Accumulation of Immunosuppressive CD11b+ Myeloid Cells Correlates with the Failure to Prevent Tumor Growth in the Anterior Chamber of the Eye

Kyle C. McKenna and Judith A. Kapp
J Immunol August 1, 2006, 177 (3) 1599-1608; DOI: https://doi.org/10.4049/jimmunol.177.3.1599

Abstract

The purpose of these studies is to determine why an immunogenic tumor grows unchecked in the anterior chamber (a.c.) of the eye. The OVA-expressing EL4 tumor, E.G7-OVA, was injected into the a.c. or skin of immunocompetent and immunodeficient mice. Tumor growth and tumor-specific immune responses were monitored. Ocular tumor-infiltrating leukocytes were characterized phenotypically and functionally. Growth of E.G7-OVA was inhibited when limiting numbers of cells were injected in the skin but not in the a.c. of C57BL/6 mice, although both routes primed OVA-specific immune responses, which prevented the growth of a subsequent injection with E.G7-OVA in the skin or opposite eye. Tumor regression was OVA-specific because growth of the parental EL-4 tumor was not inhibited in primed mice. E.G7-OVA growth in the skin was not inhibited in immunodeficient Rag−/− or CD8 T cell-deficient mice, suggesting that CD8+ CTLs mediate tumor elimination. CD8+ T cell numbers were significantly increased in eyes of mice primed with E.G7-OVA, but few were detected in primary ocular tumors. Nevertheless, growth of E.G7-OVA was retarded in the a.c. of TCR-transgenic OT-I mice, and CD8+ T cell numbers were increased within eyes, suggesting that tumor-specific CD8+ CTLs migrated into and controlled primary ocular tumor growth. E.G7-OVA did not lose antigenicity or become immunosuppressive after 13 days of growth in the eye. However, CD11b+ cells accumulated in primary ocular tumors and contained potent immunosuppressive activity when assayed in vitro. Thus, CD11b+ cells that accumulate within the eye as tumors develop in the a.c. may contribute to immune evasion by primary ocular tumors by inhibiting CTLs within the eye.

Uveal melanoma is the most common primary malignant intraocular tumor in adults (1). These malignancies arise in pigmented ocular tissues including the iris, ciliary body, and choroid and metastasize hematogenously to the liver (2, 3). Metastatic disease of the liver is the leading cause of death in patients with uveal melanoma, and there is currently no effective treatment (2, 3).

It is now well documented that malignant uveal and cutaneous melanomas express common tumor-specific Ags, e.g., tyrosinase, GP-100, and MART, that are highly immunogenic (4). These Ags are presented on the melanoma cell surface as peptides associated with MHC class I molecules and target malignant cells for recognition by CD8+ CTL via the TCR. Patients with cutaneous melanomas that lose expression of MHC class I have a shorter life span, supporting the idea that CD8+ CTL control tumor growth (5). However, cutaneous melanoma patients vaccinated with melanoma-specific Ags in one study showed no evidence of tumor regression despite increased melanoma-specific CD8+ CTL precursors (6). Thus, melanoma-specific CD8+ T cells were not able to promote melanoma regression when they were induced in patients after tumors had been identified.

In contrast with cutaneous melanoma, uveal melanomas that express MHC class I molecules have been associated with a more aggressive phenotype, which correlates with a poorer prognosis in comparison to uveal melanomas that have lost or decreased MHC class I expression (7). The association of a poor prognosis with MHC class I expression in uveal melanoma is believed to result from the loss of antitumor NK cell activity whose effector function is inhibited by MHC class I molecules (8), but it may also involve inhibition of CD8+ CTLs, which recognize tumor Ags presented by MHC class I on uveal melanomas. Understanding the mechanisms of CTL inhibition to ocular tumors may lead to therapeutic approaches to overcome T cell inhibition and promote ocular tumor elimination.

Ocular immune privilege is exemplified by the observation that immunogenic tumors are eliminated by tumor-specific CD8+ CTL responses when injected in the skin but not in the anterior chamber (a.c.)3 of the eye (9). Several potential explanations have been put forth to explain the evasion of CD8+ CTL by tumors developing in the eye, including the following: 1) immune ignorance due to sequestration of ocular Ags; 2) systemic tolerance in tumor-specific CTLs; 3) inhibited migration of CTL into the eye; 4) inhibition of CTL activity within the eye; and 5) loss of Ag expression by the tumor. Immune ignorance due to sequestration of ocular Ags was postulated because of the blood-aqueous barrier and the absence of demonstrable afferent lymphatics draining the a.c. (10). However, many observations indicate that ocular Ags normally escape the eye. For example, a.c. administration of soluble OVA induces T cell expansion in the draining submandibular lymph nodes and spleen (11, 12, 13), leading to the induction of tolerance in CD4+ (12, 14) and CD8+ (13, 15) T cells. In addition, tumor-specific TCR transgenic CD8+ T cells proliferate in the submandibular lymph nodes when transferred into mice bearing tumors in the a.c. (16). Moreover, P815 tumor cells injected into the a.c. grow progressively but also induce tumor-specific CD8+ CTL responses (17, 18, 19) that eliminate a subsequent injection with P815 in the opposite eye, a phenomenon termed intracamerally induced, concomitant immunity (20). Thus, sequestration of tumor Ags in the a.c., induction of systemic tumor-specific CTL tolerance, or the failure of tumor-specific CTLs to migrate into the eye do not account for immune evasion by certain ocular tumors.

To determine why an immunogenic tumor grows unchecked in the a.c. of the eye in primary ocular tumor development, we used a model of immune evasion using E.G7-OVA, an EL-4 thymoma transduced to express OVA, as a surrogate tumor Ag (21). E.G7-OVA cells can be tracked, enumerated, and isolated in vivo by flow cytometric cell sorting in Thy 1.1 congenic B6.Pl mice because E.G7-OVA expresses the Thy 1.2 allotype. In addition, the E.G7-OVA/B6 model is very useful for understanding mechanisms of immune evasion by primary ocular tumors because OVA-specific T cell responses are induced, immunodominant OVA epitopes for both CD4 (22) and CD8 (21) T cells are very well characterized, and TCR transgenic mice specific for both class I (23) and class II (24) restricted OVA epitopes are available. Our results indicate that E.G7-OVA grow progressively in the a.c. of the eye despite the induction of tumor-specific CD8+ CTL responses. Infiltration of the eye by immunosuppressive CD11b+ myeloid cells may play a role in preventing immune elimination of E.G7-OVA by CD8+ CTL.

Materials and Methods

Experimental animals

Male and female C57BL/6/J (B6; H-2b, Thy1.2+) mice were purchased from the National Cancer Institute (Frederick Cancer Research and Development Center, Frederick, MD) and from The Jackson Laboratory. C57BL/6.PL (B6.PL; H-2b, Thy1.1+), B6;129P2-B2Mtm1Unc/J (B2M−/−; H-2b, Thy1.2+) (25), and B6.129S6-Cd4tm1Knw/J (CD4−/−; H-2b, Thy1.2+) (26) were purchased from The Jackson Laboratory, and C57BL/6J-TgN (TCR-1) (23), also referred to as OT-I, were a gift from Dr. M. Bevan (University of Washington, Seattle, WA). B6.PL, B2M−/−, CD4−/−, and OT-I mice were bred and maintained in the animal facilities at Emory University. B2M−/− and CD4−/− are on a B6 background, and OT-I mice have been backcrossed for >10 generations to B6 mice and are considered congenic with B6 mice. All of the procedures on animals were approved by the Institutional Animal Care and Use Committee at Emory University, conducted according to the principals in the guidelines of the Committee on Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council, Washington, DC) and in adherence to the provisions of the Association for Research in Vision and Ophthalmology (ARVO) Statement for the use of Animals in Ophthalmic and Vision Research.

Tumor cell lines and Ag

EL-4 (H-2b, Thy 1.2+) and EL-4 transduced to express OVA (E.G7-OVA) (21), provided by Dr. M. Bevan (University of Washington, Seattle, WA), or human insulin (EL-4 insulin) (27) were grown in standard growth medium (SGM) (RPMI 1640 medium supplemented with 10% FBS, 2 mM l-glutamine, 1 mM sodium pyruvate, 50 μM 2-ME, gentamicin, penicillin, and streptomycin) at 37°C in a 5% CO2 atmosphere. All cell lines were maintained free of mycoplasma. E.G7-OVA were continuously cultured in 1.0 mg/ml neomycin to maintain the expression of the transfected OVA gene. OVA peptide 257–264 (SIINFEKL) was a gift from Dr. B. Evavold (Emory University, Atlanta, GA) and was dissolved in PBS at a concentration of 3 mM.

