| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Schepens Eye Research Institute, Department of Ophthalmology, Harvard Medical School, Boston, MA 02114; and
Department of Microbiology, Boston University School of Medicine, Boston, MA 02118
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The strict regulation of inflammatory reactions within the eye is vital in maintaining both anatomical integrity and visual function. Left unregulated, inflammation within the eye may lead to extensive ocular damage, resulting in impaired vision. Studies have demonstrated that the Fas-FasL system plays a pivotal role in preventing inflammation within the eye by triggering apoptosis of invading Fas-positive inflammatory cells, thus maintaining the immune privileged environment (3).
Despite the substantial evidence supporting the immunoprotective role for FasL within the eye, applying this phenomenon to the field of transplantation using nonocular tissues has led to conflicting results (4, 5, 6, 7, 8). Lau et al. (5) reported that myoblasts transfected with FasL protected islet allografts from immune-mediated rejection. By contrast, Kang et al. (6) reported that FasL expression on pancreatic islets triggered neutrophil infiltration and graft rejection. Taken together, these studies indicated that FasL could either be immunoprotective, or immunodestructive. The eventual outcome may depend on the level of FasL expression and/or the moderating effects of the local microenvironment.
Recently, data from Chen et al. (9) indicated that the cytokine milieu within the eye antagonized the immunodestructive capacity of FasL-expressing tumors. FasL+ colon carcinoma cells (CT-26) were rejected if injected s.c. in the flank (a nonprivileged site). By contrast, when these same tumor cells were injected into the anterior chamber (AC) of the eye, they experienced immune privilege and grew progressively. Their results suggested that aqueous humor, the fluid within the AC, blocked FasL-induced inflammation, resulting in progressive tumor growth. These data support the idea that the local environment determines whether FasL is either immunoprotective or immunodestructive.
We sought to determine whether the form and/or level of FasL expressed within the ocular environment could modify immune privilege. FasL is a type II transmembrane protein (mFasL) that can be cleaved by specific metalloproteinases to release a soluble form of the protein (sFasL) (10, 11, 12). Because sFasL has been shown to antagonize the functional activity of mFasL (13, 14), cleavage serves to both reduce the level of mFasL and produce a natural antagonist. Therefore, the overall impact of FasL expression is likely to depend on the balance between the relative levels of mFasL and sFasL. To test this premise, tumor cells expressing wild-type (wtFasL), mFasL, or sFasL were evaluated for their ability to induce ocular inflammation. We found that although neither sFasL nor wtFasL could perturb the immune privilege status of the eye, tumor cells that expressed mFasL terminated ocular immune privilege and induced a potent neutrophil-mediated inflammatory response. Interestingly, we determined that the termination of immune privilege was dependent on the level of mFasL expressed on the tumor cell surface, by demonstrating that only tumor cells that expressed high levels of mFasL were capable of inducing an inflammatory response potent enough to terminate immune privilege. Furthermore, the suppressive microenvironment of aqueous humor could not block the proinflammatory effects of tumor cells expressing high levels of mFasL. Another intriguing finding was that termination of immune privilege and ocular inflammation led to the induction of a tumor-specific adaptive immune response. Taken together, our data demonstrate that both the form and level of FasL expressed within the suppressive microenvironment of the eye are critical in determining the immune privileged status of the eye.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Adult female DBA/2, C3H/HeJ, and C3.MRL-Faslpr mice (6–8 wk) and adult male and female SCID/beige mice were purchased from Taconic Farms (Germantown, NY). Aqueous humor was obtained from adult female New Zealand White albino rabbits purchased from Millbrook Breeding Labs (Amherst, MA). All animals were treated according to the Association for Research in Vision and Ophthalmology Resolution on the Use of Animals in Research.
Aqueous humor (AqH)
AqH was obtained as previously described (15) from the ocular AC by paracentesis through a 27-gauge perfusion set (Fisher Scientific, Pittsburgh, PA) into siliconized microfuge tubes. The aqueous humor was used immediately or transiently acidified according to the following standard procedure (16). Approximately 5 µl 1 N HCl (Sigma, St. Louis, MO) was added to 100 µl AqH, bringing the pH up to 2. The acidified sample was left for 1 h at 4°C. Reneutralization of the acidified sample was conducted by adding a 1:1 mixture of 1 N NaOH (Sigma) and 1 M HEPES (Life Technologies, Gaithersburg, MD) (10 µl/100 µl AqH) to return the pH to 7.3. Both fresh and transiently acidified AqH were added to neutrophil cultures as described below.
