate inhibition of human cytomegalovirus replication J. Gen. Virol., February 1, 2005; 86(2): 285 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Munn, M. D. Sharma, and A. L. Mellor Ligation of B7-1/B7-2 by Human CD4+ T Cells Triggers Indoleamine 2,3-Dioxygenase Activity in Dendritic Cells J. Immunol., April 1, 2004; 172(7): 4100 - 4110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Adams, K. Besken, C. Oberdorfer, C. R. MacKenzie, O. Takikawa, and W. Daubener Role of Indoleamine-2,3-Dioxygenase in Alpha/Beta and Gamma Interferon-Mediated Antiviral Effects against Herpes Simplex Virus Infections J. Virol., March 1, 2004; 78(5): 2632 - 2636. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Mellor, B. Baban, P. Chandler, B. Marshall, K. Jhaver, A. Hansen, P. A. Koni, M. Iwashima, and D. H. Munn Cutting Edge: Induced Indoleamine 2,3 Dioxygenase Expression in Dendritic Cell Subsets Suppresses T Cell Clonal Expansion J. Immunol., August 15, 2003; 171(4): 1652 - 1655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Mellor and D. H. Munn Tryptophan Catabolism and Regulation of Adaptive Immunity J. Immunol., June 15, 2003; 170(12): 5809 - 5813. [Full Text] [PDF]
|
|
|
|
 |
|
 I. Kosugi, H. Kawasaki, Y. Arai, and Y. Tsutsui Innate Immune Responses to Cytomegalovirus Infection in the Developing Mouse Brain and Their Evasion by Virus-Infected Neurons Am. J. Pathol., September 1, 2002; 161(3): 919 - 928. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Terness, T. M. Bauer, L. Rose, C. Dufter, A. Watzlik, H. Simon, and G. Opelz Inhibition of Allogeneic T Cell Proliferation by Indoleamine 2,3-Dioxygenase-expressing Dendritic Cells: Mediation of Suppression by Tryptophan Metabolites J. Exp. Med., August 19, 2002; 196(4): 447 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-H. Chiou, J.-H. Liu, W.-M. Hsu, S. S.-L. Chen, S.-Y. Chang, L.-J. Juan, J.-C. Lin, Y.-T. Yang, W.-W. Wong, C.-Y. Liu, et al. Up-Regulation of Fas Ligand Expression by Human Cytomegalovirus Immediate-Early Gene Product 2: A Novel Mechanism in Cytomegalovirus-Induced Apoptosis in Human Retina J. Immunol., October 1, 2001; 167(7): 4098 - 4103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Ebrahimi, B. M. Dutia, D. G. Brownstein, and A. A. Nash Murine Gammaherpesvirus-68 Infection Causes Multi-Organ Fibrosis and Alters Leukocyte Trafficking in Interferon-{{gamma}} Receptor Knockout Mice Am. J. Pathol., June 1, 2001; 158(6): 2117 - 2125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Noda, K. Tanaka, S.-a. Sawamura, M. Sasaki, T. Matsumoto, K. Mikami, Y. Aiba, H. Hasegawa, N. Kawabe, and Y. Koga Role of Nitric Oxide Synthase Type 2 in Acute Infection with Murine Cytomegalovirus J. Immunol., March 1, 2001; 166(5): 3533 - 3541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Presti, D. L. Popkin, M. Connick, S. Paetzold, and H. W. Virgin IV Novel Cell Type-Specific Antiviral Mechanism of Interferon {gamma} Action in Macrophages J. Exp. Med., February 19, 2001; 193(4): 483 - 496. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Detrick, C. N. Nagineni, L. R. Grillone, K. P. Anderson, S. P. Henry, and J. J. Hooks Inhibition of Human Cytomegalovirus Replication in a Human Retinal Epithelial Cell Model by Antisense Oligonucleotides Invest. Ophthalmol. Vis. Sci., January 1, 2001; 42(1): 163 - 169. [Abstract] [Full Text]
|
|
|
|
|
|
M. C.-J. Cheeran, S. Hu, G. Gekker, and J. R. Lokensgard Decreased Cytomegalovirus Expression Following Proinflammatory Cytokine Treatment of Primary Human Astrocytes J. Immunol., January 15, 2000; 164(2): 926 - 933. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. F. Pais and R. Appelberg Macrophage Control of Mycobacterial Growth Induced by Picolinic Acid Is Dependent on Host Cell Apoptosis J. Immunol., January 1, 2000; 164(1): 389 - 397. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and cells treated with IFN-
) or IFN-ß
(x) the day before infection. No extracellular virus was detected in
cells continuously maintained in the presence of IFN-µ (
).
NO. CMV infection down-regulates (-) NO
production induced by IFN-