Tumor cell and Ag administration

For all injections, mice were anesthetized with of a saline solution containing 1–2 mg of ketamine (Sigma-Aldrich) and 0.2–0.4 mg of xylazine (Bayer) delivered i.m. in the thigh or i.p. For intradermal (i.d.) injections, tumors (104–106 cells) were injected into the skin of the back in 0.2 ml of PBS. For a.c. injections, the eye was anesthetized by topical drops of proparacaine HCl (Alcon Laboratories). The eye was then proptosed using forceps, and the cornea was punctured by insertion of a 30-gauge needle directly above the pupil and almost parallel to the iris. Released aqueous humor was removed with an ophthalmic sponge. A 33-gauge beveled needle fitted on a Hamilton 700 series microsyringe (Hamilton) was then inserted briefly into the incision site to wick the remaining aqueous humor from the a.c., and then the a.c. was filled by injection of 1.5 μl of air. Tumor cells (102–2.0 × 104) contained in 2 μl of PBS were then injected into the a.c. by pressing a 33-gauge blunt needle fitted on a Hamilton 700 series microsyringe against the incision. The air bubble seals the corneal puncture and prevents leaking. In most experiments, only one eye was treated, and the experiment was terminated before tumor growth ruptured the globe. In experiments where both eyes were treated, one eye was enucleated (see below) before rupture of the globe.

Enucleation

Tumor-bearing eyes were removed to minimize stress from progressive growth of ocular tumors that, during the course of the experiment, would rupture the globe. Mice were anesthetized as described above for a.c. injections. The eye was proptosed using forceps, and then the conjunctiva, muscles, and optic nerve were cut with microscissors to remove the eye from the orbit. Bleeding, which was minimal, was stopped by applying direct pressure with a surgical sponge. Mice were monitored postoperatively for signs of pain and distress, such as wasting, hunched posture, or ruffled fur. In multiple experiments, no evidence of distress or infection was observed in any mice, indicating that the procedure was well-tolerated.

Collagenase digestion of eyes and flow cytometric enumeration of tumors and tumor-infiltrating leukocytes (TIL)

Mice were euthanized by cervical dislocation, and their eyes were removed, washed in 2 ml of PBS on ice, and then incubated for 1–2 h at 37°C in a 5% CO2 atmosphere in 2 ml of RPMI 1640 medium supplemented with 58.5 U/ml Collagenase IV (Sigma-Aldrich) and 1% FBS. Eyes were then pressed between frosted glass slides to generate single-cell suspensions. The cell suspension was filtered through a nylon mesh screen to remove debris, washed with PBS, and resuspended in 0.4 ml of FACS buffer (PBS + 1% FBS). Eye suspensions (0.2 ml) were added to individual wells of a 96-well plate for staining. Cells were washed with FACS buffer, and Fc receptors were blocked by incubation with purified CD16/32 Ab (BD Pharmingen) and then incubated with combinations of FITC-, PE-, or APC-labeled anti-CD8α, anti-Thy1.2, anti-Gr-1, anti-CD11b, or anti-CD45 Abs (BD Pharmingen). Following incubation, wells were washed and resuspended in 0.5 ml of FACS buffer supplemented with 20 μl of Viaprobe (BD Pharmingen) and run on a FACSCalibur flow cytometer (BD Biosciences). The forward light scatter channel was set at E-1 with high sample voltage to simultaneously visualize large E.G7-OVA tumor cells and smaller TIL. To enumerate E.G7-OVA and TIL in tumor-challenged eyes, a known sample volume was collected on the flow cytometer at a collection rate, which was determined experimentally to be constant over multiple runs. The number of CD45+ cells in the sample was determined by the following formula: ((Number of CD45 cells collected/collection time)/collection rate (μl/sec)) * 1000 μl.

The number of E.G7-OVA, defined as Thy 1.2+ events in B6.Pl mice, or CD8+, CD11b+, or CD11b+/Gr-1+ cells within the sample was calculated by multiplying the number of CD45+ cells in the sample by the percentage of CD45+ cells of the respective cell population. In some experiments, Thy 1.2+ E.G7-OVA tumors or CD11b+ cells were isolated from collagenase-digested B6.PL eyes by FACSsorting on a FACSVantage (BD Biosciences) or MoFlo flow cytometer (DakoCytomation). CD11b+ cells were enriched from collagenase-digested eye samples before sorting using EasySep magnetic selection (StemCell Technologies). E.G7-OVA tumors were sorted to >95% purity, and CD11b+ cells were sorted to >94% purity. Flow cytometric analysis was performed using FlowJo software (Tree Star).

Ab-mediated CD8+ T cell depletion

Mice were injected with 0.4 mg of anti-CD8 Ab (clone 2.43) (provided by Dr. A. Lukacher, Emory University) i.p. given in daily doses of 0.2 mg, 0.1 mg, and 0.1 mg in a 1 mg/ml solution in sterile PBS before tumor challenge. After tumor challenge, mice were given 0.1 mg of anti-CD8 Ab twice weekly. As a control for Ab administration, another group of mice were given rat IgG (Jackson ImmunoResearch Laboratories).

CTL assay

Splenocytes (3.0 × 107) were stimulated with irradiated (20,000 rad) E.G7-OVA (0.5 × 106 cells) in 10 ml of SGM contained in a 25-cm2 flask. Following incubation for 7 days at 37°C in a 5% CO2 atmosphere, the number of live cells was determined by trypan blue exclusion, and a standard 4-h 51Cr release assay was performed using Na2Cr51O4 (DuPont) labeled E.G7-OVA or EL-4 insulin targets at various E:T ratios. Percentage of lysis was determined by the following formula: (observed – spontaneous release)/(maximal – spontaneous release) × 100.

Histology

Paraffin-embedded 5-μm sections of untreated and tumor-bearing eyes were stained with H&E.

Assay of CTL suppression by myeloid suppressor cells (MSC)

Naive splenocytes from OT-I mice were diluted 1/10 with naive splenocytes from B6.Pl mice to decrease the high frequency of OVA-specific CD8+ T cells in OT-I spleens to a lower more physiological level. Diluted OT-I splenocytes (4.0 × 106) were then added to an individual well of a 24-well plate with or without the addition of isolated myeloid suppressors (0.5–1.0 × 106 cells) and with or without SIINFEKL peptide (0.1 μg/ml) in a total volume of 2 ml of SGM. Following incubation for 4 days at 37°C in a 5% CO2 atmosphere, the number of live cells was determined, and a CTL assay was performed with E.G7-OVA or EL-4 targets at various E:T ratios.

Statistical analysis

Differences in the number of mice developing tumors in the skin were compared by a Fisher exact test. Differences in indicated cell numbers in collagenase-digested eyes or percentage of control in CTL assays were compared by a Student’s t test with equal or unequal variance as determined by a F test, and p values <0.05 were considered statistically significant.

Results

Immunological control of the growth of E.G7-OVA in the skin of B6 mice

Previous studies have shown that E.G7-OVA forms tumors when 106 cells are injected in the skin (28) or 4.0 × 106 cells are injected in the peritoneal cavity (29) of syngeneic B6 mice. Consistent with these observations, B6 mice uniformly developed skin tumors in our hands when injected with ≥105 E.G7-OVA cells in the skin, which grew progressively over time, whereas only a rare mouse developed a tumor when 104 tumor cells were injected into the skin (Fig. 1⇓ and Table I⇓). The failure of E.G7-OVA to form tumors at low doses was largely the result of tumor elimination by the adaptive immune response because 44% of immune-deficient Rag−/− mice developed skin tumors when injected with 104 E.G7-OVA cells, a difference that was statistically significant (p = 0.04) compared with similarly challenged B6 mice.

FIGURE 1.
FIGURE 1.