Tumor cell lines
L5178Y-R tumors expressing no FasL (L5-neo), wtFasL (L5-wtFasL), sFasL (L5-sFasL), or mFasL (L5-mFasL) were produced as previously described (13). Tumor cells were grown in suspension cultures in RPMI 1640 (Life Technologies) supplemented with 10% heat-inactivated FCS (HyClone, Logan, UT), 0.01 M HEPES buffer, 2.0 mM glutamine (Life Technologies), 100 U/ml penicillin G sodium (Life Technologies), 100 g/ml streptomycin sulfate (Life Technologies), 2-ME (1 x 10-5 M; Sigma), and 800 µg/ml Geneticin (Life Technologies).
Evaluation of FasL expression by flow cytometry
Flow cytometry was used to assess surface expression of FasL on L5178Y-R tumor cells expressing no FasL, wtFasL, mFasLlow, or mFasLhigh. Tumor cells (1 x 106 cells) were stained with PE-conjugated anti-mouse FasL (MFL3; BD PharMingen, San Diego, CA) in 50 µl of staining buffer (1x PBS, 1.0% BSA, 0.02% NaN3) for 30 min on ice and washed three times with staining buffer. Cells were resuspended in 1x PBS and then analyzed on a FACScan flow cytometer (BD Biosciences, San Jose, CA), and the data were analyzed using CellQuest software (BD Biosciences).
AC tumor inoculation and growth
L5 cells were washed in HBSS and resuspended in HBSS for inoculations. Using a quantitative technique that has been described previously, 2 x 103 cells in 3 µl were injected into the AC of DBA/2, C3H/HeJ, C3.MRL-Faslpr, or SCID/beige mouse eyes (17). Slit lamp examination was used daily to determine the percentage of the AC occupied by tumor cells. Tumor growth and ensuing inflammation were also examined histologically. Tumor-containing eyes were enucleated, fixed in 10% paraformaldehyde, embedded in paraffin, sectioned, and stained with H&E.
Subcutaneous tumor inoculation and growth
L5 cells (2 x 106 cells) were washed with HBSS and injected s.c. into the rear flanks of syngeneic DBA/2 mice (13). Tumor growth was followed for up to 3 wk by caliper measurements of perpendicular diameters.
Preparation of neutrophils
One milliliter of 9% casein solution (w/v in 0.9% saline) was injected into the peritoneal cavity of DBA/2 mice. A repeat injection was given 24 h later. Three hours after the second injection, the mice were euthanized, and the peritoneal cavity was washed twice with 5 ml of 1 x PBS plus 0.02% EDTA. Cells from the peritoneal fluid were pooled from 4–5 mice and washed three times with 1x PBS. One milliliter of the cell suspension (3–5 x 107/ml) was added to 9 ml of Percoll gradient solution (10 ml sterile 10x PBS plus 90 ml sterile Percoll (Pharmacia Biotech, Piscataway, NJ)). The mixture was ultracentrifuged for 20 min at 60,650 x g. The polymorphonuclear neutrophils were collected from the second opaque layer, washed with PBS, and checked for purity and viability. The purity of murine neutrophil preparations was determined by cytospin to be 98% neutrophils. The neutrophils were maintained in serum-free medium (RPMI 1640 supplemented with 0.01 M HEPES buffer, 1 mM sodium pyruvate (Life Technologies), 1 mM nonessential amino acids (Life Technologies), 100 U/ml penicillin G sodium, 100 g/ml streptomycin sulfate, 1 mg/ml BSA (Sigma), 50 mM 2-ME, and 1/500 diluted (insulin, transferrin, sodium selinite liquid culture supplement (ITS + 1)) liquid culture supplement (Sigma)).
Cytokine determinations
Murine neutrophils (300,000 cells/well) were incubated in a 24-well plate (Falcon; NJ) with serum free medium alone, or L5 tumor cells (300,000 cells/well) expressing either: 1) no FasL, 2) wtFasL, 3) sFasL, or 4) mFasL. The total volume was 1 ml of medium per well. For a positive control, neutrophils were incubated with 1 µg/ml LPS (SIGMA). In experiments that use aqueous humor (AqH), murine neutrophils (60,000 cells/well) were added to a 96-well, U-bottom plate (Falcon; BD Biosciences) with serum-free medium, 50% fresh AqH, or 50% transiently acidified AqH. The neutrophils were incubated in the presence or absence of L5 tumor cells (120,000 cells/well) expressing 1) no FasL, 2) wtFasL, or 3) mFasL. The total volume was 200 µl medium per well. For a positive control, neutrophils were incubated with 1 µg/ml LPS (Sigma).
