Cornea
Role of Human Corneal Endothelial Cells in T-Cell– Mediated Alloimmune Attack In Vitro Imad Lahdou,1 Christoph Engler,2 Stefan Mehrle,3 Volker Daniel,1 Mahmoud Sadeghi,1 Gerhard Opelz,1 and Peter Terness1 1
University Hospital Heidelberg, Department of Immunology, Heidelberg, Germany University Eye Hospital Mannheim, Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany 3University Hospital Heidelberg, Department of Molecular Virology, Heidelberg, Germany 2
Correspondence: Imad Lahdou, Institute of Immunology, Department of Transplantation Immunology, University of Heidelberg, Im Neuenheimer Feld 350, D-69120 Heidelberg, Germany; [email protected]. de. IL, CE, and SM contributed equally to the work presented here and should therefore be regarded as equivalent authors. Submitted: February 25, 2013 Accepted: November 24, 2013 Citation: Lahdou I, Engler C, Mehrle S, et al. Role of human corneal endothelial cells in T-cell–mediated alloimmune attack in vitro. Invest Ophthalmol Vis Sci. 2014;55:1213– 1221. DOI:10.1167/iovs.13-11930
PURPOSE. Human corneal endothelial cells (HCEC) are a potential target of immune attack after corneal transplantation. The aim of this in vitro study was to investigate the role of HCEC during the alloimmune response of T-cells by examining cytokine profiles, function of the immunosuppressive enzyme indoleamine 2,3-dioxigenase (IDO), major histocompatibility complex (MHC-I/-II), T-cell proliferation, and the induction of cell death. METHODS. Real-time PCR and RP-HPLC were used to determine IDO expression and activity. Multiplex assay was performed for quantification of cytokine levels. T-cell proliferation was assessed by thymidine incorporation, and HCEC cell death was measured by flow cytometry. RESULTS. Human corneal endothelial cells induce strong proliferation of allogeneic T-cells and an increase of proinflammatory cytokines such as interleukin-1a (IL-1a), IL-1b, IL-6, interferon-gamma (IFN-c), and tumor necrosis factor-alpha (TNF-a). Tumor necrosis factoralpha (and to a lesser extent IFN-c) induces apoptosis. Moreover, IFN-c strongly upregulates MHC-II molecules and IDO activity in HCEC as reflected by high kynurenine (Kyn) concentrations. Interestingly, the T-cell response was not affected by increased IDO activity, since blocking of IDO did not affect the proliferation rate. Indoleamine 2,3-dioxigenase– induced Kyn levels did not exceed concentrations of 175 6 20 lM. Concentrations of ‡400 lM Kyn were required to suppress T-cell proliferation. CONCLUSIONS. Our data show that T-cell attack on HCEC leads to increased concentrations of proinflammatory cytokines. Inflammatory cytokines induce apoptosis and upregulate MHC-II molecules and IDO in HCEC. Although increased IDO activity does not influence the T-cell response, it constitutes an inflammatory marker of the alloimmune response toward HCEC. Keywords: human corneal endothelial cells, IDO, alloimmune response, cell death
he cornea has been widely known to be an immunoprivileged site due to its tight regulation of immune cell entry. Keratoplasty surgery has been performed for over a century and constitutes a successful form of solid tissue transplantation.1,2 At a closer look at the clinical data, however, corneal transplant survival rates turn out to be not much better than those of other organs: The Australian Graft Registry Report 2012 (http:// hdl.handle.net/2328/25860 [in the public domain]) shows a 5year mean graft survival of 72% for penetrating keratoplasty and 67% for lamellar keratoplasty—figures comparable to 5-year kidney or liver allograft survival rates (69.3% and 68.4%, respectively, organs stemming from deceased donors) (2009 Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients: Transplant Data 1999–2008 [http://optn. transplant.hrsa.gov/ar2009/113_dh.pdf; in the public domain]). The United Kingdom’s organ donation and transplantation Activity Report 2011/12 shows a 5-year long-term graft survival rate in patients transplanted between 2004 and 2006 of 73% for cornea grafts (not distinguishing between the underlying causes), 85% for kidney grafts, and 77% for liver grafts (http:// www.organdonation.nhs.uk/statistics/transplant_activity_ report/current_activity_reports/ukt/activity_report_2011_12.
pdf [in the public domain]). In the Cornea Donor Study,3 data were more favorable: The 5-year cumulative probability of graft survival was 86%. Of all graft failures, only 4% were caused by graft rejection. Even though these data are encouraging, they clearly show that immune responses in keratoplasty surgery continue to constitute an obstacle in preventing long-term graft success in many patients. Due to the nonregenerative capacity of the human corneal endothelial layer, in the event of a decline of cell density below a critical value of 500 cells/mm2, corneal transplantation is often unavoidable.4,5 Acute rejection is the most common form of corneal allograft attack, which may target any of its three layers (epithelium, stroma, and endothelium). However, immune rejection within the corneal endothelium is usually the most severe form.5 For successful transplantation of corneal grafts, the endothelial cell count is critical. In addition to the natural attrition of corneal endothelial cells, the decline of the endothelial cell count is further exacerbated by mechanical shearing forces during the surgical procedure itself and early postoperative inflammation with cell loss by apoptosis.6 Apoptosis might play a decisive role already during ex vivo storage of corneal grafts.7
Copyright 2014 The Association for Research in Vision and Ophthalmology, Inc. www.iovs.org j ISSN: 1552-5783
1213
T
Role of HCEC in T-Cell–Mediated Alloimmune Attack Several investigations in rodents showed that CD4þ T-cells play a dominant role in acute rejection of corneal allografts.8–11 T helper (TH) 1 cells are a subset of CD4þ T-cells and defined by their production of IFN-c, which can stimulate macrophages and endothelial cells to produce proinflammatory factors.12,13 As shown in human keratoplasty patients and in rodent models of corneal transplantation, TH1 cells and IFN-c are involved in rejection.14,15 Along the same lines, increased levels of IFN-c and TNF-a were found in rejected corneas.15,16 In the last decade, the role of indoleamine 2,3-dioxigenase (IDO, IDO1, INDO) as an immunoregulatory factor has been a subject of interest, as it appears to play a role in infection, tumor evasion, autoimmune disease, and transplantation. Indoleamine 2,3-dioxigenase is expressed in many cell types and is the rate-limiting enzyme in the kynurenine (Kyn) pathway of tryptophan catabolism. Kynurenine and other catabolites were shown to inhibit the T-cell response. Therefore, the immunosuppressant role of IDO has been attributed, among others, to its role in the Kyn pathway. Indoleamine 2,3-dioxigenase expression is upregulated in response to proinflammatory stimuli such as the TH1 cytokine IFN-c.17 The immunoregulatory role of IDO in solid organ transplantation still remains controversial. Some experimental studies in animals18–20 found a causal relationship between IDO activity and prolongation of allograft survival, suggesting an immunoprotective role. A previous study showed as well that overexpression of IDO in mouse donor corneal allografts results in prolonged graft survival and that increased IDO activity may inhibit inflammatory cellular responses.20 However, clinical studies in organ transplant patients have not supported these experimental data.21–25 Brandacher et al.22 demonstrated that acute kidney graft rejection was associated with simultaneously increased serum and urinary IDO activity. We also showed that increased pretransplantation plasma Kyn levels predict acute kidney allograft rejection.24 In addition, our previous study showed that IDO expression is significantly higher in rejected corneal transplants as compared to successful grafts.23 Taking into consideration that rejection of corneal allograft is a problem in clinical transplantation, it is fundamental to understand the molecular processes of the immune response to corneal endothelial cells. This paves the way for the development of strategies to prolong corneal graft survival. The aim of this in vitro study was to investigate the response of human corneal endothelial cells (HCEC) to the alloimmune attack of T-cells with particular focus on the role of IDO. We addressed this question by assessing cytokine profiles, the expression and activity of IDO enzyme, T-cell proliferation, human leukocyte antigen (HLA) class I and class II expression, and apoptosis in HCEC under the influence of inflammatory stimuli and under cocultivation with allogeneic T-cells.
MATERIALS
AND
METHODS
All research methods described adhere to the tenets of the Declaration of Helsinki.
Cell Culture Human Corneal Endothelial Cells. Cell lines of HCEC were obtained from the Hamburg Eye Bank (J. Bednarz, University Hospital Hamburg-Eppendorf, Hamburg, Germany) as previously published.26 Cells were grown in RPMI 1640, with or without L-tryptophan, when we added D-tryptophan to RPMI 1640 (Promocell, Heidelberg, Germany) supplemented with 10% heat-inactivated fetal bovine serum (Promocell), 1% L-glutamine (Promocell), and 1% Penicillin/Streptomycin mix
IOVS j March 2014 j Vol. 55 j No. 3 j 1214 (Promocell). In some experiments, as described below, cells were stimulated with either interferon-c (IFN-c, recombinant [Escherichia coli], Catalog No. 11050494001; Roche Diagnostics, Mannheim, Germany) or tumor necrosis factor-a (TNF-a, human recombinant, No. T0157; Sigma-Aldrich, St. Louis, MO), alone or in combination. Prior to stimulation, HCEC were grown to a density of 1 3 106 cells per well. Cells were grown in six-well plates (CellStar; Greiner Bio-One, Frickenhausen, Germany) at 378C in a humidified atmosphere of 5% CO2 and stimulated for 3 days with the following reagents (concentrations): IFN-c (500 U/mL) or TNF-a (100 ng/mL). The human embryonic kidney 293 cell line (HEK 293 cells) was obtained from the Leibniz Institute DSMZ (German Collection of Microorganisms and Cell Cultures; DSMZ, No. ACC 305; Braunschweig, Germany). The DSMZ cell culture protocol was strictly followed. Human embryonic kidney 293 cells were cocultured with allogeneic lymphocytes in the same way as HCEC in order to measure cytokine profiles during the alloimmune response in both cell lines. Peripheral Blood Mononuclear Cells. Peripheral blood mononuclear cells (PBMC) were isolated from heparinized blood of healthy volunteers by density gradient centrifugation on Lymphodex (Inno-Train Diagnostik, Kronberg, Germany).27 Cells were washed three times in PBS, and viability was assessed by staining with trypan blue (Sigma-Aldrich GmbH, Steinheim, Germany). Cells were then cultured in a humidified atmosphere of 5% CO2 for 3 days.