E.G7-OVA is rejected in the skin by adaptive immunity. Various numbers of E.G7-OVA tumors were injected into the skin of B6 and B6 Rag−/− mice. Tumor size was determined at indicated time points by the product of length and width measurements made with calipers. Symbols indicate the mean tumor size ± 1 SD. Only mice that developed tumors in B6 Rag−/− were included in calculating the mean tumor size.

View this table:
Table I.

E.G7-OVA tumors are rejected in the skin of B6 mice by CD8+ T cellsa

To determine whether T cells were responsible for tumor elimination in the skin, CD8 T cell-deficient (B2M−/−) and CD4 T cell-deficient (CD4−/−) mice were injected with 104 E.G7-OVA cells. B6 and Rag−/− mice that received a similar tumor challenge were included as controls. The majority of Rag−/− mice but only a few immune-competent B6 mice developed tumors, and this difference was statistically significant (Table I⇑). The number of CD4−/− mice that developed tumors was comparable to B6 mice. In contrast, the number of B2M−/− mice that developed tumors was comparable to Rag−/− mice and significantly different from B6 mice. These data suggest that CD8+ T cells are responsible for tumor elimination in the skin and that CD8 T cell induction and expression of effector function is CD4 independent. In addition, 43% of B6 mice in which CD8+ T cells were depleted by Ab developed skin tumors when injected with 104 E.G7-OVA cells, whereas similarly challenged B6 mice given control Ab did not (Table I⇑), further substantiating the contribution of CD8 T cells to tumor regression.

The specificity of tumor immunity in B6 mice was determined in a CTL assay using E.G7-OVA and EL-4-insulin targets. E.G7-OVA was created by transfecting EL-4 with a vector in which OVA and neomycin phosphotransferase (neo) expression, a selectable marker, were expressed under control of ubiquitous promoters (21). A similar expression vector was used to express human insulin and neo genes in EL-4 insulin cells (27). Hence, EL-4 insulin target cells control for CTL activity directed against neo or shared tumor Ags expressed by E.G7-OVA and EL-4. In naive untreated mice, splenic CTL responses were at background levels (Fig. 2⇓). Splenic CTL responses in mice injected in the skin with 106 E.G7-OVA cells preferentially lysed E.G7-OVA but not EL-4 insulin targets, suggesting that immunity was OVA-specific. Tumor-specific CTL responses were not detected in the spleens of mice that eliminated 104 E.G7-OVA cells injected into the skin (data not shown), which likely reflects a limited CTL expansion in the spleen and/or trafficking of the CTL to the tumor site. A larger tumor dose most likely induces sufficient expansion of tumor-specific CTL precursors to reach a detectable level.

FIGURE 2.
FIGURE 2.

Administration of E.G7-OVA in the skin induces OVA-specific CTL. Fourteen days after administration of E.G7-OVA (106 cells) i.d., splenocytes were harvested, stimulated for 1 wk with irradiated E.G7-OVA, and assayed in a standard 4-h CTL assay with the indicated targets. Lytic activity at a 100:1 (Experiment 1) and 200:1 (Experiment 2) E:T ratios are displayed.

Progressive growth of E.G7-OVA in the a.c. of B6 mice

Previous studies have monitored ocular tumor development after the injection of tumor cells in the a.c. by determining the percentage of the a.c. that was filled by cells through slit lamp examination (30). A limitation of this approach is that tumor cells cannot be distinguished from infiltrating leukocytes. We have used an assay using immunofluorescent Abs and flow cytometry of collagenase-digested eyes to enumerate tumor cells and infiltrating leukocytes, simultaneously. Representative results of an experiment (Fig. 3⇓), in which E.G7-OVA cells were added to untreated eye suspensions from B6.PL mice, validate this approach. E.G7-OVA can be easily distinguished from ocular cells because the majority of the E.G7-OVA (>90%) display cell surface expression of CD45 (data not shown), whereas only a minority of cells (<1%) in normal untreated eyes of B6.PL mice are CD45+ (Fig. 3⇓A). The small percentage of CD45+ bone marrow-derived cells in untreated eyes are most likely contained within blood vessels and/or parenchymal tissues of the eye. Although these leukocytes are CD45+ they can be distinguished from tumors because E.G7-OVA expresses the Thy1.2 allotype, whereas no cells in Thy 1.1 congenic B6.Pl mice express Thy 1.2 (Fig. 3⇓B). This experiment also demonstrates that E.G7-OVA do not express CD8, which further distinguishes tumors from infiltrating CTL.

FIGURE 3.
FIGURE 3.

Visualizing E.G7-OVA in the eye by flow cytometric analysis. Cell viability and expression of CD45 were determined in untreated collagenase-digested eyes with and without the addition of 2.0 × 105 CD45+ E.G7-OVA tumors (A). Viaprobe- (live), CD45+ cells, demarcated by a box (numbers indicate the percentage of total cells), were analyzed for expression of Thy 1.2 and CD8α (B). Numbers indicate the percentage of live CD45+ cells. Serial 5-fold dilutions of E.G7-OVA beginning at 106 cells were added to collagenase-digested eyes, and the number of E.G7-OVA was determined (C). Symbols indicate the mean value of three independent experiments in which 5.0 × 104 total events were collected. Error bars represent one SD. The dashed line indicates the threshold of detection for this assay.

To enumerate tumors or infiltrating leukocytes in collagenase-digested eyes, the percentage of CD45+ cells that were tumors or respective TIL was multiplied by the absolute number of CD45+ cells within samples, which was determined by flow cytometric cell counts. To validate this assay, 5-fold dilutions, beginning at 106 E.G7-OVA cells, were added to collagenase-digested eyes, and the number of E.G7-OVA cells (CD45+, Thy1.2+, CD8 negative) was determined. The number of E.G7-OVA cells significantly correlated (r2 = 0.99; slope = 1) with the number of E.G7-OVA cells added to collagenase-digested eyes (Fig. 3⇑C). When 5.0 × 104 total events were collected, the threshold of E.G7-OVA detection was 100 cells.

Using this flow cytometry assay, tumor cell numbers were determined at several time points following injection of various numbers of E.G7-OVA into the a.c. of B6.Pl or B6.Rag−/− mice (Fig. 4⇓A). E.G7-OVA grew progressively after injection with 104 tumor cells in all B6.PL mice and in immunodeficient B6.Rag−/− mice. Tumor growth was confined to the anterior and posterior compartments of the a.c. (compare Fig. 4⇓, B and C). Decreasing the number of cells injected into the eye by 10- to 100-fold delayed the kinetics of tumor cell growth and the length of time for tumors to completely fill the a.c., as has been previously reported (31). However, all of the mice injected with 102 E.G7-OVA developed ocular tumors. These data indicate that tumor-specific CD8+ CTL did not eliminate doses ≥102 tumor cells injected in the a.c. of B6 mice, whereas doses as high as 104 were rejected when injected into the skin.

FIGURE 4.
FIGURE 4.

E.G7-OVA grows progressively in the a.c. of the eye. Absolute numbers of E.G7-OVA were determined at the indicated time points by flow cytometric analysis of collagenase-digested eyes of B6.Pl and B6.Rag−/− mice injected in the a.c. with E.G7-OVA (102–104 cells) (A). Equivalent sample volumes containing varying numbers of events (range, 5.0 × 104–55.9 × 104 events) were collected, with the exception of one day-3 mouse challenged with 104 E.G7-OVA cells in which 5.0 × 104 cells were collected. Each symbol represents the mean tumor number ± 1 SD of 3–7 mice per group. The number of mice with tumors was scored on the final time point shown. Representative H&E-stained whole eye sections from untreated mice (B), and mice injected with 2.0 × 104 E.G7-OVA cells 10 days earlier (C) are shown at ×4 magnification.

