Immunological rejection is the major cause of human corneal allograft failure. We hypothesized that local production of IL-4 or the p40 subunit of IL-12 (p40 IL-12) by the grafted cornea might prolong allograft survival. Replication-deficient adenoviral vectors encoding ovine IL-4 or p40 IL-12 and GFP were generated and used to infect ovine corneas ex vivo. mRNA for each cytokine was detected in infected corneas, and the presence of secreted protein in corneal supernatants was confirmed by bioassay (for IL-4) or immunoprecipitation (for p40 IL-12). Sheep received uninfected or gene-modified orthotopic corneal allografts. Postoperatively, untreated corneas (n = 13) and corneas expressing GFP (n = 6) were rejected at a median of 21 and 20 days, respectively. Corneas expressing IL-4 (n = 6) underwent rejection at 18.5 days (p > 0.05 compared with controls) and histology demonstrated the presence of eosinophils. In contrast, corneas expressing p40 IL-12 (n = 9) showed prolonged allograft survival (median day to rejection = 45 days, p = 0.003). Local intraocular production of p40 IL-12 thus prolonged corneal graft survival significantly, but local production of the prototypic immunomodulatory cytokine IL-4 induced eosinophilia, inflammation, and rejection. These findings have important implications for the development of novel strategies to improve human corneal graft survival.

Despite the immune-privileged nature of the eye (1, 2), effector immune responses occur within the ocular environment and irreversible rejection is the major cause of human corneal allograft failure (3). The majority of the available evidence suggests that corneal allograft rejection is a CD4-positive T cell-dependent response (4, 5) during which proinflammatory cytokines, including IFN-γ, are expressed within the eye (6, 7, 8).

Gene transfer allows the modification of tissues for subsequent transplantation. We hypothesized that the expression of cDNA encoding an immunomodulatory cytokine by a donor cornea might prevent corneal allograft rejection. We have previously shown that transfer of cDNA encoding ovine IL-10 to donor corneal endothelium ex vivo can prolong corneal allograft survival significantly (9), although a beneficial effect was not observed in all grafts. We were thus interested to investigate whether local production of other immunomodulatory cytokines would be more effective than IL-10 in prolonging corneal graft survival.

We selected IL-4 and the p40 subunit of IL-12 for further study. IL-4, considered central to the development of Th2 responsiveness, is frequently detected in surviving allografts (10, 11) and has been associated with the acquisition of tolerance (12), although its expression in surviving and rejecting allografts is extremely variable (10, 13, 14). IL-4 may play what has been described as a conditional role in permitting the development of regulatory cells (15) and in supporting allograft survival (15, 16, 17, 18), but may not be obligatory for graft survival (13, 14, 16, 19, 20). The proinflammatory cytokine IL-12 promotes the proliferation of Th1-type cells and subsequent production of IFN-γ (21). Biologically active IL-12 is a glycosylated heterodimer of 70 kDa, consisting of disulfide-linked 40- and 35-kDa subunits (22). The genes encoding each chain map to different chromosomes, and their expression is regulated independently (23). Both subunits need to be produced by the same cell to result in functional p70 IL-12 (24). The p40 IL-12 subunit is produced in vast excess over p35 IL-12 and acts as a natural antagonist for IL-12 in vivo (25, 26, 27).

We constructed adenoviral vectors encoding either full-length ovine IL-4 or p40 IL-12, used them to infect donor ovine corneas ex vivo, and transplanted the gene-modified corneas to the eyes of outbred sheep to investigate the effects of local production of IL-4 and p40 IL-12 within the eye on corneal allograft outcome.

Ovine p40 IL-12 and ovine IL-4 cDNAs have been described previously (28, 29). A replication-deficient E1-, E3-deleted adenovirus type 5 encoding GFP under the transcriptional control of a CMV promoter (Ad-GFP) was the kind gift of Prof. B. Vogelstein (Johns Hopkins University, Baltimore, MD). Two replication-deficient E1-, E3-deleted adenovirus type 5 vectors encoding either ovine p40 IL-12 and GFP (Ad-p40) or ovine IL-4 and GFP (Ad-IL-4) under the transcriptional control of separate CMV promoters were generated using the AdEasy system (30). Maps of the recombinant plasmids are shown in Fig. 1. Restriction enzyme analyses demonstrated the presence of the insert for IL-4 (408 bp) or p40-IL-12 (1040 bp) in the recombined infectious plasmids. Each recombinant plasmid was linearized by restriction enzyme digestion with PacI and transfected into 293 cells (CRL 1573; American Type Culture Collection) using liposomal gene transfer. The success of the transfection and the progress of virus replication were monitored by assessing GFP expression in transfected cells at the fluorescence microscope. Vectors were propagated in E1A, E1B trans-complementing 293 cells (31). Virus was purified from infected 293 cell culture supernatants over caesium chloride density gradients (32) and titered by fluorescence microscopy. Titers of different batches varied from 5 × 109–5 × 1011 PFU/ml.

FIGURE 1.

Maps of the recombinant plasmids pAd IL-4 and pAd p40 IL-12. Plasmids were linearized using the PacI cutting sites, resulting in loss of the gene encoding kanamycin resistance (kanr) and the gene encoding the origin of replication (ori). CMV, CMV immediate early promoter; LITR, left hand-inverted terminal repeat; IL-4, full-length cDNA-encoding ovine IL-4; p40 IL-12, full-length cDNA-encoding p40 subunit of ovine IL-12.

FIGURE 1.

Maps of the recombinant plasmids pAd IL-4 and pAd p40 IL-12. Plasmids were linearized using the PacI cutting sites, resulting in loss of the gene encoding kanamycin resistance (kanr) and the gene encoding the origin of replication (ori). CMV, CMV immediate early promoter; LITR, left hand-inverted terminal repeat; IL-4, full-length cDNA-encoding ovine IL-4; p40 IL-12, full-length cDNA-encoding p40 subunit of ovine IL-12.

Close modal

Endotoxin was measured with the Limulus amoebocyte lysate test (BioWhittaker) according to the manufacturer’s protocol. The lower level of detection of endotoxin was 1 pg (0.06 endotoxin units)/ml.

Corneas were excised from fresh sheep eyes obtained within 3 h of donor death from a local abattoir (Lobethal Abattoirs, Lobethal, Australia) and were organ-cultured in vitro as previously described (9). Corneas were infected with 0.8–1 × 107 PFU/cornea Ad-p40, Ad-IL-4, or Ad-GFP at room temperature for 2 h, followed by incubation in 3 ml of complete medium (HEPES-buffered RPMI 1640 medium (ICN Pharmaceuticals) supplemented with 10% v/v heat-inactivated (56°C, 30 min) FCS, 100 IU/ml penicillin, 100 μg/ml streptomycin, 2.5 μg/ml amphotericin B, and 2 mM l-glutamine (all from Invitrogen Life Technologies)). After 24 h, a further 2 ml of the same medium was added. For longer incubations thereafter, the volume of medium was increased to a total of 15 ml. Similarly processed but uninfected corneas served as controls.