For both culture systems, neutrophils and tumor cells were cultured
together for 18 h at 37°C and 5% CO2. At
the end of the culture period, the plates were centrifuged at 500
x g for 5 min, and 50 µl of supernatant were harvested
from triplicate cultures. The concentration of IL-1
and
macrophage-inflammatory protein-2 (MIP-2) was measured using the
Quantikine M Mouse Immunoassay kit according to the manufacturer’s
protocol (R&D Systems, Minneapolis, MN).
Assay for protective antitumor immunity
DBA/2 mice that had been injected with 2 x 103 L5-mFasL cells into the AC and had rejected the tumor were subsequently challenged (40 days after the initial AC injection) with a s.c. challenge of 2 x 106 L5-neo cells in the flank. Tumor growth was determined by measuring tumor diameter every two days with calipers.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first series of experiments were designed to determine whether
the form of FasL expressed on tumor cells altered their ability to
experience ocular immune privilege. L5 tumor cells were transfected
with four FasL cDNAs: 1) wtFasL; 2) mFasL; 3) sFasL, or 4) no FasL
(empty vector; L5-neo). These tumor cells were characterized previously
(13). Briefly, soluble FasL was detected in supernatants
from L5-sFasL and L5-wtFasL cells, but not in supernatants from
L5-mFasL cells. L5-sFasL cells secreted significantly more sFasL than
L5-wtFasL cells. L5-mFasL cells expressed a 5- to 10-fold higher level
of mFasL, as compared with L5-wtFasL cells (Fig. 1
A). The lower level on
L5-wtFasL cells was due, at least in part, to the cleavage of mFasL by
matrix metalloproteinase enzymes, because addition of the
metalloproteinase inhibitor KB8301 increased the amount of membrane
FasL on L5-wtFasL cells (data not shown).
|
Tumor cells were injected into syngeneic DBA/2 mice at either 1) a
s.c. site in the flank (nonprivileged site) or 2) the AC of the eye
(immune-privileged site). Tumor diameter was measured at regular
intervals and was used to assess s.c. tumor growth (Fig. 2), whereas histological analysis and
survival was used to assess tumor growth within the AC of the eye (Fig. 3). As a positive control, L5-neo (FasL
negative) tumors grew progressively in both the AC and s.c. sites,
indicating that in the absence of FasL, the L5 tumors alone are not
immunogenic (Fig. 2 and Fig. 3A). By definition, a tumor
benefits from immune privilege within the AC if the tumor grows
progressively within the eye but is rejected from the nonprivileged
s.c. site. L5-wtFasL cells were rejected from the s.c. site (Fig. 2),
but experienced immune privilege within the eye and 100% of the mice
succumbed to progressive tumor growth (Fig. 3A). This same
result was observed and reported by Chen et al. 9) and
supports the hypothesis that wtFasL in the ocular environment
contributes to immune privilege. Interestingly, tumors that expressed
sFasL grew progressively in both s.c. and AC sites, displaying a growth
pattern indistinguishable from L5-neo tumors (Figs. 2 and 3,
B–D). Taken together, these data indicate that tumor cells
expressing wtFasL experience immune privilege within the eye and grow
progressively. In addition, tumor cells expressing sFasL fail to induce
an antitumor inflammatory response in both nonprivileged and privileged
sites
|
|
), but to our
surprise, L5-mFasL cells experienced no immune privilege within the AC
of the eye; tumors were rejected completely and all mice survived
indefinitely (Fig. 3A). Tumor rejection within the eye was
examined histologically by comparing rejecting L5-mFasL tumors (Fig. 3, E–G) with progressively growing L5-neo tumors (Fig. 3, B–D). Three days after injection into the AC, L5-neo and
L5-mFasL tumors were present within the anterior angle. By day 6
postinoculation the L5-neo tumors were growing progressively and by day
11 had filled 100% of the AC. By contrast, tumor regression and
fibrosis was observed in the AC of mice injected with L5-mFasL cells on
days 6 and 11 postinoculation. Complete rejection of mFasL-expressing
tumors was observed by day 32 and resulted in phthisis of the
tumor-containing eye. Phthisis typically occurs following an intense
inflammatory response leading to nonspecific destruction of normal
ocular tissue (18). Thus, L5-mFasL cells, but not
L5-wtFasL cells, break through the immune privileged status of the
eye. mFasL alone stimulates ocular inflammation
Experiments from other laboratories found that rejection of
nonocular tumors expressing FasL coincided with infiltration of
neutrophils (9, 13) To determine whether neutrophil
infiltration coincided with rejection of tumors from the anterior
chamber of the eye, histological analysis of the L5 derivatives was
performed 9 days postinoculation. Tumor cells expressing no FasL (Fig. 4A) or L5-sFasL (Fig. 4B) grew to fill the AC with solid tumor, and there was no
evidence of inflammatory infiltrate. By contrast, tumor regression was
observed in mice that received inoculation of L5-mFasL cells (Fig. 4C). Furthermore, this regression was associated with an
intense inflammatory response characterized by neutrophils present in
both the cornea and AC (Fig. 4D). These results demonstrate
that neutrophil infiltration coincides with rejection of L5-mFasL
tumors in the eye.