Real-Time Polymerase Chain Reaction Total RNA was isolated using RNeasy-plus Mini Kit spin columns (Qiagen GmbH, Hilden, Germany). Quantification of total RNA was performed in a NanoDrop (ND-1000; NanoDrop Technologies, Wilmington, DE) spectrophotometer. One microgram of total RNA was used for cDNA synthesis by reverse transcription. First strand synthesis was performed with SuperScript III and Super Mix for quantitative reverse transcription polymerase chain reaction (qRT-PCR) module (Invitrogen, Darmstadt, Germany). The manufacturer’s protocol was strictly followed. The real-time PCR analysis was performed with the SYBR GreenER Two-Step qRT-PCR Kit Universal (Invitrogen) in an Applied Biosystems-7500 real-time PCR machine (Applied Bioscience, Darmstadt, Germany). Experiments were undertaken to monitor changes in the expression of IDO under immunological stimulation. Glycerinaldehyde-3-phosphate-dehydrogenase (GAPDH) was used as an endogenous control gene to normalize for varying starting amounts of RNA. The following oligonucleotide sequences were used: GAPDH, sense primer 5 0 -GAA GGT GAA GGT CGG AGT-3 0 , antisense primer 5 0 -AGA TGG TGA TGG GAT TTC-3 0 ; IDO, sense primer 5 0 -CAG CTG CTT CTG CAA TCA AA-3 0 , antisense primer 5 0 -AGC GCC TTT AGC AAA GTG TC-3 0 . The instrumental settings were as follows: initial denaturation step of 10 minutes at 958C, followed by 40 cycles of two-step PCR (denaturation 958C for 10 seconds, annealing and extension 608C for 60 seconds). At the end, a melting curve analysis was carried out on the formed products. Relative quantification was calculated as published before.28
Reversed-Phase High-Pressure Liquid Chromatography Culture supernatants were harvested to quantify the concentration of the IDO metabolites tryptophan and Kyn using RPHPLC as described previously.27 To remove residual proteins, 10% (vol/vol) 2.4 M perchloric acid was added to the sample, and the mixture was incubated for 5 minutes at room temperature. After centrifugation (5850g for 15 minutes at
IOVS j March 2014 j Vol. 55 j No. 3 j 1215
Role of HCEC in T-Cell–Mediated Alloimmune Attack 48C), supernatants were transferred into new vials. The pH value was adjusted to 7.0, and 100 lL filtered supernatant was injected into a C-18 column (Supelco, Aschaffenburg, Germany). Samples were eluted with PBS over 30 minutes. Experiments were undertaken to investigate the IDO-mediated tryptophan metabolism. Tryptophan was monitored by means of its native fluorescence at the excitation wavelength of 285 nm and emission wavelength of 360 nm, whereas Kyn was detected simultaneously by ultraviolet (UV) absorption at the wavelength of 230 nm. The peaks of Kyn and tryptophan were identified by comparison with the retention times of previously determined standard compounds. Quantification was based on the ratios of the peak areas of the compound to the internal standard.21
lasers (405, 488, and 633 nm). BD FACS DIVA 5.0 (BD Biosciences) software was used for analysis. Flow cytometry was used for the following purposes:
Lymphocyte Proliferation Assay Isolated PBMC (as described above) were either cocultured with HCEC or stimulated by OKT3 (Muromonab-CD3, monoclonal antibody targeting the CD3 receptor; Sigma-Aldrich, Taufkirchen, Germany). [3H]-Thymidine was used for the following experiments:
In order to find out whether IDO activity suppresses Tcell proliferation, HCEC were cocultured with allogeneic PBMC in medium with various concentrations of Trp (25, 250, and 500 lM). DL 1-methyl-tryptophan (DL 1-MT) was used to inhibit IDO activity. Human corneal endothelial cells were cocultured with allogeneic PBMC (1:1) in triplicates in 96-well plates in a total volume of 200 lL/well. Plates were incubated for 72 hours at 378C in a humidified 5% CO2 atmosphere. Cell proliferation was measured by an 18-hour pulse with [3H]-thymidine (5 lCi/mL; Amersham Pharmacia Biotech, N¨ umbrecht, Germany). Proliferation assays were measured in a betacounter (Inotech Biosystems, Lansing, MI), and counts per minute (cpm) were calculated as published before.28 In order to investigate the suppressive levels of Kyn, Tcells were stimulated with OKT3 in medium with various concentrations of Kyn (100, 200, 400, 800, and 1600 lM).
Carboxyfluorescein Diacetate Succinimidyl Ester Proliferation Assay Carboxyfluorescein diacetate succinimidyl ester (CFSE) dye was purchased from Life Technologies GmbH (Darmstadt, Germany). The manufacturer’s instructions were strictly followed. Proliferation was measured by flow cytometry. Reduction in fluorescence indicated cell division.
Multiplex Assay Interleukin-1a receptor antagonist (IL-1a RA), interleukin-1a (IL-1a), IL-1b, IL-2, interleukin-2 receptor (IL-2R), IL-4, IL-6, IL-8, IFN-c, and TNF-a were determined in cell culture supernatants using a multiplex assay (R&D Systems, Wiesbaden, Germany), which uses multianalyte profiles (MAPs) based on Luminex xMAP (Luminex Corporation, Austin, TX) technology. The manufacturer’s protocol was strictly followed. The levels of cytokines in supernatants were estimated by mean fluorescence intensity (MFI).
Flow Cytometry Analysis Flow cytometry was performed on a BD fluorescence-activated cell sorting (FACS, flow cytometry) CANTO II (BD Biosciences, Heidelberg, Germany) flow cytometer equipped with three
Assessment of changes in surface protein expression of HLA-I (A, B, C, labeled with phycoerythrin [PE], Catalog No. 555553) and -II (DR, labeled with fluorescein isothiocyanate [FITC], Catalog No. 555811; both BD Biosciences) on HCEC by stimulation with proinflammatory cytokines (IFN-c and TNF-a). After 72 hours of culturing, HCEC were collected, centrifuged twice at 1800g for 6 minutes, and washed with PBS/0.1% BSA. Human corneal endothelial cells were resuspended in 100 lL PBS/0.1% BSA, and IgG block reagent was added to avoid unspecific binding. After 5 minutes of incubation on ice, 4 lL stained monoclonal antibodies (antiHLA-ABC and -DR) was added. HCEC were incubated in the dark for 30 minutes, and cells were centrifuged at 3000g for 5 minutes and resuspended in 200 lL PBS/ 0.1% BSA. Investigation of HLA-DR expression in HCEC cocultured with allogeneic PBMC, HCEC, and PBMC: HLA typing of HCEC yielded HLA-A03, HLA-A31, HLA-B08, HLA-B27, HLA-C02, and HLA-C07. Peripheral blood mononuclear cells from blood of healthy volunteers were negative for HLA-A03 and positive for HLA-A02. The following antibodies were used for FACS analysis: anti-HLA-DR labeled with PE (Catalog No. 555812; BD Biosciences), HLA-A03 (Lot No. 009, Ref. BIH0269; One Lambda, Canoga Park, CA), and HLA-A02 labeled with FITC (Catalog No. 551285; BD Biosciences). Detection of cell death (7-AAD Annexin V) (BD Biosciences) in HCEC by stimulation with IFN-c/TNF-a and during incubation with various concentrations of Kyn (100, 200, 400, 800, and 1600 lM). The manufacturer’s protocol was strictly followed.
Statistical Analysis Data are based on at least three independent experiments. Mean values with standard deviation are given for all data. Statistical differences were calculated using unpaired Student’s t-test. A value of P 0.05 was regarded as statistically significant. All statistical analyses were performed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA).
RESULTS Coculturing HCEC With Allogeneic Lymphocytes Results in Increased Proinflammatory Cytokine Production To understand the cross-talk between lymphocytes and endothelial cells in corneal allograft rejection, we investigated different cytokines that are released during the alloimmune reaction. For this purpose, HCEC were cocultured with allogeneic PBMC for 72 hours. Peripheral blood mononuclear cells and HCEC alone served as controls. Cytokine levels were quantified in culture supernatants. Significant increases were found in the levels of the proinflammatory cytokines IL-1a, IL1b, IL-6, IFN-c, and TNF-a (P 0.01, for all investigated parameters, compared to PBMC alone, Fig. 1, upper row). On the other hand, levels of IL-1a RA, IL-2, IL-2R, IL-4, IL-5, IL-10, and IL-17 in the supernatants were similar to those of the controls (P ¼ NS, data not shown).
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1216
FIGURE 1. Cocultivation of HCEC with allogeneic lymphocytes (PBMC) results in increased proinflammatory cytokine production. Cytokine levels were determined in the supernatants in monoculture of HCEC, PBMC, or HEK 293, and in cocultures of HCEC with PBMC (1:1) (upper row) or cocultures of HEK 293 with PBMC (1:1) (lower row) by multiplex assay and expressed by mean fluorescence intensity (MFI). Left: IL-1a, IL-1b, and IL-6. Right: IFN-c and TNF-a. Bars represent means 6 standard deviation. Statistically significant differences in expression are shown for all investigated parameters (upper row).
To prove that the observed increase in cytokine levels in the supernatant of HCEC cocultured with allo-PBMC was not merely a generic response, in a set of experiments a different cell lineage (HEK 293 cells) was used. Results showed no significant increase of cytokine levels of supernatant of HEK 293 cells cocultured with allo-PBMC as compared to control supernatant (PBMC alone; Fig. 1, lower row).
Increasing IDO Function Induced by Coculturing of HCEC With Allogeneic Lymphocytes In a previous study we demonstrated upregulation of IDO expression and its enzymatic activity in HCEC under stimulation with IFN-c.23 Interferon-c is one of the main cytokines participating in allograft rejection.29–31 The increase of IFN-c levels in the supernatant of HCEC cocultured with allo-lymphocytes led us to raise the question whether IDO is upregulated during rejection of corneal transplants. For that purpose, HCEC were cocultured with allogeneic PBMC. Human corneal endothelial cells or PBMC alone served as negative controls, and HCEC or PBMC stimulated with IFNc served as positive controls. The results showed strong increase of IDO mRNA expression as well as IDO enzymatic activity by coculturing HCEC with allo-PBMC (Fig. 2). Indoleamine 2,3-dioxigenase degraded approximately 175 6 20 lM Trp, which resulted in 175 6 20 lM Kyn, accordingly. Of note, stimulation of cells with IFN-c upregulates IDO expression and activity in HCEC (1 3 106) approximately 100/ 10 times (respectively) more than in PBMC (1 3 106). The IDO blocker DL 1-MT inhibited this activity. This shows that HCEC were the main, if not the only, source of IDO production (Fig. 2, right).
HCEC Induce Proliferation of Allogeneic T-Cells in Spite of Highly Increased IDO Activity Human corneal endothelial cells are the most vulnerable cells during acute rejection. These cells, however, do not express major histocompatibility complex (MHC) class II molecules (HLA-II)32 and are therefore not expected to directly induce T-cell proliferation. Moreover, as shown by the above findings, IDO is upregulated upon coculture of HCEC with allogeneic PBMC. As a consequence, even if Tcell proliferation occurred, it would be inhibited by the decreased tryptophan and elevated Kyn levels induced by IDO. To our surprise, the findings showed vigorous T-cell proliferation (11.3-fold increase, P 0.001, compared to PBMC alone) in cocultures with allogeneic HCEC. This reaction was not significantly influenced by the IDO blocker DL 1-MT (P > 0.05), in spite of its ability to abrogate Kyn production in culture medium containing up to 175 6 20 lM tryptophan. These findings question the causative involvement of IDO in HCEC-induced T-cell proliferation in this setting (Fig. 3, left). Additional experiments were carried out to clarify if PBMC were the proliferating cells in coculture with HCEC as observed in [3H]-thymidine assay. Freshly isolated allogeneic PBMC were labeled with CFSE and cocultured with HCEC for 3 days. Peripheral blood mononuclear cells showed three cycles of proliferation in coculture, in contrast to the lack of proliferation of PBMC alone (Fig. 3, middle). In order to find out which levels of Kyn are suppressive in the test system, PBMC were stimulated with OKT3 in media containing increasing concentrations of Kyn (100, 200, 400, 800, and 1600 lM). Data showed that T-cell proliferation was suppressed by approximately 25% at concentrations of 400 lM, 61% at 800 lM, and 75% at 1600 lM Kyn, respectively (Fig.
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1217
FIGURE 2. Human corneal endothelial cells cocultured with allogeneic lymphocytes (PBMC) display increased IDO expression and activity. Indoleamine 2,3-dioxigenase expression and enzymatic activity were investigated by cocultivation of HCEC with allo-PBMC (1:1). Human corneal endothelial cells or PBMC alone served as negative controls. Stimulation of HCEC or PBMC with IFN-c served as positive control. Cells were cultured in 250 lM L-Trp medium for 72 hours. Left: IDO gene expression was detected by using RT-PCR. Right: IDO activity was estimated by measuring the degradation of L-Trp and accumulation of Kyn in supernatants using RP-HPLC. The IDO blocker DL-1-methyl-tryptophan (MT) was used to inhibit enzymatic IDO activity. Addition of IFN-c induced an increase in kynurenine production in both PBMC and HCEC (P 0.001). Addition of DL 1-MT (3 mM) to the coculture medium sharply reduced kynurenine production (P 0.001). Bars represent means 6 standard deviation; ns, nonsignificant.