Fresh corneas were infected with Ad-GFP, Ad-p40, or Ad-IL-4 as described above or were incubated in medium without any viral vector. Corneas were harvested at 2 h and at 3, 7, 10, 14, and 21 days after infection. A central 8-mm diameter full-thickness disc of cornea was trephined and snap-frozen in liquid nitrogen. Each disc was pulverized in a prechilled stainless steel mortar and pestle. Total RNA was extracted with Total RNA Extraction Reagent (Advanced Biotechnologies), treated with DNase (Invitrogen Life Technologies) and reverse-transcribed using a commercially available first-strand cDNA synthesis kit (Amersham Pharmacia Biotech U.K.) according to the manufacturer’s recommendations. To control for residual ovine genomic or viral DNA contamination, samples were subjected to the same reverse-transcription step after inactivation of the reverse transcriptase at 95°C for 60 min. Dilutions of cDNA were amplified in a 25-μl total volume by PCR. The concentration of reagents in the final reaction mixture was 10 mM Tris-HCl, pH 8.3, 0.15 M KCl (PerkinElmer Roche Molecular Systems), 0.2 mM each dNTP (Amersham Pharmacia Biotech), 1.5 mM MgCl2, 1 μM each primer, 1 U of AmpliTaq-Gold (all from PerkinElmer Roche Molecular Systems) and 5 μl of sample. Primer sequences amplified a 374-bp region for IL-4 (5′-TTAATGGGTCTCACCTCCC-3′, 5′-TTCCAAGAGGTCTCTCAGCG-3′), a 461-bp region for p40 IL-12 (5′-TTTGGAGATGCTGGGCAGTACA-3′, 5′-GATGATGTCCCTGATGAAGAAGC-3′) and a 317-bp region for the housekeeping gene, β-actin (5′-ATCATGTTTGAGACCTTCAA-3′, 5′-CATCTCTTGCTCGAAGTCCA-3′). After one cycle of 15 min at 94°C, 40 cycles of amplification were performed, each consisting of 30 s at 55°C (annealing), 30 s at 72°C (extension), and 1 min at 94°C (denaturation), with a final extension cycle of 30 s at 55°C, 20 s at 72°C, and 10 s at 35°C. Both H2O as the sample (negative control) and cDNA from ovine lymph node lymphocytes stimulated in vitro with Con A (Sigma-Aldrich) at 25 μg/ml for 48 h (positive control) were amplified with every experiment. Amplification products were analyzed by electrophoresis on 1.5% agarose gels.

Corneas were infected with Ad-IL-4 or Ad-GFP and organ-cultured as described above, but the final culture volume was restricted to 5 ml. Infected corneas were monitored at the fluorescence microscope for GFP expression to confirm successful infection. Culture supernatant was harvested after 3 days and stored at −20°C. Supernatants from uninfected corneas were used as negative controls.

The biological activity of IL-4 in supernatant of Ad-IL-4-infected corneas was assessed using a modified colorimetric assay (33). TF1.8 cells (CRL-2003; American Type Culture Collection) routinely maintained in the presence of GM-CSF (PeproTech) (34) in HEPES-buffered RPMI 1640 medium supplemented with 10% v/v heat-inactivated FCS, 100 IU/ml penicillin, 100 μg/ml streptomycin sulfate, and 2 mM l-glutamine were starved of GM-CSF overnight and seeded into 96-well flat-bottom plates (Falcon) at 2 × 104 cells/well. Supernatants from Ad-IL-4, Ad-GFP, or uninfected organ-cultured ovine corneas were added at 5% final concentration. TF1.8 cells incubated with 2 ng/ml GM-CSF or recombinant human IL-4 (PeproTech) ranging from 0.1 to 1 ng/ml were used as positive controls. Cells were incubated for 48 h at 37°C in 5% CO2 in air and the colorimetric reaction developed using the Cell Titer Proliferation kit (Promega) according to the manufacturer’s instructions. Absorbency at 490 nm was determined half hourly for 4 h using an ELISA plate reader (Multiscan EX; Thermo LabSystems).

Ovine corneas infected with 0.8–1.0 × 107 PFU Ad-GFP or Ad-p40 as described above were organ-cultured for 3 days. Uninfected corneas served as negative controls. Corneas were washed with medium and incubated in HEPES-buffered RPMI 1640 without glutamine, methionine, and cysteine (ICN Biochemicals) for 30 min at 32°C. This medium was replaced with 300 μl of medium supplemented with 50 μCi Trans35S-Label (ICN Radiochemicals). The corneas were incubated for a further 4 h at 32°C. Culture supernatants were harvested and immediately incubated with 45 μl of heat-inactivated normal rabbit serum (ICN Biochemicals) at 4°C for 1 h. Two 1 h cycles of Protein A adsorption were performed using 200 μl of Pansorbin (Calbiochem) in 500 mM Tris-HCl, pH 7.2, 25 mM NaCl, 0.1% w/v SDS, 1% w/v sodium deoxycholate, 1% v/v Triton X-100 buffer (radio-immunoprecipitation assay (RIPA)4 buffer) per cycle. The supernatants were collected, 15 μl of rabbit anti-human polyclonal anti-IL-12 Ab (Genzyme) was added, and the samples were incubated at 4°C overnight on a rotating shaker. Pansorbin (100 μl in RIPA buffer supplemented with 1 mg/ml OVA) was added for 1 h at 4°C. The samples were washed twice at 400 × g, 4°C for 15 min with RIPA buffer and resuspended in 30 μl of Laemmli-reducing electrophoresis buffer. SDS-PAGE was performed using the MiniProtean system (Bio-Rad) with a 3% stacking gel and 12.5% resolving gel. Coomassie blue-stained gels were incubated in the fluorescence signal-enhancer Amplify (Amersham Pharmacia Biotech) for 30 min at room temperature, dried under vacuum at 80°C, and exposed to an x-ray film (XAR5; Eastman Kodak) for 60 h at −80°C.

Approval for all experimentation was obtained from the institutional Animal Welfare Committee. Adult female Merino cross-breed sheep were acclimatized in groups of at least two animals for at least 1 wk in indoor pens and were fed water and chaff supplemented with lucerne hay ad libitum. Twelve-millimeter diameter penetrating corneal transplantation was performed as previously described (8) in the right eye only. The same experienced surgeon performed all of the graft procedures. Postoperative care and inspection were as previously described (8) and every graft was examined using the slit lamp each day. Groups of sheep received unmodified corneal grafts, or corneas optimally infected with Ad-GFP or Ad-p40 or Ad-IL-4 immediately before transplantation (9). Test and control grafts were performed in random order and the surgeon was unaware of the treatment that had been applied to the donor cornea. Corneal graft rejection was defined and documented as reported previously (8).

Photography was performed using a Nikon 801 camera equipped with a 105-mm Nikon Macro lens. A Kodak Wratten No. 47 blue excitation filter (Eastman Kodak) was placed over the flash, resulting in transmission of light at 400–500 nm with a peak at 440 nm. Additionally, a Hoya K2 yellow barrier filter mounted on the lens was used to eliminate wavelengths outside the GFP emission spectrum. In preliminary experiments f5.6 and f8.0 were determined as optimal apertures. Exposure time was 1/60 s. Photographs were taken in a darkened room with just enough light to allow focusing.

Eyes were harvested immediately postmortem and the cornea excised, fixed in buffered formalin, embedded in paraffin wax, cut at 8 μm, and stained with H&E. For immunohistochemistry, an indirect immunoperoxidase technique was used as previously described (8). Hybridoma culture supernatants containing mouse mAbs to sheep cell surface determinants were obtained from the Centre for Animal Biotechnology (University of Melbourne, Parkville, Australia) and included: SBU 41.19, anti-MHC class I monomorphic epitope; SBU 28.1, anti-MHC class II monomorphic epitope; SBU 1-11-32, anti-CD45 leukocyte-common (unrestricted) Ag; SBU 44.38, anti-CD4 and SBU 38.65, anti-CD8; and SBU 20.27, anti-CD1 (35). Culture supernatants from the hybridomas P3X63Ag8 (IgG1 isotype; European Collection of Animal Cell Cultures) and SAL5 (IgG2a isotype; gift of Dr. L. Ashman, Institute of Medical and Veterinary Science, Adelaide, South Australia, Australia) were used as negative controls. All sections were scored at the light microscope according to the following schema: −, no staining; ±, very rare positively stained cells; +, few positively stained cells; ++, moderate number of positively stained cells; +++, many positively stained cells.