Induction of systemic OVA-specific tumor immunity by E.G7-OVA

One potential explanation for the growth of E.G7-OVA in the eye but not the skin is that tumor Ags are sequestered in the eye, and hence the immune system is ignorant of ocular tumors. However, previous studies have shown that mice injected with P815 cells in the a.c. generated tumor-specific CTL responses (19) and were immune to a secondary tumor challenge with P815 cells in the skin or the opposite eye (20). To determine whether a.c. administration of E.G7-OVA induced tumor immunity, B6 mice were injected with E.G7-OVA (5.0 × 103 cells) in the a.c. of eye. Eleven days later, the tumor containing eyes were removed, mice were challenged with 5.0 × 105 E.G7-OVA or EL-4 cells in the skin, and tumor growth was monitored. All of the control mice, which were not exposed to E.G7-OVA in the a.c., developed tumors when challenged with E.G7-OVA or EL-4 in the skin (Table II⇓). By contrast, >90% of the mice previously injected with E.G7-OVA in the a.c. rejected a subsequent challenge with E.G7-OVA. Rejection was OVA specific because E.G7-OVA regressed, but the parental EL-4 cells grew into tumors in most of the mice. E.G7-OVA and EL-4 grew with similar kinetics in the skin of control mice (Fig. 5⇓). However, EL-4 tumors developing in the skin of mice previously primed with E.G7-OVA a.c. were slightly smaller than EL-4 tumors in control mice (Fig. 5⇓). This difference was statistically significant at day 14 (p = 0.05) and day 20 (p = 0.03) and may indicate that a.c. administration of E.G7-OVA also induced weak immune responses to tumor Ags, other than OVA, that are shared by E.G7-OVA and EL-4. However, in another experiment, EL-4 tumors in E.G7-OVA-primed mice were not significantly different in size from EL-4 tumors in control mice (data not shown). Only one mouse that was given E.G7-OVA via the a.c. developed a tumor after challenge with E.G7-OVA in the skin, and this tumor was markedly smaller than E.G7-OVA tumors in control mice (Fig. 5⇓). These data indicate that the administration of E.G7-OVA in the a.c. induced a systemic tumor-specific immune response, demonstrating that tumor Ags were not sequestered in the eye in this model.

FIGURE 5.
FIGURE 5.

E.G7-OVA administered via the a.c. induces primarily OVA-specific tumor immunity. Mice were untreated or injected with E.G7-OVA (5.0 × 103 cells) in the a.c. and challenged with E.G7-OVA or EL-4 (5.0 × 105 cells) i.d. in the skin, and tumor growth was measured. Symbols indicate the mean tumor size ± 1 SD. The number of mice within treatment groups that developed skin tumors is displayed in Table II⇓, experiment (Exp.) 1, and only mice that developed tumors were used to calculate the mean tumor size.

View this table:
Table II.

a.c. administration of E.G7-OVA induces OVA-specific immune responsesa

CD11b+ myeloid cells accumulate in primary ocular tumors

Another potential explanation for immune evasion of E.G7-OVA tumors developing in the a.c. is that tumor-specific CD8+ CTLs may fail to migrate into the eye. To address this possibility, CD45+ leukocytes infiltrating primary ocular tumors were analyzed for CD8, CD11b, and Gr-1 expression at multiple time points after a.c. administration of E.G7-OVA. Ten days after injection of 104 E.G7-OVA cells into the a.c., 62 ± 6% of live CD45+ cells were Thy1.2+ E.G7-OVA tumors, <1% were CD8+, and 38 ± 6% were Thy 1.2 negative and CD8 negative (Fig. 6⇓A), suggesting that CD8+ T cells did not accumulate in the primary ocular tumor. Alternatively, CD8 T cells may have migrated into the eye, but the CD8 molecules down modulated on CTL that were activated in the eye. However, 38 ± 8% of CD45+ cells within the eye were CD11b+ (Fig. 6⇓B), which accounts for both CD8-negative and Thy 1.2-negative populations. Roughly half of the CD45+ cells that expressed CD11b coexpressed Gr-1, a marker expressed by monocytes (32), neutrophils (32), and MSC (33). The number of CD8+ T cells in tumor-containing eyes on day 10 was increased 3-fold from untreated control eyes, but this difference was not statistically significant (p = 0.14), whereas CD11b+ cells were significantly increased at all time points tested and accumulated to almost equal numbers with tumor cells as the tumors grew progressively (Fig. 6⇓C). By day 10, CD11b+ cells that did not express Gr-1 increased 9-fold, and CD11b+ cells that expressed Gr-1 increased 145-fold.

FIGURE 6.
FIGURE 6.

CD11b+ cells accumulate in the a.c. as tumors develop. B6.PL mice were injected with E.G7-OVA (104 cells) in the a.c. of the eye. Tumor growth and leukocyte infiltration were monitored by flow cytometric analysis of collagenase-digested eyes labeled with anti-CD45, anti-Thy1.2, anti-CD8, anti-CD11b, and anti-Gr-1 Abs. Ten days after the a.c. injection of E.G7-OVA, gated CD45+ events in the eye were evaluated for expression of Thy1.2 (E.G7-OVA) and CD8 (A) and CD11b and Gr-1 (B). Plots shown are from one mouse, which is representative of three mice in each treatment group. Absolute numbers of tumors and indicated leukocyte populations within tumor-bearing eyes were determined at indicated time points (C). Equivalent sample volumes containing varying numbers of events (range, 5.0 × 104–37.3 × 104 events) were collected, with the exception of one day-3 mouse challenged with 104 E.G7-OVA cells in which 5.0 × 104 cells were collected. Collagenase-digested eye suspensions used to determine tumor numbers in mice injected with 104 E.G7-OVA via the a.c. in Fig. 4⇑ were also used to determine the absolute numbers of indicated TIL in this figure. E.G7-OVA tumor numbers from Fig. 4⇑ are shown for comparison. Each symbol represents the mean measurement ± 1 SD of indicated cell populations from three to seven individual mice. Day 0 values are from eight untreated eyes of B6.Pl mice in which 2.0 × 105 events were collected for analysis.

Tumor-specific CD8+ CTL can migrate into the eye

The low frequency of CD8+ T cells in primary ocular tumors could result from the failure of large numbers of CD8+ CTL to migrate into the eye or the death of CD8+ CTL within the eye. To determine whether tumor-specific CTL could migrate into the eye, we tested whether tumor-specific immunity induced by administration of E.G7-OVA via the a.c. could cause rejection of a subsequent challenge with E.G7-OVA in the opposite eye. B6.PL mice were given PBS or injected with E.G7-OVA (2.0 × 104 cells) in the a.c. of the right eye. Seven days later, treated eyes were removed, and mice were challenged with E.G7-OVA in the a.c. of the left eye. E.G7-OVA tumors and CD8+ cells were enumerated in the left eyes 4 days after challenge, and representative flow cytometry plots are shown in Fig. 7⇓. The percentage of Thy 1.2+ E.G7-OVA cells was significantly reduced in E.G7-OVA-primed mice compared with mice injected with PBS, indicating that tumor burden was significantly reduced in the opposite eye as the result of priming with E.G7-OVA via the a.c. In contrast, the percentage of CD8+ cells increased dramatically in eyes of mice previously injected with E.G7-OVA in the opposite eye compared with uninjected mice. Similar results were observed when the absolute number of cells in the eyes was determined in multiple mice (Table III⇓). These data further substantiate the interpretation that a.c. administration of E.G7-OVA induced tumor-specific immune responses and demonstrate that tumor-specific CD8+ CTL can migrate into the eye, where they prevented the growth of the tumor presumably by lysing the E.G7-OVA.

FIGURE 7.
FIGURE 7.

Mice primed with E.G7-OVA via the a.c. reject a subsequent E.G7-OVA challenge in the opposite eye. Groups of B6.Pl mice (four per group) were given PBS or E.G7-OVA (2.0 × 104 cells) in the a.c. of the right eye. Seven days later, tumor-bearing eyes were enucleated. Three days later, the mice were injected with E.G7-OVA (4.0 × 103 cells) via the a.c. of the left eye. The frequency of Thy1.2+ E.G7-OVA cells, CD8+, and Thy1.2−/CD8− was determined by flow cytometric analysis 4 days after secondary tumor challenge. Flow cytometry plots display cells contained within a live (Viaprobe negative) CD45+ gate from one eye that is representative of four in each treatment group.

View this table:
Table III.