3, right). Concentrations of 200 lM did not cause statistically significant suppression (P > 0.05). These results indicate that Kyn is suppressive only at high concentrations.
IFN-c Upregulates HLA-II Expression in HCEC To find out whether proinflammatory cytokines generated during HCEC–T-cell interaction affect the expression of HLA-I and/or -II in HCEC, the effect of IFN-c and TNF-a as classical proinflammatory cytokines was tested. Flow cytometry readings revealed equal expression of HLA-I antigens (HLA-A, -B, -C) on both treated and untreated HCEC (Fig. 4, upper row). However, HLA-II (HLA-DR) expression was upregulated upon stimulation with IFN-c (Fig. 4, lower row, P ¼ 0.002, as compared to untreated HCEC). Human corneal endothelial cells stimulated with TNF-a did not express HLA-II (Fig. 4, lower row, P ¼ NS as compared to untreated HCEC and P 0.001 as compared to stimulated HCEC with IFN-c).
The Expression of HLA-DR in an Allogeneic System We aimed to investigate the HLA-DR expression of HCEC cocultured with allogeneic PBMC; HCEC alone and PBMC alone served as controls. To this end, HLA-A02/A03þ HCEC and A02þ/A03 PBMC were coincubated and analyzed for DR expression by double staining (for HLA-A and -DR). The percentage of cells with expression of both HLA-A3þ and HLADR was 3% in HCEC alone and 9% in cocultures, whereas the percentage of cells with HLA-DR and HLA-A2þ was 22.1% in PBMC alone and 60.8% in cocultures (Fig. 5, right).
TNF-a or IFN-c Induces Apoptosis in HCEC We investigated whether the release of proinflammatory cytokines or the accumulation of Kyn affects the viability of HCEC. Cell death was determined by detection of 7-aminoactinomycin (7-AAD, a marker for the early phase of apoptosis) and Annexin V (a marker of cell death) upon
FIGURE 3. Increased IDO activity in HCEC does not suppress T-cell proliferation. T-cell proliferation was measured by incorporation of [3H]thymidine (cpm). Left: Human PBMC and HCEC were cultured alone or in combination in 250 lM L-Trp with or without addition of 3 mM DL-1methyl-tryptophan (MT). Cocultured cells showed a much higher cpm count as compared to PBMC alone (P 0.001). The addition of DL 1-MT did not significantly change cpm. Middle: Carboxyfluorescein diacetate succinimidyl ester dye was used as a tracer of T-cells in order to demonstrate that allogeneic T-cells proliferate. Right: PBMC were cultivated in the presence of Muromonab-CD3 (OKT3) in culture media containing 250 lM LTrp with increasing concentrations of Kyn (100–1600 lM). Concentrations ‡400 lM exhibited a statistically significant (P 0.001) effect on T-cell proliferation. Bars represent means 6 standard deviation; ns, not significant.
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1218
FIGURE 4. IFN-c upregulates expression of HLA class II molecules on HCEC: HLA class I and II surface expression on HCEC was determined by flow cytometry. Upper row: HLA-I (MHC-I) expression, from left to right: isotype, without stimulation, with IFN-c stimulation, with TNF-a stimulation. Lower row: HLA-II (MHC-II), from left to right: isotype, expression without stimulation, with IFN-c stimulation, with TNF-a stimulation.
treating HCEC with IFN-c or TNF-a (Fig. 5, first and second rows), or by adding various concentrations of Kyn (100, 200, 400, 800, and 1600 lM; Fig. 6, graph, third row). The results showed a statistically not significant increase of 7-AAD after incubation with IFN-c (P ¼ 0.13, compared to untreated HCEC, Fig. 5, first row) and a stronger, statistically significant increase after TNF-a treatment (P ¼ 0.02, compared to untreated HCEC, Fig. 5, first row). For Annexin V, similarly, an increase was observed after incubation with IFN-c (P ¼ 0.13, compared to untreated HCEC, Fig. 6, second row), whereas treatment with TNF-a caused stronger, statistically significant increase (P ¼ 0.03, compared to untreated HCEC, Fig. 5, second row). In experiments using Kyn, apoptosis occurred only at concentrations of >800 lM (P 0.01, compared to treatment without Kyn, Fig. 6, third row). These concentrations clearly exceed those expected in vivo.
IFN-c Upregulates the Expression of IDO mRNA in the Presence of L- or D-Tryptophan For a better understanding of the role of IDO in HCEC, we examined if its expression is affected in the presence of the isomers L- or D-tryptophan. Human corneal endothelial cells were cultured with either L- or D-tryptophan (25 lM), with or without IFN-c, for 72 hours. Findings showed that stimulation of HCEC with IFN-c leads to increased IDO mRNA expression in the presence of both isoforms (L- or D-) of Trp (see Supplementary Fig. S1).
IDO in HCEC Degrades L- or D-Tryptophan Equally To investigate whether IDO can equally degrade both isoforms (L- or D-) of tryptophan in the cell culture media, HCEC were cultured with various concentrations of L- or D-tryptophan (25, 250, and 500 lM), with or without IFN-c, for 72 hours
FIGURE 5. Expression of HLA-DR after coculture of HCEC with allogeneic PBMC cells had the following characteristics: HCEC were HLA-A03þ/HLAA02, whereas PBMC were HLA-A03/HLA-A02þ. Double staining (for HLA-A and DR) and subsequent FACS analysis showed that HCEC were predominantly HLA-A03þ/HLA-A02 (in over 98%) and PBMC HLA-A02þ/HLA-A03 (in over 90%). Percentage of cells with expression of both HLAA03þ and HLA-DR was 3% in HCEC alone, and 9% in HCEC cocultured with allo-PBMC. In contrast, percentage of cells with expression of HLA-DR and HLA-A02þ was 22.1% in PBMC (alone), and in coculture (HCEC and allo-PBMC) was 60.8% (right). Bars represent means 6 standard deviation.
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1219
FIGURE 6. The proinflammatory cytokines TNF-a and IFN-c induce apoptosis in HCEC. Cell death was assessed by detection of 7-actinoaminomycin (7-AAD, marker for the early phase of apoptosis for cell death, first row) and Annexin V (marker for cell death, second row) upon treatment of HCEC with IFN-c or TNF-a, or by adding various concentrations of Kyn (100, 200, 400, 800, and 1600 lM, third row). Untreated HCEC served as control (left in each row). First and second rows: Both IFN-c and TNF-a led to an increase of 7-AAD and Annexin V. This increase was statistically significant (P 0.05) only in the case of TNF-a. Third row: Concentrations of >800 lM Kyn led to a statistically significant (P 0.01) increase of Annexin V.
(Supplementary Fig. S2, left). Without IFN-c-stimulation, neither the degradation of L- or D-tryptophan nor, of course, accumulation of Kyn was detectable in the supernatant of HCEC. Stimulation with IFN-c triggered activation of IDO and resulted in degradation of both isoforms of tryptophan and accumulation of Kyn (Supplementary Fig. S2, left). Interestingly, degradation of both isoforms of tryptophan (L- and D-) and accumulation of Kyn were approximately similar. To our knowledge, this is the first report on IDO activity in HCEC proving degradation of both isoforms of tryptophan. Another set of experiments was carried out to analyze the blocking of IDO activity. Human corneal endothelial cells were incubated with 250 lM L-tryptophan, IFN-c, and DL-1MT in concentrations of 1, 2, and 3 mM. As shown in Supplementary Figure S2, right, at a concentration of 3 mM, DL 1-MT completely inhibited Kyn production. This extremely high concentration seems to be required when HCEC are in an immunologically active state.
DISCUSSION In this study we aimed to investigate the response of HCEC to the alloimmune attack of T-cells. To better understand the interaction and cross-talk between HCEC and allo-lymphocytes, pro- and anti-inflammatory cytokine levels were determined under cocultivation, showing elevation of multiple proinflammatory cytokines (IL-1a, IL-1b, IL-6, IFN-c, and TNF-a). Supernatants of HEK 293 cells/PBMC cocultures showed a similar or even lower cytokine release compared to the supernatant of PMBC. However, it was markedly weaker than the cytokine release in the supernatant of HCEC/PBMC cocultures, proving that the observed cytokine induction in HCEC was not a generic response, but rather a cell-specific one. These findings are in accordance with previous studies pointing to these cytokines as mediators of the alloimmune response in corneal transplantation.33–36 Bosnar et al.33 demonstrated that in corneal recipients with inflammatory diseases, enhanced production of IL-1a and TNF-a is positively correlated with graft rejection. Since IFN-c is a strong inducer of IDO, its increased levels in supernatants of HCEC/PBMC cocultures prompted us to investigate whether IDO is induced in these cultures. Our
results showed strong upregulation of IDO expression and activity. Interestingly, upon stimulation with IFN-c, upregulation of IDO is approximately 100 times higher in HCEC than in PBMC, indicating that HCEC are the main source of IDO production. In addition, our data reveal that IDO in HCEC is functionally expressed in the presence of both L- and Dtryptophan, and that its enzymatic activity is able to degrade both isoforms of tryptophan equally. Indoleamine 2,3-dioxigenase activity was completely inhibited by DL 1-MT at a concentration of 3 mM. Several previous studies described the potential immunosuppressive role of IDO. In order to examine if IDO modulates the response of activated allo-lymphocytes, a lymphocyte proliferation assay was performed. Experiments carried out in the presence or absence of the IDO inhibitor DL 1-MT did not show significant differences in ratios of T-cell proliferation induced by HCEC in settings with high IDO activity. Since Kyn is a potential mediator of IDO-induced immunosuppression, we tested which Kyn concentration is able to inhibit T-cell proliferation. The results showed that only high concentrations of Kyn (‡400 lM) inhibited the proliferation of activated T lymphocytes. Since the plasma concentration of tryptophan is approximately 60 lM in normal healthy humans,37,38 it is questionable whether the concentration of Kyn required for T-cell suppression can be reached in vivo. Failure to reach suppressive Kyn concentrations in vivo might explain the positive (and not negative) correlation of IDO expression with kidney graft rejection found in clinical studies.21,22,24 Another explanation would be that Kyn is not the main mediator of suppression in this system. Our previous study showed that IDO expression in corneal endothelial cells is significantly higher in rejected corneas as compared to accepted grafts.23 Based on our current in vitro results, we interpret this finding as a consequence of increased IFN-c expression during an inflammatory alloimmune response. Along the same lines, Reeve et al. 39 demonstrated an increase of IDO gene expression in kidney transplant biopsies with acute rejection. In the normal cornea, cells do not express MHC class II molecules, with exception of the limbal margins.32 In order to examine HLA (HLA-I or -II) expression of HCEC in the presence of proinflammatory cytokines (such as IFN-c and TNF-a, which are commonly generated during rejection), FACS studies were
Role of HCEC in T-Cell–Mediated Alloimmune Attack performed. We found that HCEC express HLA-I constitutively: Treated and untreated HCEC express equal levels of HLA-I. On the other hand, expression of HLA-II (HLA-DR) was detectable only after stimulation with IFN-c. Stimulation with TNF-a was not capable of inducing the expression of HLA-II. Using additional markers, a 3-fold upregulation of HLA-II in cocultured HCEC was detected. Peripheral blood mononuclear cells showed a similar behavior. Decay attack factor (DAF) is a contributor to the immune privilege of the cornea. We aimed to find out whether the presence or absence of DAF molecule (CD55) on HCEC can correlate with alloimmune response. Our data showed that 93% of HCEC expressed DAF molecules on their surface (data not shown). The MHC class II antigens expressed on the surface of antigen-presenting cells (APCs) play a pivotal role in the primary T-cell response.40 Expression of MHC class II antigens is inducible by immunological stimulation and has been detected on many cell types that do not express class II antigens under physiological conditions.41–43 In renal grafts, MHC class II antigens have been demonstrated to elicit strong immune responses,44,45 and compatibility at the HLA-DR (class II) locus was found to be of overriding importance for graft survival when compared to compatibility at the HLA-A and -B (class I) loci during the first posttransplant year.46 Hence, the induction of MHC class II antigens on HCEC further underlines the important role of these cells in immunological responses. We also investigated the induction of cell death in HCEC in the presence of IFN-c and TNF-a in the culture media. Both cytokines were able to affect the viability of HCEC. Tumor necrosis factor-alpha alone induced a very high rate of apoptosis in HCEC in vitro. Based on these results, we conclude that proinflammatory cytokines (especially TNF-a) produced during the alloimmune response might cause endothelial cell loss during corneal allograft rejection. Sagoo et al.6 showed that inflammatory cytokines (TNF-a, IFN-c, and IL-1a) induced high rates of apoptosis in mouse corneal endothelium. Moreover, Niederkorn et al.47 demonstrated high susceptibility of murine corneal epithelial and endothelial cells to TNF-a–induced apoptosis. To our knowledge, our work is the first report on proinflammatory cytokine induction of apoptosis in HCEC. As a consequence, avoidance of major postoperative inflammatory episodes should be an imperative paradigm in treatment. In summary, we believe that the results presented here improve current knowledge on the important immunological role of HCEC and IDO in cornea transplantation and provide the basis for further in vivo studies. Although IDO activity was increased during the immune attack of HCEC, it did not play an immunosuppressive role. These findings support a possible correlation of IDO expression/activity with graft rejection in clinical transplantation, but do not support a graft protective role of IDO. Consequently, increased IDO activity might be a predictive marker for rejection, but not for good graft function.