Iris flat mounts were dissected, fixed in 2% paraformaldehyde at 4°C for 20 min, and stained as described by McMenamin (36). In brief, tissue pieces were immersed in the appropriate mAb at 4°C overnight, washed twice with Dulbecco’s A PBS and incubated for 1 h at 4°C with 1/750 diluted FITC-conjugated anti-mouse Ab (Silenus) in PBS supplemented with 1% normal swine serum (Silenus), 1% v/v FCS, and 1% v/v normal sheep serum. Tissue pieces were washed twice with PBS and mounted in PBS-glycerol. Sections were analyzed using a BX50 Olympus microscope with a Sony digital camera and MacImage 1.6.2 image analysis software and scored as described above.

For analysis of the bioassay for IL-4, Tukey’s HSD test was used in pairwise comparisons. Corneal graft survival data were analyzed with the Mann-Whitney U test, corrected for ties. Kaplan-Meier plots were constructed using SPSS version 11.0.2 software.

Ovine corneas were infected in vitro with Ad-IL-4, Ad-p40, or Ad-GFP and organ-cultured for up to 21 days. Uninfected corneas were similarly cultured as controls. Using RT-PCR on these corneas, the expected 374-bp product for IL-4 was amplified at 2 h after infection and at every subsequent time point tested from all corneas infected with Ad-IL-4 and from Con A-stimulated ovine lymph node cells; no product was amplified from normal uninfected corneas (Fig. 2,A and Table I). Similarly, the expected 461-bp product for p40 IL-12 was amplified from all corneas infected with Ad-p40 but not from corneas infected with Ad-GFP (Fig. 2,B and Table I). Controls in which the reverse transcriptase had been inactivated did not amplify at any stage. None of the uninfected corneas or the corneas infected with Ad-GFP showed detectable mRNA for either IL-4 or p40 IL-12 at any time. Because the gene transfer vectors carried cDNA for GFP, expression of this marker protein was able to be visualized. In organ-cultured corneas, expression of GFP was observed at 24 h after infection and was maximal at 3–4 days after infection, by which time 80–100% of corneal endothelial cells were expressing GFP.

FIGURE 2.

Expression of IL-4 (A) or p40 IL-12 (B) in corneas infected with Ad-IL-4 or Ad-p40. Corneas infected with Ad-IL-4 or Ad-p40 or Ad-GFP were organ-cultured for 7 days (Ad-IL-4) or 21 days (Ad-p40, Ad-GFP) before RNA extraction and RT-PCR. Dilutions (1/1, 1/10, 1/100) of PCR products from duplicate amplifications were loaded on to a 1.5% agarose gel. The expected sizes of the amplified specific products were 374 bp for IL-4 and 461 bp for p40 IL-12. Controls included duplicate PCRs with water (no cDNA), cDNA from uninfected cornea, or cDNA from Con A-stimulated ovine lymph node cells (lymph node). RT-PCR for the housekeeping gene β-actin (with loading of similar amounts of product as for test samples) was found to be positive for all samples except the no DNA control (data not shown). The data presented are illustrative and representative of additional data obtained at different time points.

FIGURE 2.

Expression of IL-4 (A) or p40 IL-12 (B) in corneas infected with Ad-IL-4 or Ad-p40. Corneas infected with Ad-IL-4 or Ad-p40 or Ad-GFP were organ-cultured for 7 days (Ad-IL-4) or 21 days (Ad-p40, Ad-GFP) before RNA extraction and RT-PCR. Dilutions (1/1, 1/10, 1/100) of PCR products from duplicate amplifications were loaded on to a 1.5% agarose gel. The expected sizes of the amplified specific products were 374 bp for IL-4 and 461 bp for p40 IL-12. Controls included duplicate PCRs with water (no cDNA), cDNA from uninfected cornea, or cDNA from Con A-stimulated ovine lymph node cells (lymph node). RT-PCR for the housekeeping gene β-actin (with loading of similar amounts of product as for test samples) was found to be positive for all samples except the no DNA control (data not shown). The data presented are illustrative and representative of additional data obtained at different time points.

Close modal
Table I.

Detection of mRNA for IL-4 and p40 IL-12 by RT-PCR at different time points after adenoviral vector-mediated gene transfer to organ-cultured ovine corneas in vitro

Cornea Infected with:Detection of mRNA for IL-4aDetection of mRNA for p 40 IL-12a
2 h3 db7 d10 d14 d21 d2 h3 d7 d10 d14 d21 d
Uninfected c NTd e NT NT NT − − − NT NT NT 
Ad-GFP − NT NT NT − − NT − NT NT NT − 
Ad-IL-4 NT NT − NT NT NT NT 
Ad-p40 IL-12 − NT NT − NT NT 
Cornea Infected with:Detection of mRNA for IL-4aDetection of mRNA for p 40 IL-12a
2 h3 db7 d10 d14 d21 d2 h3 d7 d10 d14 d21 d
Uninfected c NTd e NT NT NT − − − NT NT NT 
Ad-GFP − NT NT NT − − NT − NT NT NT − 
Ad-IL-4 NT NT − NT NT NT NT 
Ad-p40 IL-12 − NT NT − NT NT 
a

Three corneas were tested at each time point shown.

b

d, Days after infection.

c

+, Detection of the specific mRNA species in all three corneas tested.

d

NT, Not tested.

e

−, No detectable specific mRNA species in any of the three corneas tested.

Proliferation of TF1.8 cells was induced by supernatant of Ad-IL-4-infected corneas. Fig. 3 shows representative data from one of two experiments: at a concentration of 5%, the supernatant of Ad-IL-4-infected corneas induced a substantial level of proliferation, while the negative control supernatants induced significantly less proliferation (p < 0.01).

FIGURE 3.

Detection of biologically active IL-4 in supernatant of corneas infected with Ad-IL-4. IL-4-dependent TF1.8 cells were cultured in medium alone to determine baseline proliferation, with added 2 ng/ml GM-CSF or 0.1–1 ng/ml recombinant human IL-4 (rIL-4) as positive controls, or with corneal supernatants. Supernatants from uninfected corneas (cont), or corneas infected with Ad-GFP (GFP) or Ad-IL-4 (IL-4) collected after 3 days of corneal organ culture were added at a final concentration of 5%. TF1.8 cell proliferation was quantified spectrophotometrically at 490 nm. The mean ± SD of triplicate wells is shown. Data are representative of two experiments; ∗, p < 0.001 compared with controls. Statistical analysis was performed with Tukey’s HSD test.

FIGURE 3.

Detection of biologically active IL-4 in supernatant of corneas infected with Ad-IL-4. IL-4-dependent TF1.8 cells were cultured in medium alone to determine baseline proliferation, with added 2 ng/ml GM-CSF or 0.1–1 ng/ml recombinant human IL-4 (rIL-4) as positive controls, or with corneal supernatants. Supernatants from uninfected corneas (cont), or corneas infected with Ad-GFP (GFP) or Ad-IL-4 (IL-4) collected after 3 days of corneal organ culture were added at a final concentration of 5%. TF1.8 cell proliferation was quantified spectrophotometrically at 490 nm. The mean ± SD of triplicate wells is shown. Data are representative of two experiments; ∗, p < 0.001 compared with controls. Statistical analysis was performed with Tukey’s HSD test.

Close modal

Immunoprecipitation on supernatants from corneas infected with Ad-p40 and Ad-GFP showed that a protein of the expected size, 40 kDa, was precipitated from the supernatant of Ad-p40-infected corneas but not from supernatants of control corneas (Fig. 4).

FIGURE 4.

Detection of p40 IL-12 in supernatant of corneas infected with Ad-p40 by immunoprecipitation. Uninfected ovine corneas and corneas infected with Ad-GFP or with 6 × 106 or 6 × 107 PFU Ad-p40 under optimized conditions were labeled with 35S for 4 h. Culture supernatants were pooled within each group of three corneas and immunoprecipitation for p40 IL-12 was performed. Precipitates were run on a 10% SDS-PAGE, which was dried and exposed to x-ray film for 3 days. The expected size of the p40 IL-12 product was 40 kDa.