|
Following recruitment to the site of inflammation, neutrophils
play two major roles: 1) eradication of cellular debris through
phagocytosis and release of hydrolytic enzymes and reactive oxygen
products; and 2) recruitment and activation of additional neutrophils
and other inflammatory cells through the production of proinflammatory
cytokines and chemokines (19, 20). To determine whether
L5-mFasL tumor cells could directly activate neutrophils to secrete the
chemokine MIP-2 and proinflammatory cytokine IL-1, casein-elicited
neutrophils were incubated (in vitro) with the L5 derivatives.
Neutrophils were obtained from 1) DBA/2, 2)
C3.MRL-Faslpr, or 3) C3H/HeJ mice. Cells from
C3.MRL-Faslpr mice do not express functional Fas
receptors and were used to demonstrate that cytokine secretion from
neutrophils was triggered by the Fas receptor. The results were
identical when neutrophils were used from either DBA/2 or C3H/HeJ mice
(data from DBA/2 mice not shown).
Interestingly, L5-mFasL cells stimulated Fas+
neutrophils to secrete significant amounts of MIP-2 (11.0 ± 0.4
pg/ml) and IL-1 (20.0 ± 4.6 pg/ml), whereas L5-sFasL cells
were unable to stimulate Fas+ neutrophils to
secrete either MIP-2 or IL-1 (Fig. 5).
wtFasL-stimulated neutrophils secrete a small amount of both MIP-2 and
IL-1. These data demonstrate that mFasL, but not sFasL, is capable
of directly activating neutrophils to secrete both MIP-2 and
IL-1.
|
only in neutrophils that expressed functional Fas receptors (MIP-2:
11.0 ± 0.4 pg/ml C3H vs 0 pg/ml MRLlpr)
(IL-1: 20.0 ± 4.6 pg/ml C3H vs 0 pg/ml
MRLlpr). As a positive control, LPS (a
FasR-independent activator) stimulated production of both MIP-2 and
IL-1 equally well in neutrophils with and without functional FasR.
Taken together, these data confirm that: 1) mFasL is capable of
directly activating neutrophils; and 2) this activation is dependent on
a functional Fas-FasL system.
AqH inhibits neutrophil production of IL-1
Previous studies by Chen et al. (9) demonstrated that
the microenvironment of the eye, not the amount of FasL, determines
whether wtFasL on tumor cells will activate neutrophils. However, these
studies did not address the role of the form of FasL expressed on the
tumor cells. To determine whether the microenvironment of the eye,
particularly TGF-, inhibits the production of IL-1 by neutrophils
stimulated with tumor cells expressing mFasL, casein-elicited
neutrophils and L5 derivatives were incubated (in vitro) in the
presence or absence of 50% AqH.
AqH contains many immunosuppressive factors that contribute to the
immunosuppressive environment of the eye (21). However,
only TGF- has been shown to directly inhibit FasL activation of
neutrophils (9). Whereas TGF- is expressed at high
levels in the AqH, the predominant form in fresh unperturbed AqH is
latent (22). The level of activated TGF- in AqH is
greatly influenced by 1) the species used, 2) isolation technique, 3)
handling, and 4) storage (16, 22, 23). Therefore, studies
using fresh AqH to assess the suppressive effects of TGF- may be
quite variable. On the basis of these findings we assessed the ability
of 1) fresh and 2) transiently acidified AqH to block IL-1
production by neutrophils stimulated with the different L5 derivatives.
Transient acidification of AqH activates all latent TGF-.