Intracamerally induced concomitant immunitya

The association of tumor rejection in the eye with increased CD8+ T cell numbers suggests that increasing the frequency of tumor-specific T cells may overcome regional inhibition of CTL responses in the eye. To determine whether increasing the frequency of tumor-specific CD8+ CTL would promote elimination of primary ocular tumors, E.G7-OVA (103 cells) were injected into the a.c. of OT-I mice because >90% of CD8+ T cells in OT-I mice express a TCR that recognizes OVA peptide 257–264 presented by MHC class I H-2 Kb (13). Furthermore, the E.G7-OVA used in this experiment was isolated from collagenase-digested eyes of B6.Pl mice 10–13 days after a.c. administration of tumors to simultaneously evaluate whether any changes in the antigenicity of E.G7-OVA were induced by growing in the eye. The number of E.G7-OVA tumors in eyes of OT-I mice was 86-fold less (Experiment 1) and 54-fold less (Experiment 2) than similarly challenged B6.Pl mice (Fig. 8⇓). It is important to note that E.G7-OVA was defined in OT-I mice as Thy 1.2+CD8− cells, and OT-I mice express Thy 1.2. Therefore, the number of E.G7-OVA in these experiments potentially includes a background number of Thy 1.2+, CD8-negative cells that are not tumors. However, the size and granularity of Thy 1.2+, CD8− cells in tumor-challenged OT-1 mice was equivalent to cultured E.G7-OVA tumor cells and distinct from leukocytes (data not shown), suggesting that Thy 1.2+ tumor cells rather than leukocytes were being measured in the OT-I eyes. Our data suggest that CD8+ CTL infiltrated primary ocular tumors and that increasing the frequency of tumor-specific CD8+ T cells inhibited the growth of primary ocular tumor cells. These data also demonstrate that E.G7-OVA did not lose antigenicity or become immune-suppressive after 13 days of growth in the eye.

FIGURE 8.
FIGURE 8.

E.G7-OVA numbers are reduced in the a.c. of OT-I mice. B6.PL and OT-1 mice (5 mice/group) were injected with E.G7-OVA (103 cells) that was isolated after 10 (Experiment 1) or 13 (Experiment 2) days of growth in the a.c. of B6.Pl mice. Ten (Experiment 1) or 14 days (Experiment 2) later, E.G7-OVA cells were enumerated in collagenase-digested eyes. A total of 5.0 × 104 cells was collected for analysis. E.G7-OVA were defined as CD45+/Thy1.2+/CD8-negative cells in OT-I mice that express the Thy 1.2 allotype. The number of CD45+/Thy1.2+/CD8-negative cells in collagenase-digested eyes (n = 6) from three naive OT-I mice was 3.2 ± 0.8 × 103 cells. A total of 105 events was collected for analysis.

The next experiment was designed to determine whether the decreased frequency of E.G7-OVA in the eyes of OT-I, compared with B6.PL, mice correlated with an increase in the number of CD8+ T cells and an increase or decrease in the number of CD11b+ cells infiltrating the eye. The number of CD8+ T cells detected in the eyes of OT-1 mice on day 0 was 15-fold greater (Fig. 9⇓B) than that in B6.PL mice (Fig. 6⇑C). Seven days after injection of 104 E.G7-OVA, 33-fold more CD8+ T cells were detected in OT-I than B6.PL eyes, whereas 45-fold more E.G7-OVA cells were detected in B6.PL than OT-1 eyes (Fig. 9⇓A), which confirms our previous experiments. This suggests that increasing the number of tumor-specific CD8+ T cells within the eye contributes to reducing the number of tumor cells. At day 7, the number of CD11b+Gr-1+ and CD11b+Gr-1− cells were significantly less in OT-1 than B6.Pl recipients. Similar results were observed in OT-1 mice on day 7 in a second experiment (Fig. 9⇓B). At this time point, the eyes of OT-I mice were almost indistinguishable from naive unchallenged mice by visual inspection, and only a few mice displayed very slight a.c. cloudiness. E.G7-OVA tumor cells increased 11-fold from day 7 to day 9 (Fig. 9⇓A) and 49-fold on day 13 in comparison to background numbers of Thy 1.2+, CD8− cells in OT-I T cells at day 0 (Fig. 9⇓B), whereas the number of CD8+ T cells remained fairly constant and similar to the number observed in OT-I mice before tumor cells were injected. By day 13, the number of tumor cells in the OT-1 recipients was similar to that observed in B6.PL mice on day 7, which indicates that increased numbers of CD8+ T cells in OT-1 mice delayed, but did not prevent, the growth of E.G7-OVA. Both CD11b+Gr-1− and CD11b+Gr-1+ cells increased in parallel with the increasing growth of E.G7-OVA, but the number of CD11b+Gr-1+ cells was virtually the same as the number of tumor cells in both strains of mice at all times after tumor injection, suggesting that there might be a causal relationship between these cells.

FIGURE 9.
FIGURE 9.

Increased numbers of CD8+ T cells and decreased numbers of CD11b+ myeloid cells is associated with delayed kinetics of tumor growth in the a.c. of OT-I mice. B6.Pl and OT-I mice (5 mice/group) were injected with 104 E.G7-OVA in the a.c. Seven and 9 days later, tumors and indicated leukocyte populations were enumerated in collagenase-digested eyes (A). The kinetics of tumor growth and leukocyte infiltration in OT-I mice injected with 104 E.G7-OVA in the a.c. are shown (B). Each symbol represents the mean measurement ± 1 SD of indicated cell populations from four to nine individual mice. Day 0 values are from six untreated eyes of three OT-I mice in which 105 events were collected for analysis.

Myeloid cells within primary ocular tumors suppress CD8+ CTL responses

CD11b+ cells have been shown to inhibit tumor-specific CTL responses in other murine tumor models (33, 34, 35, 36, 37), raising the possibility that CD11b+ cells may be inhibiting tumor-specific CD8+ CTL responses. To test this hypothesis, CD11b+ cells were sorted from tumor-bearing eyes or normal spleen cells by flow cytometry and then added to cultures of spleen cells from naive OT-I mice that were stimulated with the cognate SIINFEKL peptide. CTL responses were measured 4 days later (Fig. 10⇓A). CTL cultures that were stimulated with SIINFEKL developed CTL activity against OVA-expressing E.G7-OVA targets, but not against the parental cell line EL-4 (data not shown), in comparison to unstimulated cultures. In five independent experiments (Fig. 10⇓B), CD11b+ cells isolated from ocular tumors significantly inhibited the lytic activity of peptide-stimulated cultures (37 ± 27% of control). Lytic activity was also significantly reduced after the addition of naive splenic CD11b+ cells (68 ± 29% of control), but their specific activity was slightly, albeit not significantly (p = 0.12), less than that of the CD11b+ cells from the eye of tumor-bearing mice. These data suggest that myeloid cells infiltrating primary ocular tumors may suppress CD8+ CTL responses, thereby contributing to primary ocular tumor growth.

FIGURE 10.
FIGURE 10.

CD11b+ cells isolated from ocular tumors suppress CTL responses. OT-I splenocytes (4.0 × 106 cells) were stimulated with or without SIINFEKL peptide (0.1 μg/ml) and with or without the addition of CD11b+ regulators (1.0 × 106 cells) that were isolated by flow cytometric cell sorting from spleens of naive B6 mice or from collagenase-digested eyes of B6.Pl mice injected with E.G7-OVA (5.0 × 103 cells) in the a.c. 10 days earlier. CTL responses were assayed with E.G7-OVA targets at the indicated E:T ratios 4 days later (A). The lytic activity in OT-I cultures with the addition of CD11b+ regulators (0.5–1.0 × 106 cells) from the spleen or eye was normalized to the lytic activity of OT-I cultures without regulators at the same E:T ratio (50:1–67:1) within the same experiment and is presented as percentage of control (B). Each symbol represents an independent experiment.