FIGURE 4.

Detection of p40 IL-12 in supernatant of corneas infected with Ad-p40 by immunoprecipitation. Uninfected ovine corneas and corneas infected with Ad-GFP or with 6 × 106 or 6 × 107 PFU Ad-p40 under optimized conditions were labeled with 35S for 4 h. Culture supernatants were pooled within each group of three corneas and immunoprecipitation for p40 IL-12 was performed. Precipitates were run on a 10% SDS-PAGE, which was dried and exposed to x-ray film for 3 days. The expected size of the p40 IL-12 product was 40 kDa.

Close modal

Sheep were randomly assigned to receive either untreated or gene-modified corneal grafts. In the sheep receiving a gene-modified cornea, expression of GFP was monitored in vitro in the remnant donor cornea (the corneoscleral rim) remaining after transplantation, in vivo in the graft, and postmortem. GFP was always observed in the donor rim after in vitro organ culture for 24 h, confirming successful viral infection of the donor cornea. Further, GFP expression was visualized in the living sheep in 18 of 21 adenovirus-infected grafts at times postoperatively ranging from 2 to 36 days. Expression of GFP by fixed donor corneal tissues harvested postmortem was observed to varying extents in 16 of these 21 grafts, in residual islands of endothelial cells still attached to the underlying Descemet’s membrane. GFP expression was more clearly evident in the corneoscleral rims examined 24 h after infection than in residual endothelial cells in the corneas retrieved postmortem, suggesting a loss of transgene expression over time. However, a few GFP-expressing corneal endothelial cells were still visible in one sheep killed at 121 days postoperatively. No expression of GFP was observed in normal, uninfected corneal grafts either in vitro or in vivo at any time. In all sheep where GFP was visualized, GFP expression was restricted to the donor cornea.

Corneal graft outcome was determined by daily examination of each graft at the slit lamp. Over the first few postoperative days, some infiltrating leukocytes were observed in the anterior chamber of all grafted eyes. Fibrin was also present in the majority of anterior chambers, but resolved gradually over a few days. By 10 days postoperatively, all grafts were clear and, except in the group that had received corneas treated with Ad-IL-4 (vide infra), the grafted eyes were noninflamed. Blood vessels crossed from the corneal periphery into the graft at medians ranging from 7 to 9 days among the experimental groups. In the absence of immunosuppression, untreated corneal grafts and grafts treated with the control virus Ad-GFP underwent rejection at a median of 21 and 20 days, respectively (Fig. 5). There was no statistically significant difference in graft survival between these two control groups (p > 0.05).

FIGURE 5.

Kaplan-Meier plot of survival of gene-modified ovine corneal allografts. Sheep were randomly assigned to receive untreated corneal allografts (n = 13), Ad-GFP-treated grafts (n = 6), Ad-IL-4-treated grafts (n = 6), or Ad-p40 IL-12-treated grafts (n = 9). Grafts were examined daily and rejection was defined as spreading edema in a previously clear, thin corneal graft or the appearance of an epithelial or endothelial rejection line.

FIGURE 5.

Kaplan-Meier plot of survival of gene-modified ovine corneal allografts. Sheep were randomly assigned to receive untreated corneal allografts (n = 13), Ad-GFP-treated grafts (n = 6), Ad-IL-4-treated grafts (n = 6), or Ad-p40 IL-12-treated grafts (n = 9). Grafts were examined daily and rejection was defined as spreading edema in a previously clear, thin corneal graft or the appearance of an epithelial or endothelial rejection line.

Close modal

Grafted eyes containing an Ad-IL-4-treated donor cornea developed severe inflammation within the first postoperative week. To exclude the possibility that this inflammatory response was caused by bacterial endotoxin contamination of the Ad-IL-4, the level of endotoxin in the various viral stocks was measured and found to be <25 pg endotoxin/ml, a level insufficient to account for the inflammation observed. Ten days after the graft procedure, two Ad-IL-4-treated corneal grafts developed ulcerations followed by antibiotic-resistant corneal infections caused by a different organism in each case, and were excluded from graft survival analysis. Technically successful corneal grafts modified with the vector-encoding ovine IL-4 were rejected at a median of 18.5 days (Fig. 5), not significantly different from the graft survival observed in untreated controls (p = 0.059) or the Ad-GFP-treated controls (p = 0.330).

Ad-p40-treated grafts exhibited a median time to rejection of 45 days (p = 0.002 and p = 0.003 compared with unmodified controls and Ad-GFP infected grafts, respectively; Fig. 5). All grafted corneas became neovascularized by invasion of limbal vessels, which occurred first in the superior and inferior quadrants of the cornea, but in the sheep with long-surviving corneal grafts that had been treated with Ad-p40, the new corneal vessels failed to maintain patency and became “ghost” vessels.

The right (engrafted) and left (control) corneas were processed for histological analysis and immunoperoxidase staining after euthanasia of each recipient. H&E-stained sections of each rejected graft showed a vascularized corneal stroma and graft edema with infiltrating mononuclear cells. Mononuclear cells were seen attached to residual corneal endothelium in some sections, but generally only very few or no endothelial cells remained in rejected corneal grafts indicating that, as expected, the corneal endothelium was the major target of the rejection process. There was no difference in the histological appearance of rejected corneal grafts from any experimental group, except that the Ad-IL-4-treated grafts were found to contain large numbers of eosinophils in addition to the infiltrate described above. Eosinophils were not seen in corneas from any of the other experimental groups, including untreated controls.

To examine the composition of infiltrating cells, immunoperoxidase staining for a variety of cell surface Ags was performed on 31 of 34 corneal grafts (Table II). No staining was observed with the negative control Abs. Rejected corneas from all treatment groups appeared similar for all of the markers examined. In conjunction with the results of the histological analysis, the immunohistochemistry indicated that, with the notable exception of the eosinophilia observed in all grafts that had been treated with Ad-IL-4, the gross composition of the infiltrate in rejecting grafts was not otherwise skewed by the local expression of immunomodulatory cytokines. The graft from one sheep bearing a long-surviving Ad-p40-treated corneal graft, which had remained clear for >121 days, was also processed for endpoint histologic analysis and immunoperoxidase staining. This cornea showed a histological appearance similar to that of a normal control cornea, with expression of MHC class II Ag being essentially confined to the limbus.

Table II.

Cellular infiltrate in rejected untreated and gene-modified corneal grafts as assessed by immunoperoxidase staining with murine anti-ovine mAbs

Ab SpecificityUntreated (n = 13)aAd-GFP Treated (n = 4)aAd-IL-4 Treated (n = 6)aAd-p40 IL-12 Treated (n = 8)a
CD45 ++/+++ ++/+++ ++/+++ ++/+++ 
CD1 −/(+) −/(+) (+)/+ −/(+) 
CD4 +/++ +/++ ++ 
CD8 (+)/+ (+)/+ 
MHC class I ++ ++ ++ ++ 
MHC class II ++ ++ ++ ++ 
Ab SpecificityUntreated (n = 13)aAd-GFP Treated (n = 4)aAd-IL-4 Treated (n = 6)aAd-p40 IL-12 Treated (n = 8)a
CD45 ++/+++ ++/+++ ++/+++ ++/+++ 
CD1 −/(+) −/(+) (+)/+ −/(+) 
CD4 +/++ +/++ ++ 
CD8 (+)/+ (+)/+ 
MHC class I ++ ++ ++ ++ 
MHC class II ++ ++ ++ ++ 
a

n, Number of corneas examined. Cellular infiltrate in rejected corneal allografts was assessed microscopically at postmortem as follows: −, no positive cells; +, very rare positive cells; +, few positive cells; ++, moderate number of positive cells; +++, many positive cells.