Acknowledgments Helmut Simon, Martina Kutsche-Bauer, and Regina Seemuth provided skillful technical assistance. Disclosure: I. Lahdou, P; C. Engler, None; S. Mehrle, P; V. Daniel, None; M. Sadeghi, None; G. Opelz, P; P. Terness, P
References 1. Niederkorn JY. Immunology and immunomodulation of corneal transplantation. Int Rev Immunol. 2002;21:173–196.
IOVS j March 2014 j Vol. 55 j No. 3 j 1220 2. Niederkorn JY, Larkin DF. Immune privilege of corneal allografts. Ocul Immunol Inflamm. 18:162–171. 3. Cornea Donor Study Investigator Group, Gal RL, Dontchev M, et al. The effect of donor age on corneal transplantation outcome results of the cornea donor study. Ophthalmology. 2008;115:620–626.e6. 4. George AJ, Larkin DF. Corneal transplantation: the forgotten graft. Am J Transplant. 2004;4:678–685. 5. Hori J, Joyce NC, Streilein JW. Immune privilege and immunogenicity reside among different layers of the mouse cornea. Invest Ophthalmol Vis Sci. 2000;41:3032–3042. 6. Sagoo P, Chan G, Larkin DF, George AJ. Inflammatory cytokines induce apoptosis of corneal endothelium through nitric oxide. Invest Ophthalmol Vis Sci. 2004;45:3964–3973. 7. Barcia RN, Dana MR, Kazlauskas A. Corneal graft rejection is accompanied by apoptosis of the endothelium and is prevented by gene therapy with bcl-xL. Am J Transplant. 2007;7:2082–2089. 8. Ayliffe W, Alam Y, Bell EB, McLeod D, Hutchinson IV. Prolongation of rat corneal graft survival by treatment with anti-CD4 monoclonal antibody. Br J Ophthalmol. 1992;76: 602–606. 9. He YG, Ross J, Niederkorn JY. Promotion of murine orthotopic corneal allograft survival by systemic administration of antiCD4 monoclonal antibody. Invest Ophthalmol Vis Sci. 1991; 32:2723–2728. 10. Hegde S, Beauregard C, Mayhew E, Niederkorn JY. CD4(þ) Tcell-mediated mechanisms of corneal allograft rejection: role of Fas-induced apoptosis. Transplantation. 2005;79:23–31. 11. Yamada J, Kurimoto I, Streilein JW. Role of CD4þ T cells in immunobiology of orthotopic corneal transplants in mice. Invest Ophthalmol Vis Sci. 1999;40:2614–2621. 12. Cua DJ, Coffman RL, Stohlman SA. Exposure to T helper 2 cytokines in vivo before encounter with antigen selects for T helper subsets via alterations in antigen-presenting cell function. J Immunol. 1996;157:2830–2836. 13. Pober JS, Gimbrone MA Jr, Lapierre LA, et al. Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J Immunol. 1986;137:1893–1896. 14. Pepose JS, Nestor MS, Gardner KM, Foos RY, Pettit TH. Composition of cellular infiltrates in rejected human corneal allografts. Graefes Arch Clin Exp Ophthalmol. 1985;222:128– 133. 15. Torres PF, De Vos AF, van der Gaag R, Martins B, Kijlstra A. Cytokine mRNA expression during experimental corneal allograft rejection. Exp Eye Res. 1996;63:453–461. 16. Zhu S, Dekaris I, Duncker G, Dana MR. Early expression of proinflammatory cytokines interleukin-1 and tumor necrosis factor-alpha after corneal transplantation. J Interferon Cytokine Res. 1999;19:661–669. 17. Takikawa O. Biochemical and medical aspects of the indoleamine 2,3-dioxygenase-initiated L-tryptophan metabolism. Biochem Biophys Res Commun. 2005;338:12–19. 18. Li J, Meinhardt A, Roehrich ME, et al. Indoleamine 2,3dioxygenase gene transfer prolongs cardiac allograft survival. Am J Physiol Heart Circ Physiol. 2007;293:H3415–H3423. 19. Yu G, Dai H, Chen J, et al. Gene delivery of indoleamine 2,3dioxygenase prolongs cardiac allograft survival by shaping the types of T-cell responses. J Gene Med. 2008;10:754–761. 20. Beutelspacher SC, Pillai R, Watson MP, et al. Function of indoleamine 2,3-dioxygenase in corneal allograft rejection and prolongation of allograft survival by over-expression. Eur J Immunol. 2006;36:690–700. 21. Holmes EW, Russell PM, Kinzler GJ, et al. Oxidative tryptophan metabolism in renal allograft recipients: increased kynurenine
Role of HCEC in T-Cell–Mediated Alloimmune Attack
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32. 33.
synthesis is associated with inflammation and OKT3 therapy. Cytokine. 1992;4:205–213. Brandacher G, Cakar F, Winkler C, et al. Non-invasive monitoring of kidney allograft rejection through IDO metabolism evaluation. Kidney Int. 2007;71:60–67. Serbecic N, Lahdou I, Scheuerle A, Hoftberger R, Aboul-Enein F. Function of the tryptophan metabolite, L-kynurenine, in human corneal endothelial cells. Mol Vis. 2009;15:1312–1324. Lahdou I, Sadeghi M, Daniel V, et al. Increased pretransplantation plasma kynurenine levels do not protect from but predict acute kidney allograft rejection. Hum Immunol. 71:1067–1072. Suarez-Fuentetaja N, Domenech-Garcia N, Paniagua-Martin MJ, et al. Indoleamine, 2-3 dioxygenase activity could be an early marker of graft rejection in heart transplantation. Transplant Proc. 44:2645–2648. Bednarz J, Teifel M, Friedl P, Engelmann K. Immortalization of human corneal endothelial cells using electroporation protocol optimized for human corneal endothelial and human retinal pigment epithelial cells. Acta Ophthalmol Scand. 2000; 78:130–136. Terness P, Chuang JJ, Bauer T, Jiga L, Opelz G. Regulation of human auto- and alloreactive T cells by indoleamine 2,3dioxygenase (IDO)-producing dendritic cells: too much ado about IDO? Blood. 2005;105:2480–2486. Dennis-Sykes CA, Miller WJ, McAleer WJ. A quantitative Western Blot method for protein measurement. J Biol Stand. 1985;13:309–314. Hu H, Kwun J, Aizenstein BD, Knechtle SJ. Noninvasive detection of acute and chronic injuries in human renal transplant by elevation of multiple cytokines/chemokines in urine. Transplantation. 2009;87:1814–1820. Karczewski M, Karczewski J, Poniedzialek B, Wiktorowicz K, Glyda M. Cytometric analysis of TH1/TH2 cytokines in the urine of patients undergoing kidney transplantation. Ann Transplant. 2009;14:25–28. Xu J, Wei J, Zhu X, et al. Increased plasma indoleamine 2,3dioxygenase activity and interferon-c levels correlate with the severity of acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2013;19:196–201. Streilein JW. New thoughts on the immunology of corneal transplantation. Eye (Lond). 2003;17:943–948. Bosnar D, Dekaris I, Gabric N, Markotic A, Lazic R, Spoljaric N. Influence of interleukin-1alpha and tumor necrosis factoralpha production on corneal graft survival. Croat Med J. 2006; 47:59–66.
IOVS j March 2014 j Vol. 55 j No. 3 j 1221 34. Jin Y, Chai LJ, Hong J. Early expression of tumor necrosis factor alpha and nuclear factor kappa B after Descemet stripping endothelial keratoplasty in rabbits. Cornea. 31:913–919. 35. Maier P, Heizmann U, Bohringer D, Kern Y, Reinhard T. Predicting the risk for corneal graft rejection by aqueous humor analysis. Mol Vis. 2011;17:1016–1023. 36. van Gelderen BE, Van Der Lelij A, Peek R, et al. Cytokines in aqueous humour and serum before and after corneal transplantation and during rejection. Ophthalmic Res. 2000; 32:157–164. 37. Kudo Y, Boyd CA, Sargent IL, Redman CW. Decreased tryptophan catabolism by placental indoleamine 2,3-dioxygenase in preeclampsia. Am J Obstet Gynecol. 2003;188:719– 726. 38. Silk JD, Lakhal S, Laynes R, et al. IDO induces expression of a novel tryptophan transporter in mouse and human tumor cells. J Immunol. 2011;187:1617–1625. 39. Reeve J, Einecke G, Mengel M, et al. Diagnosing rejection in renal transplants: a comparison of molecular- and histopathology-based approaches. Am J Transplant. 2009;9:1802–1810. 40. Rosenthal AS, Shevach EM. Function of macrophages in antigen recognition by guinea pig T lymphocytes. I. Requirement for histocompatible macrophages and lymphocytes. J Exp Med. 1973;138:1194–1212. 41. Hall BM, Bishop GA, Duggin GG, Horvath JS, Philips J, Tiller DJ. Increased expression of HLA-DR antigens on renal tubular cells in renal transplants: relevance to the rejection response. Lancet. 1984;2:247–251. 42. Volc-Platzer B, Majdic O, Knapp W, et al. Evidence of HLA-DR antigen biosynthesis by human keratinocytes in disease. J Exp Med. 1984;159:1784–1789. 43. Iwata M, Yagihashi A, Roat MI, Zeevi A, Iwaki Y, Thoft RA. Human leukocyte antigen-class II-positive human corneal epithelial cells activate allogeneic T cells. Invest Ophthalmol Vis Sci. 1994;35:3991–4000. 44. Moen T, Albrechtsen D, Flatmark A, et al. Importance of HLADR matching in cadaveric renal transplantation: a prospective one-center study of 170 transplants. N Engl J Med. 1980;303: 850–854. 45. Young E, Stark WJ, Prendergast RA. Immunology of corneal allograft rejection: HLA-DR antigens on human corneal cells. Invest Ophthalmol Vis Sci. 1985;26:571–574. 46. Opelz G, Wujciak T, Dohler B, Scherer S, Mytilineos J. HLA compatibility and organ transplant survival. Collaborative Transplant Study. Rev Immunogenet. 1999;1:334–342. 47. Niederkorn JY, Mayhew E, Mellon J, Hegde S. Role of tumor necrosis factor receptor expression in anterior chamberassociated immune deviation (ACAID) and corneal allograft survival. Invest Ophthalmol Vis Sci. 2004;45:2674–2681.