In the absence of AqH, L5-mFasL tumor cells stimulate neutrophils to
produce significant amounts of IL-1 (60 ± 5.8 pg/ml), whereas
L5-wtFasL tumor cells stimulate neutrophils to produce much smaller
amounts of IL-1 (6.0 ± 0.8 pg/ml) (Fig. 6). Neutrophil production of IL-1 in
the presence of fresh AqH was reduced by 47% when stimulated with
wtFasL tumor cells and by only 23% when stimulated with mFasL tumor
cells. A much more substantial reduction was observed when the same
experiment was performed using activated AqH. IL-1 production by
neutrophils was reduced by nearly 90% when stimulated with wtFasL
tumor cells, but only by 50% when stimulated with mFasL tumor cells.
Taken together, these data demonstrate that the microenvironment of the
eye alone is unable to block the proinflammatory effects of tumor cells
expressing high levels of mFasL.
|
To ascertain whether the level of mFasL determines the initiation
of ocular inflammation and/or the termination of immune privilege, a
clone of L5178Y-R tumor cells was established that expressed a lower
level of mFasL (L5-mFasLlow), as compared with
the level of mFasL on tumor cells used in the previous experiments
(Fig. 1B). These clones were injected into both the AC and
s.c. sites. As a positive control, L5-neo tumor cells (FasL negative)
grew progressively in either the s.c. (Fig. 7A), or the AC sites (Fig. 7B). As shown here and in the previous experiments, L5-mFasL
tumors expressing high levels of mFasL (L5-mFasL) were rejected from
either the s.c. or the AC sites. By contrast,
L5-mFasLlow tumors were rejected from the s.c.
site (Fig. 7A), but not the AC site (Fig. 7B).
Histological studies revealed a low level of inflammation present early
following injection of L5-mFasLlow tumors into
the AC site, as demonstrated by a neutrophil infiltrate (data not
shown). However, the level of infiltration was insufficient to
terminate immune privilege. We conclude from these data that only
tumors with high levels of mFasL terminate immune privilege, whereas
immune privilege remains intact for tumors that express low levels of
mFasL.
|
L5178Y-R lymphoma cells expressing mFasL are not rejected in SCID/beige mice
SCID/beige mice were used to determine whether neutrophils, in the
absence of an adaptive immune response, are capable of completely
eliminating L5-mFasL tumors from the eye. SCID/beige mice received
injections of L5 cells expressing the different forms of FasL. Tumor
growth within the AC was assessed by slit lamp microscopy, and the
results are displayed as the percentage of the AC containing tumor
(Fig. 8A). Survival of
tumor-bearing mice was also observed (Fig. 8B). As expected,
L5-neo or L5-sFasL tumors grew progressively, and no mice survived past
day 20 postinoculation. Interestingly, although L5-mFasL tumors
eventually grew progressively, there was a significant delay in tumor
growth during the first 20 days postinoculation. Moreover, although all
mice eventually succumbed to L5-mFasL tumors, these mice survived
significantly longer (day 30) as compared with the other groups of mice
(day 20). We conclude that, in the absence of T cells, NK cells, and B
cells, neutrophils are capable of eliminating some tumor cells and
slowing the growth of ocular tumors. However, expression of mFasL is
unable to bring about complete elimination of all tumor cells from the
immune privileged ocular compartment of SCID/beige mice. These data
suggest that an early nonspecific neutrophil-mediated inflammatory
response is followed by an adaptive immune response that is ultimately
responsible for rejecting the ocular tumor.
|
To determine whether the delay in ocular tumor growth in
SCID/beige mice coincides with infiltration of neutrophils into the
tumor-containing eye, a histological analysis was performed. As a
control, tumor growth in SCID/beige mice was compared with tumor growth
in DBA/2 mice. Tumor cells expressing no FasL grew to fill the AC with
solid tumor, and there was no evidence of any inflammatory infiltrate
in either SCID/beige or DBA/2 mice (Fig. 9). By contrast, an intense
neutrophil-mediated inflammatory response was present in the cornea and
AC of both DBA/2 and SCID/beige mice injected with L5-mFasL cells (Fig. 9). These data indicate that the delay in growth of AC L5-mFasL tumors
coincides with extensive neutrophil infiltration. However, in the
absence of an adaptive immune response, the neutrophils alone are not
sufficient to eliminate ocular tumors completely.
|
To determine whether rejection of L5-mFasL ocular tumors results
in long term systemic protective immunity, DBA/2 mice that had
completely eliminated L5-mFasL tumors from the AC were given a
secondary tumor challenge in the flank (s.c.) with L5-neo tumor cells.