During corneal graft rejection, leukocytes are released from the iris and ciliary body into the anterior chamber. Iris flat mounts from eyes with rejected corneal grafts and from the normal partner eyes were stained for CD45 and MHC class II Ag (Table III). No staining with the negative control Ab was seen in any iris. In normal iris tissue, MHC II-positive cells were scattered throughout the iris. CD45-positive cells were also observed, especially within blood vessels. Rejected corneal grafts showed large numbers of CD45-positive cells within the iris tissue, and the number of MHC class II-positive cells also increased together with the intensity of staining, and their dendritic appearance appeared more pronounced in rejected corneal grafts. No obvious difference among iris tissues taken from rejected corneal grafts was seen for any of the treatment groups. When sections were examined at the fluorescence microscope, no expression of GFP was observed in any iris removed from an eye bearing a gene-modified cornea, suggesting that no free adenovirus had been introduced into the eye by the donor cornea. No difference was observed between the irides of the two eyes in one sheep with a long-surviving Ad-p40-treated corneal graft.

Table III.

Immunohistochemistry on iris flat mounts from normal sheep eyes, and from eyes with rejected normal and gene-modified corneal grafts

Donor Cornea PretreatmentNumber of EyesAb Specificity
CD45aMHC class IIa
Nil-normal eye 16 ++ 
Untreated +++ +++ 
Ad-GFP +++ +++ 
Ad-p40 IL-12 +++ +++ 
Ad-IL-4 +++ +++ 
Donor Cornea PretreatmentNumber of EyesAb Specificity
CD45aMHC class IIa
Nil-normal eye 16 ++ 
Untreated +++ +++ 
Ad-GFP +++ +++ 
Ad-p40 IL-12 +++ +++ 
Ad-IL-4 +++ +++ 
a

Iris tissues from normal eyes and eyes with rejected untreated or gene-modified grafts were stained with mAbs in conjunction with a secondary fluorescein-conjugated Ab and assessed at the light microscope. An isotype-matched mAb of irrelevant specificity was used as the negative control: no staining was observed on any iris with this mAb. The median reactivity of all iris tissue examined in each group is shown, where: +, some positive cells; ++, moderate number of positive cells; +++, many positive cells.

Recombinant replication-defective adenoviruses encoding full-length ovine IL-4 or the p40 subunit of IL-12 were constructed, characterized, and used to transfect ovine corneal endothelium ex vivo, before orthotopic corneal transplantation in outbred sheep. The sheep is a useful preclinical model for such experiments because corneal transplantation is similar to the corresponding surgery in humans, because ovine corneas can be successfully organ-cultured for up to 3 wk in corneal storage medium, because the ovine corneal endothelium is similar to its human counterpart in that it is nonreplicative, and because graft rejection in the sheep is similar to the corresponding process in humans (8, 9). Furthermore, adenoviral vectors infect ovine corneal endothelium efficiently (37), producing relatively long-term transgene expression in a proportion of animals without evidence of associated immunogenicity in vivo (9).

We were able to infect ovine corneas in vitro with adenoviral vectors containing genes for GFP, or GFP and p40 IL-12, or GFP and IL-4. Expression of GFP was confirmed by detection of specific mRNA, and by direct visualization of the reporter protein in infected tissue. Expression of p40 IL-12 was confirmed by detection of specific mRNA, and by immunoprecipitation of specific protein from supernatants of infected corneas. Expression of IL-4 was confirmed by detection of specific mRNA, and by a bioassay in which proliferation of the IL-4-responsive human myeloid cell line, TF1.8, was determined. Normal corneal endothelial cells are well known to produce a number of stimulatory and inhibitory factors (38), and supernatants from both uninfected and Ad-GFP-infected corneas were somewhat stimulatory for TF1 cells. However, corneal supernatants from Ad-IL-4-infected corneas were able to augment TF1.8 cell proliferation significantly, compared with these control supernatants. The possibility that IL-4 production by Ad-IL-4-infected corneas was inducing secretion of another growth factor–that is, that IL-4 may have been acting indirectly–was not formally excluded.

Local expression of p40 IL-12, but not IL-4, by ovine donor corneal endothelium prolonged orthotopic corneal allograft survival significantly in a proportion of sheep recipients. A number of grafts treated with the Ad-p40 vector showed survival for 3–4 mo in the absence of any additional immunosuppression and, specifically, in the absence of treatment with topical glucocorticosteroids (currently the gold-standard treatment in clinical corneal transplantation). Given that adenovirus is a nonintegrative vector and that the viral DNA remains episomal, this outcome in an outbred model is the more remarkable. That not every graft expressing p40 IL-12 showed prolongation of survival may have reflected occasional poor gene transfer. It was noteworthy that two of the three grafts in which the reporter gene GFP could not be visualized postmortem showed rejection at the same tempo as control grafts, suggestive of either inadequate gene transfer at the outset or accelerated loss of the transgene.

The competitive antagonist activity of p40 IL-12 for IL-12 (26, 27) has led to its use as an immunosuppressant in a number of experimental transplantation models. Murine cardiac myoblasts and pancreatic islets expressing p40 IL-12 after retrovirus- and adenovirus-mediated gene transfer, respectively, exhibited prolonged allograft survival (39, 40). However, in at least one mouse model, both anti-IL-12 Ab treatment and treatment with p40 IL-12 recombinant protein exacerbated cardiac allograft rejection (41). Recent evidence suggests that p40 IL-12 can recruit macrophages (42) and stimulate development of activated CD8 cells while inhibiting CD4 T cell function both in vivo and in vitro (43). However, the antagonist activity of p40 IL-12 appears to be dominant over its proinflammatory activities (26, 44, 45). Certainly, our p40 IL-12-modified corneal grafts did not become unusually inflamed and 89% of grafts to which cDNA for p40 IL-12 had been transferred survived for longer than the median survival exhibited by the controls. Once rejection was initiated, however, the process appeared identical with that occurring in control grafts, as assessed by corneal histology and by immunohistochemical analysis of cells infiltrating the cornea and the iris. In particular, local expression of p40 IL-12 by graft corneal endothelium did not skew the composition of the subsequent cellular infiltrate, at least at a gross level. The cellular infiltrate observed was consistent with that expected during a Th1 response, and rejection appeared to proceed in a Th1-type manner.

IL-12 may drive the immune response to corneal antigenic peptides presented within the draining lymph nodes (46), and mRNA for p40 IL-12 has been detected in rejecting corneal tissue (47). Thus, IL-12 appears a reasonable target for early immunomodulatory intervention to prevent corneal allograft rejection and our data suggest that further studies using an integrative viral vector that would permit sustained and prolonged expression of p40 IL-12 within the anterior segment of the eye may be of some value.

In contrast to our finding with p40 IL-12, and in agreement with earlier published work in the rat (48), gene transfer of cDNA encoding IL-4 to ovine corneal endothelium ex vivo, before transplantation, did not significantly prolong corneal allograft survival. IL-4 appears to play several distinct roles within the eye. Injection of Ag into the anterior chamber induces anterior chamber-associated immune deviation (ACAID) (2, 49), an aberrant response in which subsequent intraocular injection of Ag is associated with lack of IFN-γ production and failure to develop a delayed type hypersensitivity response. The involvement of IL-4 appears likely. However, while splenocytes from mice with ACAID produce significantly higher amounts of IL-4 than conventionally sensitized mice, ACAID can still be induced in IL-4 gene-targeted mice, indicating that IL-4 is not an absolute requirement for this particular Th2-type response (50). Surprisingly, a proinflammatory effect of IL-4 in the rat eye was also observed in experimental autoimmune uveitis, an organ-specific autoimmune disease (51). In this model, rIL-4 exacerbated rather than ameliorated uveitis and the effect was only partially reversed by treatment with anti-IL-4 Ab (51). Furthermore, IL-4 gene knockout mice experienced less severe disease than did normal animals in a mouse model of endotoxin-induced uveitis (52).