Role of Human Corneal Endothelial Cells in T-Cell– Mediated Alloimmune Attack In Vitro Imad Lahdou,1 Christoph Engler,2 Stefan Mehrle,3 Volker Daniel,1 Mahmoud Sadeghi,1 Gerhard Opelz,1 and Peter Terness1 1
University Hospital Heidelberg, Department of Immunology, Heidelberg, Germany University Eye Hospital Mannheim, Medical Faculty Mannheim of the University of Heidelberg, Mannheim, Germany 3University Hospital Heidelberg, Department of Molecular Virology, Heidelberg, Germany 2
Correspondence: Imad Lahdou, Institute of Immunology, Department of Transplantation Immunology, University of Heidelberg, Im Neuenheimer Feld 350, D-69120 Heidelberg, Germany; [email protected]. de. IL, CE, and SM contributed equally to the work presented here and should therefore be regarded as equivalent authors. Submitted: February 25, 2013 Accepted: November 24, 2013 Citation: Lahdou I, Engler C, Mehrle S, et al. Role of human corneal endothelial cells in T-cell–mediated alloimmune attack in vitro. Invest Ophthalmol Vis Sci. 2014;55:1213– 1221. DOI:10.1167/iovs.13-11930
PURPOSE. Human corneal endothelial cells (HCEC) are a potential target of immune attack after corneal transplantation. The aim of this in vitro study was to investigate the role of HCEC during the alloimmune response of T-cells by examining cytokine profiles, function of the immunosuppressive enzyme indoleamine 2,3-dioxigenase (IDO), major histocompatibility complex (MHC-I/-II), T-cell proliferation, and the induction of cell death. METHODS. Real-time PCR and RP-HPLC were used to determine IDO expression and activity. Multiplex assay was performed for quantification of cytokine levels. T-cell proliferation was assessed by thymidine incorporation, and HCEC cell death was measured by flow cytometry. RESULTS. Human corneal endothelial cells induce strong proliferation of allogeneic T-cells and an increase of proinflammatory cytokines such as interleukin-1a (IL-1a), IL-1b, IL-6, interferon-gamma (IFN-c), and tumor necrosis factor-alpha (TNF-a). Tumor necrosis factoralpha (and to a lesser extent IFN-c) induces apoptosis. Moreover, IFN-c strongly upregulates MHC-II molecules and IDO activity in HCEC as reflected by high kynurenine (Kyn) concentrations. Interestingly, the T-cell response was not affected by increased IDO activity, since blocking of IDO did not affect the proliferation rate. Indoleamine 2,3-dioxigenase– induced Kyn levels did not exceed concentrations of 175 6 20 lM. Concentrations of ‡400 lM Kyn were required to suppress T-cell proliferation. CONCLUSIONS. Our data show that T-cell attack on HCEC leads to increased concentrations of proinflammatory cytokines. Inflammatory cytokines induce apoptosis and upregulate MHC-II molecules and IDO in HCEC. Although increased IDO activity does not influence the T-cell response, it constitutes an inflammatory marker of the alloimmune response toward HCEC. Keywords: human corneal endothelial cells, IDO, alloimmune response, cell death
he cornea has been widely known to be an immunoprivileged site due to its tight regulation of immune cell entry. Keratoplasty surgery has been performed for over a century and constitutes a successful form of solid tissue transplantation.1,2 At a closer look at the clinical data, however, corneal transplant survival rates turn out to be not much better than those of other organs: The Australian Graft Registry Report 2012 (http:// hdl.handle.net/2328/25860 [in the public domain]) shows a 5year mean graft survival of 72% for penetrating keratoplasty and 67% for lamellar keratoplasty—figures comparable to 5-year kidney or liver allograft survival rates (69.3% and 68.4%, respectively, organs stemming from deceased donors) (2009 Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients: Transplant Data 1999–2008 [http://optn. transplant.hrsa.gov/ar2009/113_dh.pdf; in the public domain]). The United Kingdom’s organ donation and transplantation Activity Report 2011/12 shows a 5-year long-term graft survival rate in patients transplanted between 2004 and 2006 of 73% for cornea grafts (not distinguishing between the underlying causes), 85% for kidney grafts, and 77% for liver grafts (http:// www.organdonation.nhs.uk/statistics/transplant_activity_ report/current_activity_reports/ukt/activity_report_2011_12.
pdf [in the public domain]). In the Cornea Donor Study,3 data were more favorable: The 5-year cumulative probability of graft survival was 86%. Of all graft failures, only 4% were caused by graft rejection. Even though these data are encouraging, they clearly show that immune responses in keratoplasty surgery continue to constitute an obstacle in preventing long-term graft success in many patients. Due to the nonregenerative capacity of the human corneal endothelial layer, in the event of a decline of cell density below a critical value of 500 cells/mm2, corneal transplantation is often unavoidable.4,5 Acute rejection is the most common form of corneal allograft attack, which may target any of its three layers (epithelium, stroma, and endothelium). However, immune rejection within the corneal endothelium is usually the most severe form.5 For successful transplantation of corneal grafts, the endothelial cell count is critical. In addition to the natural attrition of corneal endothelial cells, the decline of the endothelial cell count is further exacerbated by mechanical shearing forces during the surgical procedure itself and early postoperative inflammation with cell loss by apoptosis.6 Apoptosis might play a decisive role already during ex vivo storage of corneal grafts.7
Copyright 2014 The Association for Research in Vision and Ophthalmology, Inc. www.iovs.org j ISSN: 1552-5783
1213
T
Role of HCEC in T-Cell–Mediated Alloimmune Attack Several investigations in rodents showed that CD4þ T-cells play a dominant role in acute rejection of corneal allografts.8–11 T helper (TH) 1 cells are a subset of CD4þ T-cells and defined by their production of IFN-c, which can stimulate macrophages and endothelial cells to produce proinflammatory factors.12,13 As shown in human keratoplasty patients and in rodent models of corneal transplantation, TH1 cells and IFN-c are involved in rejection.14,15 Along the same lines, increased levels of IFN-c and TNF-a were found in rejected corneas.15,16 In the last decade, the role of indoleamine 2,3-dioxigenase (IDO, IDO1, INDO) as an immunoregulatory factor has been a subject of interest, as it appears to play a role in infection, tumor evasion, autoimmune disease, and transplantation. Indoleamine 2,3-dioxigenase is expressed in many cell types and is the rate-limiting enzyme in the kynurenine (Kyn) pathway of tryptophan catabolism. Kynurenine and other catabolites were shown to inhibit the T-cell response. Therefore, the immunosuppressant role of IDO has been attributed, among others, to its role in the Kyn pathway. Indoleamine 2,3-dioxigenase expression is upregulated in response to proinflammatory stimuli such as the TH1 cytokine IFN-c.17 The immunoregulatory role of IDO in solid organ transplantation still remains controversial. Some experimental studies in animals18–20 found a causal relationship between IDO activity and prolongation of allograft survival, suggesting an immunoprotective role. A previous study showed as well that overexpression of IDO in mouse donor corneal allografts results in prolonged graft survival and that increased IDO activity may inhibit inflammatory cellular responses.20 However, clinical studies in organ transplant patients have not supported these experimental data.21–25 Brandacher et al.22 demonstrated that acute kidney graft rejection was associated with simultaneously increased serum and urinary IDO activity. We also showed that increased pretransplantation plasma Kyn levels predict acute kidney allograft rejection.24 In addition, our previous study showed that IDO expression is significantly higher in rejected corneal transplants as compared to successful grafts.23 Taking into consideration that rejection of corneal allograft is a problem in clinical transplantation, it is fundamental to understand the molecular processes of the immune response to corneal endothelial cells. This paves the way for the development of strategies to prolong corneal graft survival. The aim of this in vitro study was to investigate the response of human corneal endothelial cells (HCEC) to the alloimmune attack of T-cells with particular focus on the role of IDO. We addressed this question by assessing cytokine profiles, the expression and activity of IDO enzyme, T-cell proliferation, human leukocyte antigen (HLA) class I and class II expression, and apoptosis in HCEC under the influence of inflammatory stimuli and under cocultivation with allogeneic T-cells.
MATERIALS
AND
METHODS
All research methods described adhere to the tenets of the Declaration of Helsinki.
Cell Culture Human Corneal Endothelial Cells. Cell lines of HCEC were obtained from the Hamburg Eye Bank (J. Bednarz, University Hospital Hamburg-Eppendorf, Hamburg, Germany) as previously published.26 Cells were grown in RPMI 1640, with or without L-tryptophan, when we added D-tryptophan to RPMI 1640 (Promocell, Heidelberg, Germany) supplemented with 10% heat-inactivated fetal bovine serum (Promocell), 1% L-glutamine (Promocell), and 1% Penicillin/Streptomycin mix
IOVS j March 2014 j Vol. 55 j No. 3 j 1214 (Promocell). In some experiments, as described below, cells were stimulated with either interferon-c (IFN-c, recombinant [Escherichia coli], Catalog No. 11050494001; Roche Diagnostics, Mannheim, Germany) or tumor necrosis factor-a (TNF-a, human recombinant, No. T0157; Sigma-Aldrich, St. Louis, MO), alone or in combination. Prior to stimulation, HCEC were grown to a density of 1 3 106 cells per well. Cells were grown in six-well plates (CellStar; Greiner Bio-One, Frickenhausen, Germany) at 378C in a humidified atmosphere of 5% CO2 and stimulated for 3 days with the following reagents (concentrations): IFN-c (500 U/mL) or TNF-a (100 ng/mL). The human embryonic kidney 293 cell line (HEK 293 cells) was obtained from the Leibniz Institute DSMZ (German Collection of Microorganisms and Cell Cultures; DSMZ, No. ACC 305; Braunschweig, Germany). The DSMZ cell culture protocol was strictly followed. Human embryonic kidney 293 cells were cocultured with allogeneic lymphocytes in the same way as HCEC in order to measure cytokine profiles during the alloimmune response in both cell lines. Peripheral Blood Mononuclear Cells. Peripheral blood mononuclear cells (PBMC) were isolated from heparinized blood of healthy volunteers by density gradient centrifugation on Lymphodex (Inno-Train Diagnostik, Kronberg, Germany).27 Cells were washed three times in PBS, and viability was assessed by staining with trypan blue (Sigma-Aldrich GmbH, Steinheim, Germany). Cells were then cultured in a humidified atmosphere of 5% CO2 for 3 days.