Discussion

Administration of E.G7-OVA via the a.c. generated OVA-specific CD8+ CTL responses systemically that eliminated a subsequent E.G7-OVA challenge in the skin and opposite eye but failed to eliminate the primary ocular tumor. The generation of CTL effectors after the administration of tumors in the eye is thought to be a two-step process involving expansion of CTL precursors in secondary lymphoid organs that differentiate into lytic effectors upon subsequent encounter with Ag at the site of tumor injection (38, 39). We observed a slight increase in CD8+ T cells within primary ocular tumors, and E.G7-OVA tumor growth was retarded in the a.c. of OT-I mice in which the majority of CD8+ T cells express an OVA-specific TCR. Delayed kinetics of primary ocular tumor growth in OT-I mice correlated with increased numbers of CD8+ T cells within the eye, suggesting that tumor-specific CTL can infiltrate primary ocular tumors. Because the frequency of OVA-specific CTL precursors in OT-I mice is markedly increased in comparison to B6 mice, these data suggest that the tumor cell:effector CTL ratio is critical for tumor elimination in the eye. This is compatible with the observation that P815 cells injected into the a.c. were completely eliminated when activated P815-specific CD8+ CTL effectors were coinjected at a 1:1, 3:1, or 8:1 tumor cell:effector T cell ratios (31) but not when the ratio exceeded 60:1 (31). Therefore, the growth of E.G7-OVA tumors in the eye and not the skin may be the result of fewer CD8+ CTL within primary ocular tumors compared with the numbers in primary skin tumors. Alternatively, preventing the growth of tumors in the eye may require a higher effector:tumor ratio than the skin because the various factors that maintain ocular immune privilege (reviewed in Refs. 40, 41) interfere with the lytic activity of the CD8+ T cells on a per cell basis (13, 39). In our experiments, primary ocular tumor growth was equivalent in B6 and in Rag−/−, suggesting that the few CD8+ T cells within primary ocular tumors of B6 mice were functionally inhibited. However, a.c. administration of E.G7-OVA did not induce systemic CD8+ T cell tolerance but rather primed for CD8+ CTL responses, as has been previously reported for P815 mastocytomas that had been injected into the a.c. (19).

E.G7-OVA tumors grew progressively in the a.c. of B6.PL mice and with delayed kinetics in OT-I mice despite the infiltration of CD11b+ cells that accumulated at numbers almost equal to the tumor cells. These data suggest that CD11b+ cells within primary ocular tumors are not tumoricidal, although CD11b+ cells are known to have tumoricidal properties under other circumstances. In addition, the CD11b+ cells within primary ocular tumors were particularly efficient suppressors of CD8+ CTL responses in vitro. Therefore, these CD11b+ cells may be analogous to MSC, which have been shown to promote tumor progression by suppressing CD8+ CTL responses systemically and regionally within the tumor microenvironment (reviewed by Serafini et al. (42)).

MSC accumulate in peripheral blood of patients with malignancies (43, 44) and accumulate in the spleen (33, 34, 35, 36, 37), and within tumors (45) in several different murine tumor models. The mechanism by which tumor-derived CD11b+ T cells suppress CTL responses in our system is not currently understood. However, MSC might inhibit tumor-specific CD8+ CTL activity by altering l-arginine metabolism (reviewed by Bronte et al. (46)) via inducible NO synthase or arginase I, which promotes CTL apoptosis (45, 47) or interferes with CTL signal transduction (48) in other murine tumor models. The low frequency of CD8+ T cells within primary ocular tumors would be consistent with CD8+ T cell apoptosis within the eye mediated via NO production by CD11b+ cells. This interpretation is supported by the observation that inhibition of NO production, by administration of NG-nitro-l-arginine methyl ester in an experimental model of autoimmune uveoretinitis, decreased T cell apoptosis in the retina (49).

CD11b+ cells infiltrating primary ocular tumors in our experiments were both Gr-1+ and Gr-1− cells, which were also reported to accumulate in the eyes of mice injected in the a.c. with tumors that express membrane-bound Fas ligand, where they are associated with inflammatory responses that eliminated the tumor (50). In addition, Ab-mediated depletion of CD11b+ cells mitigated endotoxin-induced uveitis, suggesting that CD11b+ cells contribute to the pathology of this disease (51). These data indicate that not all Gr-1+, CD11b+, or CD11b/Gr-1 double-positive cells are MSC. In our experiments, CD11b+ cells from naive spleens also suppressed CTL responses, suggesting that a suppressor phenotype may be the default pathway of CD11b+ cells. Environmental signals associated with inflammation may promote differentiation of CD11b+Gr-1+ cells into mature neutrophils, which are tumoricidal. For example, Gr-1+ MSC lost suppressive activity when cultured with all trans-retinoic acid, a regulator of cell differentiation that induces promyelocytes to differentiate into mature neutrophils (52). Recently, Gregory et al. (53) have presented a novel treatment for ocular tumors involving the injection of membrane Fas ligand vesicles into the eye, which induced neutrophil infiltration and promoted ocular tumor elimination. This treatment might also promote the differentiation of MSC within ocular tumors into neutrophils that are tumoricidal.

In our experiments, a.c. administration of tumors induced tumor-specific immunity that eliminated tumors subsequently injected in the opposite eye, suggesting that these protective tumor-specific CD8+ CTL responses may be inhibited by MSC within primary ocular tumors. MSC cells accumulate after immunization (54) and in chronic infection (55). Therefore, the normal physiological function of these cells may be immune suppression to prevent immunopathology. Consistent with this hypothesis, a large number of CD45+ cells that were not tumors and not CD8+ CTL accumulated in the eyes of mice previously primed with E.G7-OVA via the a.c., which displayed reduced tumor burden after a subsequent injection of E.G7-OVA in the opposite eye in comparison to unprimed mice (Fig. 7⇑ and Table III⇑). In preliminary experiments, we have found that CD11b+/Gr-1+ cells accumulate in these eyes but have not determined whether they are MSC or tumoricidal neutrophils. Hence, the ratio of MSC:effector CD8+ CTL may be critical in determining whether CTL are effective at eliminating tumors. CD11b+ cells accumulated with early kinetics in primary ocular tumors (Fig. 6⇑) at a time when CD8+ CTL precursors were most likely expanding and differentiating in secondary lymphoid organs. Therefore, when tumor-specific CD8+ CTL migrated into primary ocular tumors, they encountered a very high frequency of CD11b+ MSC. In contrast, CD8+ CTL may outnumber CD11b+ MSC after a secondary challenge with tumors in the opposite eye of primed mice because CD8+ CTL precursors have already expanded to the first tumor challenge. In support of this hypothesis, primary ocular tumors grew with delayed kinetics in OT-I mice, which was associated with increased numbers of CD8+ T cells within the eye in comparison to B6.Pl mice. Seven days after a.c. administration of E.G7-OVA to OT-I mice, the number of CD11b+/Gr-1+ cells and CD8+ T cells within eyes was equivalent, and tumor burden was low. However, on day 13, tumor burden increased and CD11b+/Gr-1+ cells outnumbered CD8+ T cells. The observation that CD11b+ cells isolated from tumor-containing eyes of B6.Pl mice inhibited CD8+ CTL responses in vitro suggests that CD11b+ cells may inhibit tumor-specific CD8+ T cell within the primary ocular tumor microenvironment in vivo.

Accumulation of MSC within human primary ocular tumors has not been reported. However, patients with uveal melanomas containing tumor-associated macrophages (TAM) were associated with larger tumors of the more aggressive epithelioid type and a poor prognosis in one study (56). These TAM may be immune suppressive because CD11b+ Gr-1+ MSC have been shown to differentiate into F4/80+ TAM after transfer into tumor-bearing mice, which suppress tumor-specific CD8+ CTL, and thereby promote tumor growth (45). Chemotherapeutic treatments that inhibit or delete MSC, such as Gemcitibine (35), may promote primary ocular tumor elimination by restoring tumor-specific CD8+ CTL activity within the eye.

Acknowledgments

We thank Dr. Michael Bevan (University of Washington, Seattle, WA) for providing E.G7-OVA; Dr. Brian Evavold (Emory University, Atlanta, GA) for providing the OVA peptide 257–264 (SIINFEKL); Dr. Aron Lukacher (Emory University, Atlanta, GA) for the anti-CD8 Ab; Robert Karaffa, Mike Hulsey, and Leah Watson (Emory University, Atlanta, GA) for excellent technical assistance in the isolation of CD11b+ cells; the L. F. Montgomery Eye Pathology Laboratory (Emory University, Atlanta, GA) for processing and preparing histological sections; and Dr. Hans Grossniklaus for reading histological sections and reviewing this manuscript.