Recently, the involvement of IL-4 in the development of allergic eye disease has also been demonstrated. Exposure of corneal fibroblasts to IL-4 leads to expression of eotaxin, which is chemotactic for eosinophils (53). This finding is particularly interesting in the light of our observations that Ad-IL-4-treated corneal grafts became very inflamed and developed a marked eosinophilic infiltrate. Eosinophils might induce damage to the cornea by release of cytotoxic arginine-rich proteins such as eosinophil peroxidase, which has been implicated in the development of corneal ulcers (54). Niederkorn and colleagues (55) have recently shown increased numbers of eosinophils in rejected human corneal grafts from atopic patients with keratoconus. We hypothesize that eosinophil degranulation within a corneal graft will fuel inflammation and exacerbate a rejection response.

The mechanics of gene transfer to the cornea have been the subject of several excellent, recent reviews (56, 57, 58, 59). Although there are many instances of successful transfer of reporter genes, accounts of successful prolongation of corneal allograft survival following gene transfer are relatively few (60). Thus far, transgenic proteins shown to have some beneficial effect following expression by the cornea include mammalian IL-10 (9), an endostatin-kringle 5 fusion protein (61), a CTLA4-Ig fusion protein (62), and a combination of IL-4 and a soluble extracellular fragment of CTLA4 (63, 64). In the current study, we provide evidence that p40 IL-12 will also modulate corneal allograft rejection. Soluble TNFR has been shown to exert a marginally beneficial effect (65), whereas IL-4 alone has been shown to be ineffective (Ref. 48 , and vide infra).

The source of transgene expression in the gene-modified corneas is the corneal endothelium, and the highest level of the secreted therapeutic product is thus most likely to be in the aqueous humor, although some cytokine may also be expected to diffuse into the corneal stroma (66). Whether expression of the cytokine in the aqueous humor or in the draining lymph node might be expected to be the more important in modulating corneal allograft rejection is uncertain. Ag introduced into the anterior chamber of the eye drains into the venous circulation (2) and hence to the spleen, whereas the normal cornea itself is generally considered to lack specific lymphatic drainage. However, given that rejecting corneal allografts almost inevitably become vascularized and that vascularized corneas do develop lymphatic channels (67, 68), it seems probable that some sensitization will occur in regional lymph nodes. Certainly the murine cornea has been shown to contain a population of immature APCs capable of migration to, and maturation in, local lymph nodes (69, 70, 71). Of interest in this context is the finding from the George and Larkin and their colleagues (62) that although ex vivo transduction of rat corneas with a CTLA4-Ig fusion protein encoded by a replication-deficient adenoviral vector prolonged rat corneal graft survival to a small but significant extent, a more substantial prolongation of allograft survival was observed after systemic treatment with the same viral vector (62). In our own experiments reported elsewhere (9) and herein, we speculate that IL-10 and p40 IL-12 were most likely to have exerted their influence proximally in the afferent arm of the allograft response, and possibly within the intraocular environment itself, by affecting Ag-presenting cell maturation, migration, or function, or by decreasing local production of IFN-γ. Given that both transgenic proteins are likely to target the same afferent pathway, albeit in different ways, we would predict that a mixture of two vectors would not produce a synergistic response.

In summary, we have demonstrated that ex vivo transfer of an appropriate cytokine cDNA to a donor cornea, a simple, relatively safe and clinically relevant intervention, has the potential to modulate corneal graft survival significantly. Local production of the immunomodulatory cytokine p40 IL-12 (but not IL-4) within the eye is sufficient to prolong corneal graft survival for over 3 mo in a proportion of outbred recipients, in the complete absence of any other immunosuppression.

We thank Scott Standfield and Kirsty Marshall for expert technical assistance, Svjetlana Kireta for assistance with the immunoprecipitation assay, and Ray Yates for assistance with sheep husbandry.

S. Klebe, D. J. Coster, and K. A. Williams are employees of Flinders University, which has submitted a provisional patent application on the use of gene therapy for human corneal transplantation.

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 the National Health and Medical Research Council of Australia, the Ophthalmic Research Institute of Australia, and the Flinders Medical Centre Foundation.

4

Abbreviations used in this paper: RIPA, radio-immunoprecipitation assay; ACAID, anterior chamber-associated immune deviation.