Because these secondary tumors are FasL negative, only mice with
systemic protective antitumor immunity would be expected to survive. As
a negative control, naive mice (without a previous ocular tumor)
received a similar tumor s.c. challenge in the flank. As expected,
tumors grew progressively within the flank of naive mice (Fig. 10). By contrast, mice that had
previously rejected L5-mFasL tumors from the AC were protected
completely and eliminated FasL-negative tumors injected s.c. We
conclude that rejection of tumors expressing mFasL from the eye results
in acquisition of systemic protective antitumor immunity. In addition,
these results imply that the early nonspecific inflammatory response
induced during rejection of L5-mFasL tumors results in activation of
tumor-specific T cells that are ultimately responsible for protective
immunity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(22, 26),
and the expression of cell surface molecules, such as FasL (3, 27), act together to maintain the immune privileged environment
of the eye. FasL is constitutively expressed in the eye and is thought to protect ocular integrity and maintain immune privilege by inducing apoptosis in infiltrating Fas+ immune cells (3). Previous studies have demonstrated that the constitutive expression of FasL on the cornea plays a significant role in corneal allograft survival (27, 28). Using a murine model of corneal transplant, both Stuart et al. (27) and Yamagami et al. (28) showed that FasL-positive grafts were accepted 45–50% of the time whereas FasL-negative corneal grafts were rejected nearly 100% of the time.
Interestingly, although expression of FasL within the eye appeared to
be immunoprotective, transplantation experiments conducted at nonocular
sites revealed that FasL could be either immunoprotective (4, 5, 8) or immunodestructive (6, 9). In an effort to
resolve these conflicting data, Chen et al. demonstrated that colon
carcinoma cells transfected with wtFasL were proinflammatory when
injected into a nonprivileged site (s.c. tissue), but the same tumor
cells failed to induce an inflammatory response when injected into an
immune-privileged site (AC) (9). Their data indicated that
TGF-, which is present in AqH within the anterior chamber, inhibited
the ability of FasL-positive tumor cells to activate neutrophils. They
concluded that the microenvironment determined the function of FasL;
within immune-privileged sites FasL is noninflammatory, and within
nonprivileged sites FasL is proinflammatory.
We observed similar results using lymphoma cells that expressed
wild-type FasL that were injected into the AC. These tumor cells grew
progressively and failed to initiate either ocular inflammation, or
neutrophil infiltration. By contrast, injection of L5-mFasL cells that
expressed only mFasL into the AC of the eye 1) terminated immune
privilege, 2) initiated tumor rejection and vigorous ocular
inflammation, 3) induced extensive infiltration of neutrophils into the
tumor-containing eye, and 4) activated neutrophils (in vitro) to
secrete MIP-2 and IL-1. These results for mFasL tumor cells were
surprising because L5-wtFasL tumors express both mFasL and sFasL and
had no effect on ocular immune privilege.
One interpretation of these data is that there is a threshold level of mFasL that is required to initiate inflammation within the immune-privileged eye. Our data demonstrate that only tumor cells expressing high levels of mFasL are able to trigger ocular inflammation and tumor rejection, whereas tumor cells expressing low levels of mFasL (L5-wtFasL and L5-mFasLlow) are only able to trigger mild inflammation that fails to terminate privilege, and the tumors grow progressively. Interestingly, both L5-mFasLhigh and L5-mFasLlow tumors triggered potent antitumor inflammation and were completely rejected from the nonimmune privileged s.c. tissue of the flank. Taken together, these data clearly support the idea that 1) high levels of mFasL are required to initiate inflammation and terminate immune privilege within the eye and 2) the threshold level of mFasL required to initiate inflammation is much higher within the immune-privileged eye than within the non-immune-privileged s.c. tissue of the flank.
The fact that low levels of mFasL initiate vigorous inflammation and
tumor rejection in nonprivileged sites, whereas significantly higher
levels of membrane FasL are required to achieve the same effect within
the immune-privileged eye, raises an important question of how the eye
down-regulates or blocks the proinflammatory function of mFasL. One
likely source of regulation is the unique ocular microenvironment
within the AC of the eye, which is filled with AqH that contains
soluble immunomodulatory factors (TGF-, 
-melanocyte-stimulating
hormone, vasoactive intestinal peptide, calcitonin gene-related
peptide) (21, 22, 26, 29, 30, 31). These soluble factors may
block directly the expansion of the inflammatory cascade initiated by
mFasL tumor cells by inhibiting the release of chemokines and
proinflammatory cytokines from Fas-positive neutrophils.
Chen et al. (9) concluded from their TGF-
studies that the microenvironment within the eye and not the level of
FasL expressed on tumor cells determined whether wtFasL tumor cells
induced inflammation. We observed similar results using tumor cells
expressing wtFasL, demonstrating that AqH blocked completely the
release of IL-1 from neutrophils stimulated with wtFasL tumor cells.