Real-Time Polymerase Chain Reaction Total RNA was isolated using RNeasy-plus Mini Kit spin columns (Qiagen GmbH, Hilden, Germany). Quantification of total RNA was performed in a NanoDrop (ND-1000; NanoDrop Technologies, Wilmington, DE) spectrophotometer. One microgram of total RNA was used for cDNA synthesis by reverse transcription. First strand synthesis was performed with SuperScript III and Super Mix for quantitative reverse transcription polymerase chain reaction (qRT-PCR) module (Invitrogen, Darmstadt, Germany). The manufacturer’s protocol was strictly followed. The real-time PCR analysis was performed with the SYBR GreenER Two-Step qRT-PCR Kit Universal (Invitrogen) in an Applied Biosystems-7500 real-time PCR machine (Applied Bioscience, Darmstadt, Germany). Experiments were undertaken to monitor changes in the expression of IDO under immunological stimulation. Glycerinaldehyde-3-phosphate-dehydrogenase (GAPDH) was used as an endogenous control gene to normalize for varying starting amounts of RNA. The following oligonucleotide sequences were used: GAPDH, sense primer 5 0 -GAA GGT GAA GGT CGG AGT-3 0 , antisense primer 5 0 -AGA TGG TGA TGG GAT TTC-3 0 ; IDO, sense primer 5 0 -CAG CTG CTT CTG CAA TCA AA-3 0 , antisense primer 5 0 -AGC GCC TTT AGC AAA GTG TC-3 0 . The instrumental settings were as follows: initial denaturation step of 10 minutes at 958C, followed by 40 cycles of two-step PCR (denaturation 958C for 10 seconds, annealing and extension 608C for 60 seconds). At the end, a melting curve analysis was carried out on the formed products. Relative quantification was calculated as published before.28
Reversed-Phase High-Pressure Liquid Chromatography Culture supernatants were harvested to quantify the concentration of the IDO metabolites tryptophan and Kyn using RPHPLC as described previously.27 To remove residual proteins, 10% (vol/vol) 2.4 M perchloric acid was added to the sample, and the mixture was incubated for 5 minutes at room temperature. After centrifugation (5850g for 15 minutes at
IOVS j March 2014 j Vol. 55 j No. 3 j 1215
Role of HCEC in T-Cell–Mediated Alloimmune Attack 48C), supernatants were transferred into new vials. The pH value was adjusted to 7.0, and 100 lL filtered supernatant was injected into a C-18 column (Supelco, Aschaffenburg, Germany). Samples were eluted with PBS over 30 minutes. Experiments were undertaken to investigate the IDO-mediated tryptophan metabolism. Tryptophan was monitored by means of its native fluorescence at the excitation wavelength of 285 nm and emission wavelength of 360 nm, whereas Kyn was detected simultaneously by ultraviolet (UV) absorption at the wavelength of 230 nm. The peaks of Kyn and tryptophan were identified by comparison with the retention times of previously determined standard compounds. Quantification was based on the ratios of the peak areas of the compound to the internal standard.21
lasers (405, 488, and 633 nm). BD FACS DIVA 5.0 (BD Biosciences) software was used for analysis. Flow cytometry was used for the following purposes:
Lymphocyte Proliferation Assay Isolated PBMC (as described above) were either cocultured with HCEC or stimulated by OKT3 (Muromonab-CD3, monoclonal antibody targeting the CD3 receptor; Sigma-Aldrich, Taufkirchen, Germany). [3H]-Thymidine was used for the following experiments:
In order to find out whether IDO activity suppresses Tcell proliferation, HCEC were cocultured with allogeneic PBMC in medium with various concentrations of Trp (25, 250, and 500 lM). DL 1-methyl-tryptophan (DL 1-MT) was used to inhibit IDO activity. Human corneal endothelial cells were cocultured with allogeneic PBMC (1:1) in triplicates in 96-well plates in a total volume of 200 lL/well. Plates were incubated for 72 hours at 378C in a humidified 5% CO2 atmosphere. Cell proliferation was measured by an 18-hour pulse with [3H]-thymidine (5 lCi/mL; Amersham Pharmacia Biotech, N¨ umbrecht, Germany). Proliferation assays were measured in a betacounter (Inotech Biosystems, Lansing, MI), and counts per minute (cpm) were calculated as published before.28 In order to investigate the suppressive levels of Kyn, Tcells were stimulated with OKT3 in medium with various concentrations of Kyn (100, 200, 400, 800, and 1600 lM).
Carboxyfluorescein Diacetate Succinimidyl Ester Proliferation Assay Carboxyfluorescein diacetate succinimidyl ester (CFSE) dye was purchased from Life Technologies GmbH (Darmstadt, Germany). The manufacturer’s instructions were strictly followed. Proliferation was measured by flow cytometry. Reduction in fluorescence indicated cell division.
Multiplex Assay Interleukin-1a receptor antagonist (IL-1a RA), interleukin-1a (IL-1a), IL-1b, IL-2, interleukin-2 receptor (IL-2R), IL-4, IL-6, IL-8, IFN-c, and TNF-a were determined in cell culture supernatants using a multiplex assay (R&D Systems, Wiesbaden, Germany), which uses multianalyte profiles (MAPs) based on Luminex xMAP (Luminex Corporation, Austin, TX) technology. The manufacturer’s protocol was strictly followed. The levels of cytokines in supernatants were estimated by mean fluorescence intensity (MFI).
Flow Cytometry Analysis Flow cytometry was performed on a BD fluorescence-activated cell sorting (FACS, flow cytometry) CANTO II (BD Biosciences, Heidelberg, Germany) flow cytometer equipped with three
Assessment of changes in surface protein expression of HLA-I (A, B, C, labeled with phycoerythrin [PE], Catalog No. 555553) and -II (DR, labeled with fluorescein isothiocyanate [FITC], Catalog No. 555811; both BD Biosciences) on HCEC by stimulation with proinflammatory cytokines (IFN-c and TNF-a). After 72 hours of culturing, HCEC were collected, centrifuged twice at 1800g for 6 minutes, and washed with PBS/0.1% BSA. Human corneal endothelial cells were resuspended in 100 lL PBS/0.1% BSA, and IgG block reagent was added to avoid unspecific binding. After 5 minutes of incubation on ice, 4 lL stained monoclonal antibodies (antiHLA-ABC and -DR) was added. HCEC were incubated in the dark for 30 minutes, and cells were centrifuged at 3000g for 5 minutes and resuspended in 200 lL PBS/ 0.1% BSA. Investigation of HLA-DR expression in HCEC cocultured with allogeneic PBMC, HCEC, and PBMC: HLA typing of HCEC yielded HLA-A03, HLA-A31, HLA-B08, HLA-B27, HLA-C02, and HLA-C07. Peripheral blood mononuclear cells from blood of healthy volunteers were negative for HLA-A03 and positive for HLA-A02. The following antibodies were used for FACS analysis: anti-HLA-DR labeled with PE (Catalog No. 555812; BD Biosciences), HLA-A03 (Lot No. 009, Ref. BIH0269; One Lambda, Canoga Park, CA), and HLA-A02 labeled with FITC (Catalog No. 551285; BD Biosciences). Detection of cell death (7-AAD Annexin V) (BD Biosciences) in HCEC by stimulation with IFN-c/TNF-a and during incubation with various concentrations of Kyn (100, 200, 400, 800, and 1600 lM). The manufacturer’s protocol was strictly followed.
Statistical Analysis Data are based on at least three independent experiments. Mean values with standard deviation are given for all data. Statistical differences were calculated using unpaired Student’s t-test. A value of P 0.05 was regarded as statistically significant. All statistical analyses were performed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA).
RESULTS Coculturing HCEC With Allogeneic Lymphocytes Results in Increased Proinflammatory Cytokine Production To understand the cross-talk between lymphocytes and endothelial cells in corneal allograft rejection, we investigated different cytokines that are released during the alloimmune reaction. For this purpose, HCEC were cocultured with allogeneic PBMC for 72 hours. Peripheral blood mononuclear cells and HCEC alone served as controls. Cytokine levels were quantified in culture supernatants. Significant increases were found in the levels of the proinflammatory cytokines IL-1a, IL1b, IL-6, IFN-c, and TNF-a (P 0.01, for all investigated parameters, compared to PBMC alone, Fig. 1, upper row). On the other hand, levels of IL-1a RA, IL-2, IL-2R, IL-4, IL-5, IL-10, and IL-17 in the supernatants were similar to those of the controls (P ¼ NS, data not shown).
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1216
FIGURE 1. Cocultivation of HCEC with allogeneic lymphocytes (PBMC) results in increased proinflammatory cytokine production. Cytokine levels were determined in the supernatants in monoculture of HCEC, PBMC, or HEK 293, and in cocultures of HCEC with PBMC (1:1) (upper row) or cocultures of HEK 293 with PBMC (1:1) (lower row) by multiplex assay and expressed by mean fluorescence intensity (MFI). Left: IL-1a, IL-1b, and IL-6. Right: IFN-c and TNF-a. Bars represent means 6 standard deviation. Statistically significant differences in expression are shown for all investigated parameters (upper row).
To prove that the observed increase in cytokine levels in the supernatant of HCEC cocultured with allo-PBMC was not merely a generic response, in a set of experiments a different cell lineage (HEK 293 cells) was used. Results showed no significant increase of cytokine levels of supernatant of HEK 293 cells cocultured with allo-PBMC as compared to control supernatant (PBMC alone; Fig. 1, lower row).
Increasing IDO Function Induced by Coculturing of HCEC With Allogeneic Lymphocytes In a previous study we demonstrated upregulation of IDO expression and its enzymatic activity in HCEC under stimulation with IFN-c.23 Interferon-c is one of the main cytokines participating in allograft rejection.29–31 The increase of IFN-c levels in the supernatant of HCEC cocultured with allo-lymphocytes led us to raise the question whether IDO is upregulated during rejection of corneal transplants. For that purpose, HCEC were cocultured with allogeneic PBMC. Human corneal endothelial cells or PBMC alone served as negative controls, and HCEC or PBMC stimulated with IFNc served as positive controls. The results showed strong increase of IDO mRNA expression as well as IDO enzymatic activity by coculturing HCEC with allo-PBMC (Fig. 2). Indoleamine 2,3-dioxigenase degraded approximately 175 6 20 lM Trp, which resulted in 175 6 20 lM Kyn, accordingly. Of note, stimulation of cells with IFN-c upregulates IDO expression and activity in HCEC (1 3 106) approximately 100/ 10 times (respectively) more than in PBMC (1 3 106). The IDO blocker DL 1-MT inhibited this activity. This shows that HCEC were the main, if not the only, source of IDO production (Fig. 2, right).
HCEC Induce Proliferation of Allogeneic T-Cells in Spite of Highly Increased IDO Activity Human corneal endothelial cells are the most vulnerable cells during acute rejection. These cells, however, do not express major histocompatibility complex (MHC) class II molecules (HLA-II)32 and are therefore not expected to directly induce T-cell proliferation. Moreover, as shown by the above findings, IDO is upregulated upon coculture of HCEC with allogeneic PBMC. As a consequence, even if Tcell proliferation occurred, it would be inhibited by the decreased tryptophan and elevated Kyn levels induced by IDO. To our surprise, the findings showed vigorous T-cell proliferation (11.3-fold increase, P 0.001, compared to PBMC alone) in cocultures with allogeneic HCEC. This reaction was not significantly influenced by the IDO blocker DL 1-MT (P > 0.05), in spite of its ability to abrogate Kyn production in culture medium containing up to 175 6 20 lM tryptophan. These findings question the causative involvement of IDO in HCEC-induced T-cell proliferation in this setting (Fig. 3, left). Additional experiments were carried out to clarify if PBMC were the proliferating cells in coculture with HCEC as observed in [3H]-thymidine assay. Freshly isolated allogeneic PBMC were labeled with CFSE and cocultured with HCEC for 3 days. Peripheral blood mononuclear cells showed three cycles of proliferation in coculture, in contrast to the lack of proliferation of PBMC alone (Fig. 3, middle). In order to find out which levels of Kyn are suppressive in the test system, PBMC were stimulated with OKT3 in media containing increasing concentrations of Kyn (100, 200, 400, 800, and 1600 lM). Data showed that T-cell proliferation was suppressed by approximately 25% at concentrations of 400 lM, 61% at 800 lM, and 75% at 1600 lM Kyn, respectively (Fig.