Disclosures

The authors have no financial conflict of interest.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • ↵1 This work was supported by a research grant from the Georgia Knights Templar Educational Foundation, a gift from Malcolm and Musette Powell, a grant from the Foundation for Fighting Blindness, a National Eye Institute core grant (P30EYO06360), a grant from Research to Prevent Blindness, and research grants (EY13459 and EY014877) from National Institutes of Health. K.C.M. is the recipient of a National Research Service Award Grant F32 EY07079 from the National Eye Institute. J.A.K. is the recipient of the Senior Scientific Investigator award from Research to Prevent Blindness.

  • ↵2 Address correspondence and reprint requests to Dr. Kyle C. McKenna, Department of Ophthalmology, Emory University, Clinic B, Room B2514, 1365 Clifton Road Northeast, Atlanta, GA 30322. E-mail address: kmckenn{at}emory.edu

  • ↵3 Abbreviations used in this paper: a.c., anterior chamber; SGM, standard growth medium; i.d., intradermal; TIL, tumor-infiltrating leukocytes; MSC, myeloid suppressor cell; neo, neomycin phosphotransferase; TAM, tumor-associated macrophage.

  • Received January 17, 2006.
  • Accepted May 4, 2006.

References

  1. ↵
    Albert, D. M., A. S. Niffenegger, J. K. Willson. 1992. Treatment of metastatic uveal melanoma: review and recommendations. Surv. Ophthalmol. 36: 429-438.
    OpenUrlCrossRefPubMed
  2. ↵
    Rajpal, S., R. Moore, C. P. Karakousis. 1983. Survival in metastatic ocular melanoma. Cancer 52: 334-336.
    OpenUrlCrossRefPubMed
  3. ↵
    Kujala, E., T. Makitie, T. Kivela. 2003. Very long-term prognosis of patients with malignant uveal melanoma. Invest. Ophthalmol. Vis. Sci. 44: 4651-4659.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    van Dinten, L. C., N. Pul, A. F. van Nieuwpoort, C. J. Out, M. J. Jager, P. J. van den Elsen. 2005. Uveal and cutaneous melanoma: shared expression characteristics of melanoma-associated antigens. Invest. Ophthalmol. Vis. Sci. 46: 24-30.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    van Duinen, S. G., D. J. Ruiter, E. B. Broecker, E. A. van der Velde, C. Sorg, K. Welvaart, S. Ferrone. 1988. Level of HLA antigens in locoregional metastases and clinical course of the disease in patients with melanoma. Cancer Res. 48: 1019-1025.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Rosenberg, S. A., J. C. Yang, D. J. Schwartzentruber, P. Hwu, F. M. Marincola, S. L. Topalian, N. P. Restifo, M. E. Dudley, S. L. Schwarz, P. J. Spiess, et al 1998. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat. Med. 4: 321-327.
    OpenUrlCrossRefPubMed
  7. ↵
    Anastassiou, G., V. Rebmann, S. Wagner, N. Bornfeld, H. Grosse-Wilde. 2003. Expression of classic and nonclassic HLA class I antigens in uveal melanoma. Invest. Ophthalmol. Vis. Sci. 44: 2016-2019.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Dithmar, S. A., D. A. Rusciano, C. A. Armstrong, M. J. Lynn, H. E. Grossniklaus. 1999. Depletion of NK cell activity results in growth of hepatic micrometastases in a murine ocular melanoma model. Curr. Eye Res. 19: 426-431.
    OpenUrlCrossRefPubMed
  9. ↵
    Niederkorn, J. Y., J. W. Streilein, J. A. Shadduck. 1981. Deviant immune responses to allogeneic tumors injected intracamerally and subcutaneously in mice. Invest. Ophthalmol. Vis. Sci. 20: 355-363.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Medawar, P.. 1948. Immunity to homologous grafted skin: III. The fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br. J. Exp. Pathol. 29: 58-69.
    OpenUrlPubMed
  11. ↵
    Egan, R. M., C. Yorkey, R. Black, W. K. Loh, J. L. Stevens, J. G. Woodward. 1996. Peptide-specific T cell clonal expansion in vivo following immunization in the eye, an immune-privileged site. J. Immunol. 157: 2262-2271.
    OpenUrlAbstract
  12. ↵
    Perez, V. L., A. Biuckians, J. W. Streilein. 2000. In-vivo impaired T helper 1 cell development in submandibular lymph nodes due to Il-12 deficiency following antigen injection into the anterior chamber of the eye. Ocul. Immunol. Inflamm. 8: 9-24.
    OpenUrlCrossRefPubMed
  13. ↵
    McKenna, K. C., Y. Xu, J. A. Kapp. 2002. Injection of soluble antigen into the anterior chamber of the eye induces expansion and functional unresponsiveness of antigen-specific CD8+ T cells. J. Immunol. 169: 5630-5637.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Mizuno, K., A. F. Clark, J. W. Streilein. 1989. Anterior chamber-associated immune deviation induced by soluble antigens. Invest. Ophthalmol. Vis. Sci. 30: 1112-1119.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Xu, Y., J. A. Kapp. 2002. γδ T cells in anterior chamber-induced tolerance in CD8+ CTL responses. Invest. Ophthalmol. Vis. Sci. 43: 3473-3479.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Boonman, Z. F., G. J. van Mierlo, M. F. Fransen, K. L. Franken, R. Offringa, C. J. Melief, M. J. Jager, R. E. Toes. 2004. Intraocular tumor antigen drains specifically to submandibular lymph nodes, resulting in an abortive cytotoxic T cell reaction. J. Immunol. 172: 1567-1574.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Niederkorn, J. Y., J. W. Streilein. 1983. Alloantigens placed into the anterior chamber of the eye induce specific suppression of delayed-type hypersensitivity but normal cytotoxic T lymphocyte and helper T lymphocyte responses. J. Immunol. 131: 2670-2674.
    OpenUrlAbstract
  18. ↵
    Niederkorn, J. Y., J. W. Streilein. 1984. Adoptive transfer of immunity to intraocular tumors in mice. Invest. Ophthalmol. Vis. Sci. 25: 336-342.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Ksander, B. R., J. W. Streilein. 1989. Analysis of cytotoxic T cell responses to intracameral allogeneic tumors. Invest. Ophthalmol. Vis. Sci. 30: 323-329.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Niederkorn, J. Y., J. W. Streilein. 1983. Intracamerally induced concomitant immunity: mice harboring progressively growing intraocular tumors are immune to spontaneous metastases and secondary tumor challenge. J. Immunol. 131: 2587-2594.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Moore, M. W., F. R. Carbone, M. J. Bevan. 1988. Introduction of soluble protein into the class I pathway of antigen processing and presentation. Cell 54: 777-785.
    OpenUrlCrossRefPubMed
  22. ↵
    Robertson, J. M., P. E. Jensen, B. D. Evavold. 2000. DO11.10 and OT-II T cells recognize a C-terminal ovalbumin 323–339 epitope. J. Immunol. 164: 4706-4712.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Hogquist, K. A., S. C. Jameson, W. R. Heath, J. L. Howard, M. J. Bevan, F. R. Carbone. 1994. T cell receptor antagonist peptides induce positive selection. Cell 76: 17-27.
    OpenUrlCrossRefPubMed
  24. ↵
    Barnden, M. J., J. Allison, W. R. Heath, F. R. Carbone. 1998. Defective TCR expression in transgenic mice constructed using cDNA-based α- and β-chain genes under the control of heterologous regulatory elements. Immunol. Cell Biol. 76: 34-40.
    OpenUrlCrossRefPubMed
  25. ↵
    Koller, B. H., P. Marrack, J. W. Kappler, O. Smithies. 1990. Normal development of mice deficient in β 2M, MHC class I proteins, and CD8+ T cells. Science 248: 1227-1230.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    McCarrick, J. W., III, J. R. Parnes, R. H. Seong, D. Solter, B. B. Knowles. 1993. Positive-negative selection gene targeting with the diphtheria toxin A-chain gene in mouse embryonic stem cells. Transgenic Res. 2: 183-190.
    OpenUrlCrossRefPubMed
  27. ↵