1
Medawar, P..
1948
. Immunity of homologous grafted skin III. The fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye.
Brit. J. Exp. Pathol.
29
:
58
-69.
2
Streilein, J. W..
1995
. Unraveling immune privilege.
Science
270
:
1158
-1159.
3
Williams, K. A., D. Roder, A. Esterman, S. M. Muehlberg, D. J. Coster.
1992
. Factors predictive of corneal graft survival: report from the Australian Corneal Graft Registry.
Ophthalmology
99
:
403
-414.
4
Ayliffe, W., Y. Alam, E. B. Bell, D. McLeod, I. V. Hutchinson.
1992
. Prolongation of rat corneal graft survival by treatment with anti-CD4 monoclonal antibody.
Br. J. Ophthalmol.
7610
:
602
-606.
5
Yamada, J., I. Kurimoto, J. W. Streilein.
1999
. Role of CD4+ T cells in immunobiology of orthotopic corneal transplants in mice.
Invest. Ophthalmol. Vis. Sci.
40
:
2614
-2621.
6
Torres, P. F., T. P. Slegers, R. Peek, N. van Rooijen, R. van der Gaag, A. Kijlstra, A. F. de Vos.
1999
. Changes in cytokine mRNA levels in experimental corneal allografts after local clodronate-liposome treatment.
Invest. Ophthalmol. Vis. Sci.
40
:
3194
-3201.
7
Yamagami, S., H. Kawashima, H. Endo, T. Tsuru, H. Shibui, Y. Kagawa, J. Hori, H. Yamagami, M. Isobe.
1998
. Cytokine profiles of aqueous humor and graft in orthotopic mouse corneal transplantation.
Transplantation
66
:
1504
-1510.
8
Williams, K. A., S. D. Standfield, R. A. D. Mills, T. Takano, D. F. P. Larkin, R. Krishnan, G. R. Russ, D. J. Coster.
1999
. A new model of orthotopic penetrating corneal transplantation in the sheep: graft survival, phenotypes of graft-infiltrating cells and local cytokine production.
Aust. N. Z. J. Ophthalmol.
27
:
127
-135.
9
Klebe, S., P. J. Sykes, D. J. Coster, R. Krishnan, K. A. Williams.
2001
. Prolongation of sheep corneal allograft survival by transfer of the gene encoding ovine interleukin 10.
Transplantation
71
:
1214
-1220.
10
Mottram, P. L., W. R. Han, L. J. Purcell, I. F. C. McKenzie, W. W. Hancock.
1995
. Increased expression of IL-4 and IL-10 and decreased expression of IL-2 and interferon-γ in long-surviving mouse heart allografts after brief CD4-monoclonal antibody therapy.
Transplantation
59
:
559
-565.
11
Nickerson, P., A. Pacheco-Silva, P. O’Connell, W. Steurer, V. Rubin Kelley, T. Strom.
1993
. Analysis of cytokine transcripts in pancreatic islet cell allografts during rejection and tolerance induction.
Transplant. Proc.
25
:
984
-985.
12
Takeuchi, T., R. Lowry, B. Konieczny.
1992
. Heart allografts in murine systems: the differential activation of Th2-like effector cells in peripheral tolerance.
Transplantation
53
:
1281
-1294.
13
Piccotti, J. R., S. Y. Chan, A. M. Van Buskirk, E. J. Eichwald, D. K. Bishop.
1997
. Are Th2 helper T lymphocytes beneficial, deleterious, or irrelevant in promoting allograft survival?.
Transplantation
63
:
619
-624.
14
Lang, T., S. M. Krams, O. M. Martinez.
1996
. Production of IL-4 and IL-10 does not lead to immune quiescence in vascularized human organ grafts.
Transplantation
62
:
776
-780.
15
Bushell, A., M. Niimi, P. J. Morris, K. J. Wood.
1999
. Evidence for immune regulation in the induction of transplantation tolerance: a conditional but limited role for IL-4.
J. Immunol.
162
:
1359
-1366.
16
Sirak, J. H., C. G. Orosz, D. C. Roopenian, E. Wakely, A. M. Van Buskirk.
1998
. Cardiac allograft tolerance: failure to develop in interleukin-4-deficient mice correlates with unusual allosensitization patterns.
Transplantation
65
:
1352
-1356.
17
Takeuchi, T., T. Ueki, S. Sunaga, K. Ikuta, Y. Sasaki, B. X. Li, N. Moriyama, J. Miyazaki, K. Kawabe.
1997
. Murine interleukin 4 transgenic heart allograft survival prolonged with down-regulation of the Th1 cytokine mRNA in grafts.
Transplantation
64
:
152
-157.
18
He, X. Y., J. C. Chen, N. Verma, K. Plain, G. Tran, B. M. Hall.
1998
. Treatment with interleukin-4 prolongs allogeneic neonatal heart graft survival by inducing T helper 2 responses.
Transplantation
65
:
1145
-1152.
19
Nickerson, P., J. Steiger, X. X. Zheng, A. W. Steele, W. Steurer, P. Roy-Chaudhury, T. B. Strom.
1997
. Manipulation of cytokine networks in transplantation: false hope or realistic opportunity for tolerance?.
Transplantation
63
:
489
-494.
20
Steiger, J., P. W. Nickerson, W. Steurer, M. Moscovitch-Lopatin, T. B. Strom.
1995
. IL-2 knockout recipient mice reject islet cell allografts.
J. Immunol.
155
:
489
-498.
21
Gately, M. K., L. M. Renzetti, J. Magram, A. S. Stern, L. Adorini, U. Gubler, D. H. Presky.
1998
. The interleukin-12/interleukin-12-receptor system: role in normal and pathologic immune responses.
Annu. Rev. Immunol.
16
:
495
-521.
22
Kobayashi, M., L. Fitz, M. Ryan, R. M. Hewick, S. C. Clark, S. Chan, R. Loudon, F. Sherman, B. Perussia, G. Trinchieri.
1989
. Identification and purification of natural killer cell stimulatory factor NKSF, a cytokine with multiple biologic effects on human lymphocytes.
J. Exp. Med.
170
:
827
-845.
23
Ma, X., M. Aste-Amezaga, G. Trinchieri.
1996
. Regulation of interleukin-12 production.
Ann. NY Acad. Sci.
795
:
13
-25.
24
Wolf, S. F., P. A. Temple, M. Kobayashi, D. Young, M. Dicig, L. Lowe, R. Dzialo, L. Fitz, C. Ferenz, R. M. Hewick, et al
1991
. Cloning of cDNA for natural killer cell stimulatory factor, a heterodimeric cytokine with multiple biologic effects on T and natural killer cells.
J. Immunol.
146
:
3074
-3081.
25
Cassatella, M. A., L. Meda, S. Gasperini, A. D’Andrea, X. Ma, G. Trinchieri.
1995
. Interleukin-12 production by human polymorphonuclear leukocytes.
Eur. J. Immunol.
25
:
1
-5.
26
Gillessen, S., D. Carvajal, P. Ling, F. J. Podlaski, D. L. Stremlo, P. C. Familletti, U. Gubler, D. H. Presky, A. S. Stern, M. K. Gately.
1995
. Mouse interleukin-12 IL-12 p40 homodimer: a potent IL-12 antagonist.
Eur. J. Immunol.
25
:
200
-206.
27
Mattner, F., S. Fischer, S. Guckes, S. Jin, H. Kaulen, E. Schmitt, E. Rude, T. Germann.
1993
. The interleukin-12 subunit p40 specifically inhibits effects of the interleukin-12 heterodimer.
Eur. J. Immunol.
23
:
2202
-2208.
28
Swinburne, S. J., G. R. Russ, R. Krishnan.
2000
. Ovine interleukin 12 has biological activity on ovine and human activated peripheral blood mononuclear cells.
Cytokine
12
:
1546
-1552.
29
Chaplin, P. J., G. Casey, R. De Rose, G. Buchan, P. R. Wood, J.-P. Scheerlinck.
2000
. The expression and biologic effects of ovine interleukin-4 on T and B cell proliferation.
J. Interferon Cytokine Res.
20
:
419
-425.
30
He, T. C., S. B. Zhou, L. T. Dacosta, J. Yu, K. W. Kinzler, B. Vogelstein.
1998
. A simplified system for generating recombinant adenoviruses.
Proc. Natl. Acad. Sci. USA
95
:
2509
-2514.
31
Graham, F. L., L. Prevec.
1995
. Methods for construction of adenovirus vectors.
Mol. Biotechnol.
3
:
207
-220.
32
Kanegae, Y., M. Makimura, I. Saito.
1994
. A simple and efficient method for purification of infectious recombinant adenovirus.
Jpn. J. Med. Sci. Biol.
47
:
157
-166.
33
Mosmann, T..
1983
. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.
J. Immunol. Methods
65
:
55
-63.
34
Temple, R., E. Allen, J. Fordham, S. Phipps, H.-C. Schneider, K. Lindauer, I. Hayes, J. Lockey, K. Pollock, R. Jupp.
2001
. Microarray analysis of eosinophils reveals a number of candidate survival and apoptosis genes.
Am. J. Resp. Cell. Mol. Biol.
25
:
425
-433.
35
Mackay, C. R., J. F. Maddox, M. R. Brandon.
1987
. Lymphocyte antigens of sheep: identification and characterization using a panel of monoclonal antibodies.
Vet. Immunol. Immunopathol.