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1217
FIGURE 2. Human corneal endothelial cells cocultured with allogeneic lymphocytes (PBMC) display increased IDO expression and activity. Indoleamine 2,3-dioxigenase expression and enzymatic activity were investigated by cocultivation of HCEC with allo-PBMC (1:1). Human corneal endothelial cells or PBMC alone served as negative controls. Stimulation of HCEC or PBMC with IFN-c served as positive control. Cells were cultured in 250 lM L-Trp medium for 72 hours. Left: IDO gene expression was detected by using RT-PCR. Right: IDO activity was estimated by measuring the degradation of L-Trp and accumulation of Kyn in supernatants using RP-HPLC. The IDO blocker DL-1-methyl-tryptophan (MT) was used to inhibit enzymatic IDO activity. Addition of IFN-c induced an increase in kynurenine production in both PBMC and HCEC (P 0.001). Addition of DL 1-MT (3 mM) to the coculture medium sharply reduced kynurenine production (P 0.001). Bars represent means 6 standard deviation; ns, nonsignificant.
3, right). Concentrations of 200 lM did not cause statistically significant suppression (P > 0.05). These results indicate that Kyn is suppressive only at high concentrations.
IFN-c Upregulates HLA-II Expression in HCEC To find out whether proinflammatory cytokines generated during HCEC–T-cell interaction affect the expression of HLA-I and/or -II in HCEC, the effect of IFN-c and TNF-a as classical proinflammatory cytokines was tested. Flow cytometry readings revealed equal expression of HLA-I antigens (HLA-A, -B, -C) on both treated and untreated HCEC (Fig. 4, upper row). However, HLA-II (HLA-DR) expression was upregulated upon stimulation with IFN-c (Fig. 4, lower row, P ¼ 0.002, as compared to untreated HCEC). Human corneal endothelial cells stimulated with TNF-a did not express HLA-II (Fig. 4, lower row, P ¼ NS as compared to untreated HCEC and P 0.001 as compared to stimulated HCEC with IFN-c).
The Expression of HLA-DR in an Allogeneic System We aimed to investigate the HLA-DR expression of HCEC cocultured with allogeneic PBMC; HCEC alone and PBMC alone served as controls. To this end, HLA-A02/A03þ HCEC and A02þ/A03 PBMC were coincubated and analyzed for DR expression by double staining (for HLA-A and -DR). The percentage of cells with expression of both HLA-A3þ and HLADR was 3% in HCEC alone and 9% in cocultures, whereas the percentage of cells with HLA-DR and HLA-A2þ was 22.1% in PBMC alone and 60.8% in cocultures (Fig. 5, right).
TNF-a or IFN-c Induces Apoptosis in HCEC We investigated whether the release of proinflammatory cytokines or the accumulation of Kyn affects the viability of HCEC. Cell death was determined by detection of 7-aminoactinomycin (7-AAD, a marker for the early phase of apoptosis) and Annexin V (a marker of cell death) upon
FIGURE 3. Increased IDO activity in HCEC does not suppress T-cell proliferation. T-cell proliferation was measured by incorporation of [3H]thymidine (cpm). Left: Human PBMC and HCEC were cultured alone or in combination in 250 lM L-Trp with or without addition of 3 mM DL-1methyl-tryptophan (MT). Cocultured cells showed a much higher cpm count as compared to PBMC alone (P 0.001). The addition of DL 1-MT did not significantly change cpm. Middle: Carboxyfluorescein diacetate succinimidyl ester dye was used as a tracer of T-cells in order to demonstrate that allogeneic T-cells proliferate. Right: PBMC were cultivated in the presence of Muromonab-CD3 (OKT3) in culture media containing 250 lM LTrp with increasing concentrations of Kyn (100–1600 lM). Concentrations ‡400 lM exhibited a statistically significant (P 0.001) effect on T-cell proliferation. Bars represent means 6 standard deviation; ns, not significant.
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1218
FIGURE 4. IFN-c upregulates expression of HLA class II molecules on HCEC: HLA class I and II surface expression on HCEC was determined by flow cytometry. Upper row: HLA-I (MHC-I) expression, from left to right: isotype, without stimulation, with IFN-c stimulation, with TNF-a stimulation. Lower row: HLA-II (MHC-II), from left to right: isotype, expression without stimulation, with IFN-c stimulation, with TNF-a stimulation.
treating HCEC with IFN-c or TNF-a (Fig. 5, first and second rows), or by adding various concentrations of Kyn (100, 200, 400, 800, and 1600 lM; Fig. 6, graph, third row). The results showed a statistically not significant increase of 7-AAD after incubation with IFN-c (P ¼ 0.13, compared to untreated HCEC, Fig. 5, first row) and a stronger, statistically significant increase after TNF-a treatment (P ¼ 0.02, compared to untreated HCEC, Fig. 5, first row). For Annexin V, similarly, an increase was observed after incubation with IFN-c (P ¼ 0.13, compared to untreated HCEC, Fig. 6, second row), whereas treatment with TNF-a caused stronger, statistically significant increase (P ¼ 0.03, compared to untreated HCEC, Fig. 5, second row). In experiments using Kyn, apoptosis occurred only at concentrations of >800 lM (P 0.01, compared to treatment without Kyn, Fig. 6, third row). These concentrations clearly exceed those expected in vivo.
IFN-c Upregulates the Expression of IDO mRNA in the Presence of L- or D-Tryptophan For a better understanding of the role of IDO in HCEC, we examined if its expression is affected in the presence of the isomers L- or D-tryptophan. Human corneal endothelial cells were cultured with either L- or D-tryptophan (25 lM), with or without IFN-c, for 72 hours. Findings showed that stimulation of HCEC with IFN-c leads to increased IDO mRNA expression in the presence of both isoforms (L- or D-) of Trp (see Supplementary Fig. S1).
IDO in HCEC Degrades L- or D-Tryptophan Equally To investigate whether IDO can equally degrade both isoforms (L- or D-) of tryptophan in the cell culture media, HCEC were cultured with various concentrations of L- or D-tryptophan (25, 250, and 500 lM), with or without IFN-c, for 72 hours
FIGURE 5. Expression of HLA-DR after coculture of HCEC with allogeneic PBMC cells had the following characteristics: HCEC were HLA-A03þ/HLAA02, whereas PBMC were HLA-A03/HLA-A02þ. Double staining (for HLA-A and DR) and subsequent FACS analysis showed that HCEC were predominantly HLA-A03þ/HLA-A02 (in over 98%) and PBMC HLA-A02þ/HLA-A03 (in over 90%). Percentage of cells with expression of both HLAA03þ and HLA-DR was 3% in HCEC alone, and 9% in HCEC cocultured with allo-PBMC. In contrast, percentage of cells with expression of HLA-DR and HLA-A02þ was 22.1% in PBMC (alone), and in coculture (HCEC and allo-PBMC) was 60.8% (right). Bars represent means 6 standard deviation.
Role of HCEC in T-Cell–Mediated Alloimmune Attack
IOVS j March 2014 j Vol. 55 j No. 3 j 1219
FIGURE 6. The proinflammatory cytokines TNF-a and IFN-c induce apoptosis in HCEC. Cell death was assessed by detection of 7-actinoaminomycin (7-AAD, marker for the early phase of apoptosis for cell death, first row) and Annexin V (marker for cell death, second row) upon treatment of HCEC with IFN-c or TNF-a, or by adding various concentrations of Kyn (100, 200, 400, 800, and 1600 lM, third row). Untreated HCEC served as control (left in each row). First and second rows: Both IFN-c and TNF-a led to an increase of 7-AAD and Annexin V. This increase was statistically significant (P 0.05) only in the case of TNF-a. Third row: Concentrations of >800 lM Kyn led to a statistically significant (P 0.01) increase of Annexin V.
(Supplementary Fig. S2, left). Without IFN-c-stimulation, neither the degradation of L- or D-tryptophan nor, of course, accumulation of Kyn was detectable in the supernatant of HCEC. Stimulation with IFN-c triggered activation of IDO and resulted in degradation of both isoforms of tryptophan and accumulation of Kyn (Supplementary Fig. S2, left). Interestingly, degradation of both isoforms of tryptophan (L- and D-) and accumulation of Kyn were approximately similar. To our knowledge, this is the first report on IDO activity in HCEC proving degradation of both isoforms of tryptophan. Another set of experiments was carried out to analyze the blocking of IDO activity. Human corneal endothelial cells were incubated with 250 lM L-tryptophan, IFN-c, and DL-1MT in concentrations of 1, 2, and 3 mM. As shown in Supplementary Figure S2, right, at a concentration of 3 mM, DL 1-MT completely inhibited Kyn production. This extremely high concentration seems to be required when HCEC are in an immunologically active state.