    Ma, H., Y. Ke, Q. Li, J. A. Kapp. 2000. Bovine and human insulin activate CD8+-autoreactive CTL expressing both type 1 and type 2 cytokines in C57BL/6 mice. J. Immunol. 164: 86-92.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Ke, Y., L. M. Kapp, J. A. Kapp. 2003. Inhibition of tumor rejection by γδ T cells and IL-10. Cell. Immunol. 221: 107-114.
    OpenUrlCrossRefPubMed
  29. ↵
    Shrikant, P., M. Mescher. 1999. Control of syngeneic tumor growth by activation of CD8+ T cells: efficacy is limited by migration away from the site and induction of nonresponsiveness. J. Immunol. 162: 2858-2866.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Clark, A. F., J. Mellon, X. Y. Li, D. Ma, H. Leher, R. Apte, H. Alizadeh, S. Hegde, A. McLenaghan, E. Mayhew, et al 1999. Inhibition of intraocular tumor growth by topical application of the angiostatic steroid anecortave acetate. Invest. Ophthalmol. Vis. Sci. 40: 2158-2162.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    Miki, S., B. Ksander, J. W. Streilein. 1993. Complete elimination (‘cure’) of progressively growing intraocular tumors by local injection of tumor-specific CD8+ T lymphocytes. Invest. Ophthalmol. Vis. Sci. 34: 3622-3634.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Lagasse, E., I. L. Weissman. 1996. Flow cytometric identification of murine neutrophils and monocytes. J. Immunol. Methods 197: 139-150.
    OpenUrlCrossRefPubMed
  33. ↵
    Gabrilovich, D. I., M. P. Velders, E. M. Sotomayor, W. M. Kast. 2001. Mechanism of immune dysfunction in cancer mediated by immature Gr-1+ myeloid cells. J. Immunol. 166: 5398-5406.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Rodriguez, P. C., D. G. Quiceno, J. Zabaleta, B. Ortiz, A. H. Zea, M. B. Piazuelo, A. Delgado, P. Correa, J. Brayer, E. M. Sotomayor, et al 2004. Arginase I production in the tumor microenvironment by mature myeloid cells inhibits T-cell receptor expression and antigen-specific T-cell responses. Cancer Res. 64: 5839-5849.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Suzuki, E., V. Kapoor, A. S. Jassar, L. R. Kaiser, S. M. Albelda. 2005. Gemcitabine selectively eliminates splenic Gr-1+/CD11b+ myeloid suppressor cells in tumor-bearing animals and enhances antitumor immune activity. Clin. Cancer Res. 11: 6713-6721.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Terabe, M., S. Matsui, J. M. Park, M. Mamura, N. Noben-Trauth, D. D. Donaldson, W. Chen, S. M. Wahl, S. Ledbetter, B. Pratt, et al 2003. Transforming growth factor-β production and myeloid cells are an effector mechanism through which CD1d-restricted T cells block cytotoxic T lymphocyte-mediated tumor immunosurveillance: abrogation prevents tumor recurrence. J. Exp. Med. 198: 1741-1752.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Sinha, P., V. K. Clements, S. Ostrand-Rosenberg. 2005. Reduction of myeloid-derived suppressor cells and induction of M1 macrophages facilitate the rejection of established metastatic disease. J. Immunol. 174: 636-645.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Ksander, B. R., J. W. Streilein. 1989. Recovery of activated cytotoxic T cells from minor H incompatible tumor graft rejection sites. J. Immunol. 143: 426-431.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Ksander, B. R., J. W. Streilein. 1990. Failure of infiltrating precursor cytotoxic T cells to acquire direct cytotoxic function in immunologically privileged sites. J. Immunol. 145: 2057-2063.
    OpenUrlAbstract
  40. ↵
    Streilein, J. W.. 2003. Ocular immune privilege: therapeutic opportunities from an experiment of nature. Nat. Rev. Immunol. 3: 879-889.
    OpenUrlCrossRefPubMed
  41. ↵
    McKenna, K. C., J. A. Kapp. 2004. Ocular immune privilege and CTL tolerance. Immunol. Res. 29: 103-112.
    OpenUrlCrossRefPubMed
  42. ↵
    Serafini, P., S. C. De, I. Marigo, S. Cingarlini, L. Dolcetti, G. Gallina, P. Zanovello, V. Bronte. 2004. Derangement of immune responses by myeloid suppressor cells. Cancer Immunol. Immunother. 53: 64-72.
    OpenUrlCrossRefPubMed
  43. ↵
    Zea, A. H., P. C. Rodriguez, M. B. Atkins, C. Hernandez, S. Signoretti, J. Zabaleta, D. McDermott, D. Quiceno, A. Youmans, A. O’Neill, et al 2005. Arginase-producing myeloid suppressor cells in renal cell carcinoma patients: a mechanism of tumor evasion. Cancer Res. 65: 3044-3048.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Almand, B., J. I. Clark, E. Nikitina, J. van Beynen, N. R. English, S. C. Knight, D. P. Carbone, D. I. Gabrilovich. 2001. Increased production of immature myeloid cells in cancer patients: a mechanism of immunosuppression in cancer. J. Immunol. 166: 678-689.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    Kusmartsev, S., D. I. Gabrilovich. 2005. STAT1 signaling regulates tumor-associated macrophage-mediated T cell deletion. J. Immunol. 174: 4880-4891.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Bronte, V., P. Zanovello. 2005. Regulation of immune responses by l-arginine metabolism. Nat. Rev. Immunol. 5: 641-654.
    OpenUrlCrossRefPubMed
  47. ↵
    Saio, M., S. Radoja, M. Marino, A. B. Frey. 2001. Tumor-infiltrating macrophages induce apoptosis in activated CD8+ T cells by a mechanism requiring cell contact and mediated by both the cell-associated form of TNF and nitric oxide. J. Immunol. 167: 5583-5593.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    Rodriguez, P. C., A. H. Zea, J. DeSalvo, K. S. Culotta, J. Zabaleta, D. G. Quiceno, J. B. Ochoa, A. C. Ochoa. 2003. l-arginine consumption by macrophages modulates the expression of CD3 ζ-chain in T lymphocytes. J. Immunol. 171: 1232-1239.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Liversidge, J., A. Dick, S. Gordon. 2002. Nitric oxide mediates apoptosis through formation of peroxynitrite and Fas/Fas-ligand interactions in experimental autoimmune uveitis. Am. J. Pathol. 160: 905-916.
    OpenUrlCrossRefPubMed
  50. ↵
    Gregory, M. S., A. C. Repp, A. M. Holhbaum, R. R. Saff, A. Marshak-Rothstein, B. R. Ksander. 2002. Membrane Fas ligand activates innate immunity and terminates ocular immune privilege. J. Immunol. 169: 2727-2735.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    Whitcup, S. M., L. R. DeBarge, H. Rosen, R. B. Nussenblatt, C. C. Chan. 1993. Monoclonal antibody against CD11b/CD18 inhibits endotoxin-induced uveitis. Invest. Ophthalmol. Vis. Sci. 34: 673-681.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    Miyauchi, J.. 1999. All-trans retinoic acid and hematopoietic growth factors regulating the growth and differentiation of blast progenitors in acute promyelocytic leukemia. Leuk. Lymphoma 33: 267-280.
    OpenUrlPubMed
  53. ↵
    Gregory, M. S., S. Koh, E. Huang, R. R. Saff, A. Marshak-Rothstein, S. Mukai, B. R. Ksander. 2005. A novel treatment for ocular tumors using membrane FasL vesicles to activate innate immunity and terminate immune privilege. Invest. Ophthalmol. Vis. Sci. 46: 2495-2502.
    OpenUrlAbstract/FREE Full Text
  54. ↵
    Bronte, V., M. Wang, W. W. Overwijk, D. R. Surman, F. Pericle, S. A. Rosenberg, N. P. Restifo. 1998. Apoptotic death of CD8+ T lymphocytes after immunization: induction of a suppressive population of Mac-1+/Gr-1+ cells. J. Immunol. 161: 5313-5320.
    OpenUrlAbstract/FREE Full Text
  55. ↵
    Noel, W., G. Raes, G. G. Hassanzadeh, B. P. De, A. Beschin. 2004. Alternatively activated macrophages during parasite infections. Trends Parasitol. 20: 126-133.
    OpenUrlCrossRefPubMed
  56. ↵
    Makitie, T., P. Summanen, A. Tarkkanen, T. Kivela. 2001. Tumor-infiltrating macrophages (CD68+ cells) and prognosis in malignant uveal melanoma. Invest. Ophthalmol. Vis. Sci. 42: 1414-1421.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section