By contrast, AqH failed to block completely the release of IL-1 from
neutrophils stimulated with tumor cells expressing high levels of
mFasL. Taken together, these data demonstrate that the form and level
of FasL are critical in determining function, because the suppressive
microenvironment within the eye cannot block the proinflammatory
effects of tumor cells expressing high levels of mFasL.
Although our studies demonstrate the suppressive nature of AqH, which
specific factor(s) blocks the proinflammatory effects of mFasL is under
investigation. Chen et al. have shown TGF- is involved; however,
there are many other factors present in the AqH, including soluble
FasL. Recent studies demonstrate that sFasL is not only noninflammatory
but can also actively inhibit mFasL-induced inflammation (13, 14). Therefore, we predict the environment within the eye uses
multiple soluble factors, including sFasL to increase the threshold
level of mFasL required to initiate vigorous inflammation. Experiments
are in progress to test this prediction.
In addition to elucidating the mechanism(s) by which the eye increases
the threshold of mFasL required to induce inflammation, it is important
to identify the target cell population FasL initially triggers on
injection into the AC. Although our studies clearly demonstrate that
neutrophil-mediated inflammation coincides with tumor rejection within
the AC of the eye, these studies do not rule out the possibility that
the FasL expressed on tumor cells initially triggers a target
population other than neutrophils, which release chemotactic factors
for neutrophils. Previous studies using a peritoneal tumor model
demonstrate that mFasL expressed on tumor cells actually triggers
resident macrophages that in turn release chemotactic factors for
neutrophils (13, 32). Furthermore, although functional Fas
is required on host cells to trigger the neutrophil response, the
neutrophils themselves do not require Fas expression (13).
Therefore, although our experiment using neutrophils from Fas-deficient
MRL-lpr mice demonstrates the importance of functional Fas for both
IL-1 and MIP-2 production, it is possible that the Fas expressed on
the 1–2% nonneutrophil cells present in our cell preparation is
actually the critical factor. This would require that only a few
Fas-expressing cells (nonneutrophils) triggered by mFasL would be
sufficient to activate neutrophils.
Because neutrophils are not normally present within the AC of the eye,
it is more likely that there is a population of cells other than
neutrophils that is the initial target of mFasL tumor cells. When
injected into the AC, the tumor cells grow first at the angle of the
iris and cornea. Therefore, it is possible that the mFasL triggers the
macrophages/dendritic cells present in the iris to release factors such
as IL-1 and MIP-2, which in turn recruit neutrophils to the site.
Although our studies clearly demonstrate that neutrophil-mediated
inflammation coincides with tumor rejection in the AC, the initial
target cell population triggered by mFasL remains unclear. Additional
studies are under way to identify this population and to assess the
importance of Fas expression on cells resident within the eye and on
the neutrophils recruited to the eye.
Another intriguing aspect of our data was the observation that activation of neutrophils alone was insufficient to eliminate completely ocular L5 tumors that express high levels of mFasL. The fact that SCID/beige mice (with a full complement of neutrophils) are unable to reject L5-mFasL tumors indicates that an adaptive immune response is necessary for complete tumor eradication. Our results also demonstrate that mice that have rejected ocular L5-mFasL tumors acquire protective systemic antitumor immunity and are capable of eliminating a subsequent tumor challenge in the flank. Expression of protective immunity is independent of FasL, because FasL- tumor cells were used in the second tumor challenge. Although additional experiments are required, these data suggest that tumor-specific T cells mediate protective immunity and that these effector cells ultimately reject the intraocular tumor cells. If this occurs, it would be surprising, because the long-standing dogma regarding the Fas-FasL system and immune privilege has been that the expression of FasL on ocular tissues maintains immune privilege by eliminating infiltrating T lymphocytes and other inflammatory cells.
Therefore the question becomes, "Why are the Fas+ T cells not eliminated by the FasL+ tumor cells?" Recent studies have demonstrated that the FasL can transduce signals via the Fas receptor using nonapoptotic pathways in monocytes (33), dendritic cells (34), and T cells (35). In addition, studies have shown that CD4+ T cells and CD8+ T cells exhibit different sensitivities to Fas-mediated apoptosis (36, 37). In particular, CD4+ Th1 cells are more sensitive to Fas-mediated apoptosis than are CD4+ Th2 or CD8+ T cells. Therefore, tumor rejection could be mediated through cytotoxic CD8+ T cells and CD4+ Th2 cells.