17
:
91
-102.
36
McMenamin, P. G..
2000
. Optimal methods for preparation and immunostaining of iris, ciliary body, and choroidal wholemounts.
Invest. Ophthalmol. Vis. Sci.
41
:
3043
-3048.
37
Klebe, S., P. J. Sykes, D. J. Coster, D. C. Bloom, K. A. Williams.
2001
. Gene transfer to ovine corneal endothelium.
Clin. Exp. Ophthalmol.
29
:
316
-322.
38
Joyce, N. C..
2003
. Proliferative capacity of the corneal endothelium.
Prog. Retin. Eye Res.
22
:
359
-389.
39
Kato, K., O. Shimozato, K. Hoshi, H. Wakimoto, H. Hamada, H. Yagita, K. Okumura.
1996
. Local production of the p40 subunit of interleukin 12 suppresses T-helper 1-mediated immune responses and prevents allogeneic myoblast rejection.
Proc. Natl. Acad. Sci. USA
93
:
9085
-9089.
40
Yasuda, H., M. Nagata, K. Arisawa, R. Yoshida, K. Fujihira, N. Okamoto, H. Moriyama, M. Miki, I. Saito, H. Hamada, et al
1998
. Local expression of immunoregulatory IL-12p40 gene prolonged syngeneic islet graft survival in diabetic NOD mice.
J. Clin. Invest.
102
:
1807
-1814.
41
Piccotti, J. R., S. Y. Chan, R. E. Goodman, J. Magram, E. J. Eichwald, D. K. Bishop.
1996
. IL-12 antagonism induces T helper 2 responses, yet exacerbates cardiac allograft rejection - evidence against a dominant protective role for T helper 2 cytokines in alloimmunity.
J. Immunol.
157
:
1951
-1957.
42
Ha, S. J., C. H. Lee, S. B. Lee, C. M. Kim, K. L. Jang, H. S. Shin, Y. C. Sung.
1999
. A novel function of IL-12p40 as a chemotactic molecule for macrophages.
J. Immunol.
163
:
2902
-2908.
43
Piccotti, J. R., S. Y. Chan, K. W. Li, E. J. Eichwald, D. K. Bishop.
1997
. Differential effects of IL-12 receptor blockade with IL-12 p40 homodimer on the induction of CD4+ and CD8+ IFN-γ-producing cells.
J. Immunol.
158
:
643
-648.
44
Heinzel, F., A. Hujer, F. Ahmed, R. Rerko.
1997
. In vitro production and function of IL-12 p40 homodimers.
J. Immunol.
158
:
4381
-4388.
45
Ling, P., M. K. Gately, U. Gubler, A. S. Stern, P. Lin, K. Hollfelder, C. Su, Y. C. Pan, J. Hakimi.
1995
. Human IL-12 p40 homodimer binds to the IL-12 receptor but does not mediate biologic activity.
J. Immunol.
154
:
116
-127.
46
Liu, Y., M. R. Dana, V. Tewari, A. W. Taylor.
2001
. Immune response to intragraft antigen in draining lymph nodes after corneal transplantation is mediated by interleukin-12.
J. Interferon Cytokine Res.
21
:
813
-819.
47
King, W. J., R. M. Comer, T. Hudde, D. F. Larkin, A. J. George.
2000
. Cytokine and chemokine expression kinetics after corneal transplantation.
Transplantation
70
:
1225
-1233.
48
Pleyer, U., E. Bertelmann, P. Rieck, C. Hartmann, H. D. Volk, T. Ritter.
2000
. Survival of corneal allografts following adenovirus-mediated gene transfer of interleukin-4.
Graefes Arch. Clin. Exp. Ophthalmol.
238
:
531
-536.
49
Li, X. Y., L. T. D’Orazio, J. Y. Niederkorn.
1996
. Role of Th1 and Th2 cells in anterior chamber-associated immune deviation.
Immunology
89
:
34
-40.
50
Kosiewicz, M. M., P. Alard, J. W. Streilein.
1998
. Alterations in cytokine production following intraocular injection of soluble protein antigen - impairment in IFN-γ and induction of TGF-β and IL-4 production.
J. Immunol.
161
:
5382
-5390.
51
Ramanathan, S., Y. A. S. de Kozak, O. Goureau, P. H. Van der Meide, P. Druet, B. Bellon.
1996
. Recombinant IL-4 aggravates experimental autoimmune uveoretinitis in rats.
J. Immunol.
157
:
2209
-2215.
52
Smith, J. R., P. H. Hart, D. J. Coster, K. A. Williams.
1998
. Mice deficient in tumor necrosis factor receptors p55 and p75, interleukin-4, or inducible nitric oxide synthase are susceptible to endotoxin-induced uveitis.
Invest. Ophthalmol. Vis. Sci.
39
:
658
-661.
53
Fukagawa, K., T. Nakajima, H. Saito, K. Tsubota, S. Shimmura, M. Natori, K. Hirai.
2000
. IL-4 induces eotaxin production in corneal keratocytes but not in epithelial cells.
Int. Arch. Allergy Immunol.
121
:
144
-150.
54
McGill, J. I., S. T. Holgate, M. K. Church, D. F. Anderson, A. Bacon.
1998
. Allergic eye disease mechanisms.
Br. J. Ophthalmol.
82
:
1203
-1214.
55
Hargrave, S., Y. Chu, D. Mendelblatt, E. Mayhew, J. Niederkorn.
2003
. Preliminary findings in corneal allograft rejection in patients with keratoconus.
Am. J. Ophthalmol.
135
:
452
-460.
56
Pleyer, U., H. Dannowski.
2002
. Delivery of genes via liposomes to corneal endothelial cells.
Drug News Perspect.
15
:
283
-289.
57
George, A. J. T., C. V. Arancibia-Carcamo, H. M. Awad, R. M. Comer, Z. Fehecari, W. J. King, M. Kadifachi, T. Hudde, C. Kerouedan-Lebosse, F. Mirza, et al
2000
. Gene delivery to the corneal endothelium.
Am. J. Respir. Crit. Care Med.
162
:
S194
-S200.
58
Jun, A. S., D. F. P. Larkin.
2003
. Prospects for gene therapy in corneal disease.
Eye
17
:
906
-911.
59
Borras, T..
2003
. Recent developments in ocular gene therapy.
Exp. Eye Res.
76
:
643
-652.
60
Williams, K. A., C. F. Jessup, D. J. Coster.
2004
. Gene therapy approaches to prolonging corneal allograft survival.
Expert Opin. Biol. Ther.
4
:
1059
-1071.
61
Murthy, R. C., T. J. McFarland, J. Yoken, S. Chen, C. Barone, D. Burke, Y. Zhang, B. Appukuttan, J. T. Stout.
2003
. Corneal transduction to inhibit angiogenesis and graft failure.
Invest. Ophthalmol. Vis. Sci.
44
:
1837
-1842.
62
Comer, R. M., W. J. King, N. Ardjomand, S. Theoharis, A. J. George, D. F. Larkin.
2002
. Effect of administration of CTLA4-Ig as protein or cDNA on corneal allograft survival.
Invest. Ophthalmol. Vis. Sci.
43
:
1095
-1103.
63
Konig Merediz, S. A., E. P. Zhang, B. Wittig, F. Hoffmann.
2000
. Ballistic transfer of minimalistic immunologically defined expression constructs for IL4 and CTLA4 into the corneal epithelium in mice after orthotopic corneal allograft transplantation.
Graefes Arch. Clin. Exp. Ophthalmol.
238
:
701
-707.
64
Zhang, E. P., J. Franke, M. Schroff, C. Junghans, B. Wittig, F. Hoffmann.
2003
. Ballistic CTLA4 and IL-4 gene transfer into the lower lid prolongs orthotopic corneal graft survival in mice.
Graefes Arch. Clin. Exp. Ophthalmol.
241
:
921
-926.
65
Rayner, S. A., D. F. Larkin, A. J. George.
2001
. TNF receptor secretion after ex vivo adenoviral gene transfer to cornea and effect on in vivo graft survival.
Invest. Ophthalmol. Vis. Sci.
42
:
1568
-1573.
66
Tonjum, A. M..
1974
. Permeability of horseradish peroxidase in the rabbit corneal epithelium.
Acta Ophthalmologica
52
:
650
-658.
67
Collin, H. B..
1966
. Endothelial cell lined lymphatics in the vascularized rabbit cornea.
Invest. Ophthalmol.
5
:
337
-354.
68
Cursiefen, C., U. Schlötzer-Schrehardt, M. Küchle, L. Sorokin, S. Breiteneder-Geleff, K. Alitalo, D. Jackson.
2002
. Lymphatic vessels in vascularized human corneas: immunohistochemical investigation using LYVE-1 and podoplanin.
Invest. Ophthalmol. Vis. Sci.
43
:
2127
-2135.
69
Liu, Y., P. Hamrah, Q. Zhang, A. W. Taylor, M. R. Dana.
2002
. Draining lymph nodes of corneal transplant hosts exhibit evidence for donor major histocompatibility complex (MHC) class II-positive cells derived from MHC class II-negative grafts.
J. Exp. Med.
195
:
259
-268.
70
Hamrah, P., Q. Zhang, Y. Liu, M. R. Dana.
2002
. Novel characterization of MHC class II-negative population of resident corneal Langerhans cell-type dendritic cells.
Invest. Ophthalmol. Vis. Sci.
43
:
639
-646.
71
Brissette-Storkus, C. S., S. M. Reynolds, A. J. Lepisto, R. L. Hendricks.
2002
. Identification of a novel macrophage population in the normal mouse corneal stroma.
Invest. Ophthalmol. Vis. Sci.
43
:
2264
-2271.