In conclusion, the current study clearly demonstrates that both the form and level of FasL expressed within the eye is critical in immune privilege. sFasL is unable to promote an inflammatory response, resulting in progressive tumor growth. By contrast, mFasL terminates immune privilege and facilitates both a potent innate and adaptive immune response, which together reject the tumor. Furthermore, there is a threshold level of mFasL required to initiate inflammation that successfully rejects tumors in the immune-privileged eye, and this threshold is higher in privileged than in nonprivileged sites.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Bruce R. Ksander, Schepens Eye Research Institute, 20 Staniford Street, Boston, MA 02114. E-mail address: ksander{at}vision.eri.harvard.edu
3 Abbreviations used in this paper: FasL, Fas ligand; L5, L5178Y-R T lymphoma cells; mFasL, membrane FasL; sFasL, soluble FasL; wtFasL, wild-type FasL; neo, no FasL; AC, anterior chamber; AqH, aqueous humor; MIP-2, macrophage-inflammatory protein-2.
Received for publication December 3, 2001. Accepted for publication June 24, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(TGF-)-producing regulatory T-cells. Curr. Eye Res. 16:900.[Medline]
. Methods Enzymol. 198:327.[Medline]
as an immunosuppressive factor in aqueous humor. Invest. Ophthalmol. Vis. Sci. 32:2201. from fibroblast-conditioned medium. J. Cell Biol. 106:1659. and a small (3500 daltons) inhibitor of thymocyte proliferation. J. Immunol. 144:3021.[Abstract]
, and the production of interferon in the absence of IL-12 during DC-T cell cognate interaction: a new role for Fas ligand in inflammatory responses. J. Exp. Med. 192:1661.This article has been cited by other articles:
|
|
 |
|
  M. S. Gregory, R. R. Saff, A. Marshak-Rothstein, and B. R. Ksander Control of Ocular Tumor Growth and Metastatic Spread by Soluble and Membrane Fas Ligand Cancer Res., December 15, 2007; 67(24): 11951 - 11958. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-C. Su, Y.-P. Lin, Y.-J. Cheng, J.-Y. Huang, W.-J. Chuang, Y.-S. Shan, and B.-C. Yang Phosphatidylinositol 3-Kinase/Akt Activation by Integrin-Tumor Matrix Interaction Suppresses Fas-Mediated Apoptosis in T Cells J. Immunol., October 1, 2007; 179(7): 4589 - 4597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. F. Ng, E. Lavik, H. Keino, A. W. Taylor, R. S. Langer, and M. J. Young Creating an Immune-Privileged Site Using Retinal Progenitor Cells and Biodegradable Polymers Stem Cells, June 1, 2007; 25(6): 1552 - 1559. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lin, E. Carlson, E. Diaconu, and E. Pearlman CXCL1/KC and CXCL5/LIX are selectively produced by corneal fibroblasts and mediate neutrophil infiltration to the corneal stroma in LPS keratitis J. Leukoc. Biol., March 1, 2007; 81(3): 786 - 792. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Chen, S. Liu, D. Park, Y. Kang, and G. Zheng Depleting Intratumoral CD4+CD25+ Regulatory T Cells via FasL Protein Transfer Enhances the Therapeutic Efficacy of Adoptive T Cell Transfer Cancer Res., February 1, 2007; 67(3): 1291 - 1298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. McKenna and J. A. Kapp Accumulation of Immunosuppressive CD11b+ Myeloid Cells Correlates with the Failure to Prevent Tumor Growth in the Anterior Chamber of the Eye J. Immunol., August 1, 2006; 177(3): 1599 - 1608. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Guo, M. Zhang, H. Tang, and X. Cao Fas signal links innate and adaptive immunity by promoting dendritic-cell secretion of CC and CXC chemokines Blood, September 15, 2005; 106(6): 2033 - 2041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Engelbert and M. S. Gilmore Fas Ligand but Not Complement Is Critical for Control of Experimental Staphylococcus aureus Endophthalmitis Invest. Ophthalmol. Vis. Sci., July 1, 2005; 46(7): 2479 - 2486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Gregory, S. Koh, E. Huang, R. R. Saff, A. Marshak-Rothstein, S. Mukai, and B. R. Ksander A Novel Treatment for Ocular Tumors Using Membrane FasL Vesicles to Activate Innate Immunity and Terminate Immune Privilege Invest. Ophthalmol. Vis. Sci., July 1, 2005; 46(7): 2495 - 2502. [Abstract] [Full Text] [PDF]
|