DISCUSSION In this study we aimed to investigate the response of HCEC to the alloimmune attack of T-cells. To better understand the interaction and cross-talk between HCEC and allo-lymphocytes, pro- and anti-inflammatory cytokine levels were determined under cocultivation, showing elevation of multiple proinflammatory cytokines (IL-1a, IL-1b, IL-6, IFN-c, and TNF-a). Supernatants of HEK 293 cells/PBMC cocultures showed a similar or even lower cytokine release compared to the supernatant of PMBC. However, it was markedly weaker than the cytokine release in the supernatant of HCEC/PBMC cocultures, proving that the observed cytokine induction in HCEC was not a generic response, but rather a cell-specific one. These findings are in accordance with previous studies pointing to these cytokines as mediators of the alloimmune response in corneal transplantation.33–36 Bosnar et al.33 demonstrated that in corneal recipients with inflammatory diseases, enhanced production of IL-1a and TNF-a is positively correlated with graft rejection. Since IFN-c is a strong inducer of IDO, its increased levels in supernatants of HCEC/PBMC cocultures prompted us to investigate whether IDO is induced in these cultures. Our
results showed strong upregulation of IDO expression and activity. Interestingly, upon stimulation with IFN-c, upregulation of IDO is approximately 100 times higher in HCEC than in PBMC, indicating that HCEC are the main source of IDO production. In addition, our data reveal that IDO in HCEC is functionally expressed in the presence of both L- and Dtryptophan, and that its enzymatic activity is able to degrade both isoforms of tryptophan equally. Indoleamine 2,3-dioxigenase activity was completely inhibited by DL 1-MT at a concentration of 3 mM. Several previous studies described the potential immunosuppressive role of IDO. In order to examine if IDO modulates the response of activated allo-lymphocytes, a lymphocyte proliferation assay was performed. Experiments carried out in the presence or absence of the IDO inhibitor DL 1-MT did not show significant differences in ratios of T-cell proliferation induced by HCEC in settings with high IDO activity. Since Kyn is a potential mediator of IDO-induced immunosuppression, we tested which Kyn concentration is able to inhibit T-cell proliferation. The results showed that only high concentrations of Kyn (‡400 lM) inhibited the proliferation of activated T lymphocytes. Since the plasma concentration of tryptophan is approximately 60 lM in normal healthy humans,37,38 it is questionable whether the concentration of Kyn required for T-cell suppression can be reached in vivo. Failure to reach suppressive Kyn concentrations in vivo might explain the positive (and not negative) correlation of IDO expression with kidney graft rejection found in clinical studies.21,22,24 Another explanation would be that Kyn is not the main mediator of suppression in this system. Our previous study showed that IDO expression in corneal endothelial cells is significantly higher in rejected corneas as compared to accepted grafts.23 Based on our current in vitro results, we interpret this finding as a consequence of increased IFN-c expression during an inflammatory alloimmune response. Along the same lines, Reeve et al. 39 demonstrated an increase of IDO gene expression in kidney transplant biopsies with acute rejection. In the normal cornea, cells do not express MHC class II molecules, with exception of the limbal margins.32 In order to examine HLA (HLA-I or -II) expression of HCEC in the presence of proinflammatory cytokines (such as IFN-c and TNF-a, which are commonly generated during rejection), FACS studies were
Role of HCEC in T-Cell–Mediated Alloimmune Attack performed. We found that HCEC express HLA-I constitutively: Treated and untreated HCEC express equal levels of HLA-I. On the other hand, expression of HLA-II (HLA-DR) was detectable only after stimulation with IFN-c. Stimulation with TNF-a was not capable of inducing the expression of HLA-II. Using additional markers, a 3-fold upregulation of HLA-II in cocultured HCEC was detected. Peripheral blood mononuclear cells showed a similar behavior. Decay attack factor (DAF) is a contributor to the immune privilege of the cornea. We aimed to find out whether the presence or absence of DAF molecule (CD55) on HCEC can correlate with alloimmune response. Our data showed that 93% of HCEC expressed DAF molecules on their surface (data not shown). The MHC class II antigens expressed on the surface of antigen-presenting cells (APCs) play a pivotal role in the primary T-cell response.40 Expression of MHC class II antigens is inducible by immunological stimulation and has been detected on many cell types that do not express class II antigens under physiological conditions.41–43 In renal grafts, MHC class II antigens have been demonstrated to elicit strong immune responses,44,45 and compatibility at the HLA-DR (class II) locus was found to be of overriding importance for graft survival when compared to compatibility at the HLA-A and -B (class I) loci during the first posttransplant year.46 Hence, the induction of MHC class II antigens on HCEC further underlines the important role of these cells in immunological responses. We also investigated the induction of cell death in HCEC in the presence of IFN-c and TNF-a in the culture media. Both cytokines were able to affect the viability of HCEC. Tumor necrosis factor-alpha alone induced a very high rate of apoptosis in HCEC in vitro. Based on these results, we conclude that proinflammatory cytokines (especially TNF-a) produced during the alloimmune response might cause endothelial cell loss during corneal allograft rejection. Sagoo et al.6 showed that inflammatory cytokines (TNF-a, IFN-c, and IL-1a) induced high rates of apoptosis in mouse corneal endothelium. Moreover, Niederkorn et al.47 demonstrated high susceptibility of murine corneal epithelial and endothelial cells to TNF-a–induced apoptosis. To our knowledge, our work is the first report on proinflammatory cytokine induction of apoptosis in HCEC. As a consequence, avoidance of major postoperative inflammatory episodes should be an imperative paradigm in treatment. In summary, we believe that the results presented here improve current knowledge on the important immunological role of HCEC and IDO in cornea transplantation and provide the basis for further in vivo studies. Although IDO activity was increased during the immune attack of HCEC, it did not play an immunosuppressive role. These findings support a possible correlation of IDO expression/activity with graft rejection in clinical transplantation, but do not support a graft protective role of IDO. Consequently, increased IDO activity might be a predictive marker for rejection, but not for good graft function.
Acknowledgments Helmut Simon, Martina Kutsche-Bauer, and Regina Seemuth provided skillful technical assistance. Disclosure: I. Lahdou, P; C. Engler, None; S. Mehrle, P; V. Daniel, None; M. Sadeghi, None; G. Opelz, P; P. Terness, P
References 1. Niederkorn JY. Immunology and immunomodulation of corneal transplantation. Int Rev Immunol. 2002;21:173–196.
IOVS j March 2014 j Vol. 55 j No. 3 j 1220 2. Niederkorn JY, Larkin DF. Immune privilege of corneal allografts. Ocul Immunol Inflamm. 18:162–171. 3. Cornea Donor Study Investigator Group, Gal RL, Dontchev M, et al. The effect of donor age on corneal transplantation outcome results of the cornea donor study. Ophthalmology. 2008;115:620–626.e6. 4. George AJ, Larkin DF. Corneal transplantation: the forgotten graft. Am J Transplant. 2004;4:678–685. 5. Hori J, Joyce NC, Streilein JW. Immune privilege and immunogenicity reside among different layers of the mouse cornea. Invest Ophthalmol Vis Sci. 2000;41:3032–3042. 6. Sagoo P, Chan G, Larkin DF, George AJ. Inflammatory cytokines induce apoptosis of corneal endothelium through nitric oxide. Invest Ophthalmol Vis Sci. 2004;45:3964–3973. 7. Barcia RN, Dana MR, Kazlauskas A. Corneal graft rejection is accompanied by apoptosis of the endothelium and is prevented by gene therapy with bcl-xL. Am J Transplant. 2007;7:2082–2089. 8. Ayliffe W, Alam Y, Bell EB, McLeod D, Hutchinson IV. Prolongation of rat corneal graft survival by treatment with anti-CD4 monoclonal antibody. Br J Ophthalmol. 1992;76: 602–606. 9. He YG, Ross J, Niederkorn JY. Promotion of murine orthotopic corneal allograft survival by systemic administration of antiCD4 monoclonal antibody. Invest Ophthalmol Vis Sci. 1991; 32:2723–2728. 10. Hegde S, Beauregard C, Mayhew E, Niederkorn JY. CD4(þ) Tcell-mediated mechanisms of corneal allograft rejection: role of Fas-induced apoptosis. Transplantation. 2005;79:23–31. 11. Yamada J, Kurimoto I, Streilein JW. Role of CD4þ T cells in immunobiology of orthotopic corneal transplants in mice. Invest Ophthalmol Vis Sci. 1999;40:2614–2621. 12. Cua DJ, Coffman RL, Stohlman SA. Exposure to T helper 2 cytokines in vivo before encounter with antigen selects for T helper subsets via alterations in antigen-presenting cell function. J Immunol. 1996;157:2830–2836. 13. Pober JS, Gimbrone MA Jr, Lapierre LA, et al. Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J Immunol. 1986;137:1893–1896. 14. Pepose JS, Nestor MS, Gardner KM, Foos RY, Pettit TH. Composition of cellular infiltrates in rejected human corneal allografts. Graefes Arch Clin Exp Ophthalmol. 1985;222:128– 133. 15. Torres PF, De Vos AF, van der Gaag R, Martins B, Kijlstra A. Cytokine mRNA expression during experimental corneal allograft rejection. Exp Eye Res. 1996;63:453–461. 16. Zhu S, Dekaris I, Duncker G, Dana MR. Early expression of proinflammatory cytokines interleukin-1 and tumor necrosis factor-alpha after corneal transplantation. J Interferon Cytokine Res. 1999;19:661–669. 17. Takikawa O. Biochemical and medical aspects of the indoleamine 2,3-dioxygenase-initiated L-tryptophan metabolism. Biochem Biophys Res Commun. 2005;338:12–19. 18. Li J, Meinhardt A, Roehrich ME, et al. Indoleamine 2,3dioxygenase gene transfer prolongs cardiac allograft survival. Am J Physiol Heart Circ Physiol. 2007;293:H3415–H3423. 19. Yu G, Dai H, Chen J, et al. Gene delivery of indoleamine 2,3dioxygenase prolongs cardiac allograft survival by shaping the types of T-cell responses. J Gene Med. 2008;10:754–761. 20. Beutelspacher SC, Pillai R, Watson MP, et al. Function of indoleamine 2,3-dioxygenase in corneal allograft rejection and prolongation of allograft survival by over-expression. Eur J Immunol. 2006;36:690–700. 21. Holmes EW, Russell PM, Kinzler GJ, et al. Oxidative tryptophan metabolism in renal allograft recipients: increased kynurenine
Role of HCEC in T-Cell–Mediated Alloimmune Attack
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32. 33.
synthesis is associated with inflammation and OKT3 therapy. Cytokine. 1992;4:205–213. Brandacher G, Cakar F, Winkler C, et al. Non-invasive monitoring of kidney allograft rejection through IDO metabolism evaluation. Kidney Int. 2007;71:60–67. Serbecic N, Lahdou I, Scheuerle A, Hoftberger R, Aboul-Enein F. Function of the tryptophan metabolite, L-kynurenine, in human corneal endothelial cells. Mol Vis. 2009;15:1312–1324. Lahdou I, Sadeghi M, Daniel V, et al. Increased pretransplantation plasma kynurenine levels do not protect from but predict acute kidney allograft rejection. Hum Immunol. 71:1067–1072. Suarez-Fuentetaja N, Domenech-Garcia N, Paniagua-Martin MJ, et al. Indoleamine, 2-3 dioxygenase activity could be an early marker of graft rejection in heart transplantation. Transplant Proc. 44:2645–2648. Bednarz J, Teifel M, Friedl P, Engelmann K. Immortalization of human corneal endothelial cells using electroporation protocol optimized for human corneal endothelial and human retinal pigment epithelial cells. Acta Ophthalmol Scand. 2000; 78:130–136. Terness P, Chuang JJ, Bauer T, Jiga L, Opelz G. Regulation of human auto- and alloreactive T cells by indoleamine 2,3dioxygenase (IDO)-producing dendritic cells: too much ado about IDO? Blood. 2005;105:2480–2486. Dennis-Sykes CA, Miller WJ, McAleer WJ. A quantitative Western Blot method for protein measurement. J Biol Stand. 1985;13:309–314. Hu H, Kwun J, Aizenstein BD, Knechtle SJ. Noninvasive detection of acute and chronic injuries in human renal transplant by elevation of multiple cytokines/chemokines in urine. Transplantation. 2009;87:1814–1820. Karczewski M, Karczewski J, Poniedzialek B, Wiktorowicz K, Glyda M. Cytometric analysis of TH1/TH2 cytokines in the urine of patients undergoing kidney transplantation. Ann Transplant. 2009;14:25–28. Xu J, Wei J, Zhu X, et al. Increased plasma indoleamine 2,3dioxygenase activity and interferon-c levels correlate with the severity of acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2013;19:196–201. Streilein JW. New thoughts on the immunology of corneal transplantation. Eye (Lond). 2003;17:943–948. Bosnar D, Dekaris I, Gabric N, Markotic A, Lazic R, Spoljaric N. Influence of interleukin-1alpha and tumor necrosis factoralpha production on corneal graft survival. Croat Med J. 2006; 47:59–66.
IOVS j March 2014 j Vol. 55 j No. 3 j 1221 34. Jin Y, Chai LJ, Hong J. Early expression of tumor necrosis factor alpha and nuclear factor kappa B after Descemet stripping endothelial keratoplasty in rabbits. Cornea. 31:913–919. 35. Maier P, Heizmann U, Bohringer D, Kern Y, Reinhard T. Predicting the risk for corneal graft rejection by aqueous humor analysis. Mol Vis. 2011;17:1016–1023. 36. van Gelderen BE, Van Der Lelij A, Peek R, et al. Cytokines in aqueous humour and serum before and after corneal transplantation and during rejection. Ophthalmic Res. 2000; 32:157–164. 37. Kudo Y, Boyd CA, Sargent IL, Redman CW. Decreased tryptophan catabolism by placental indoleamine 2,3-dioxygenase in preeclampsia. Am J Obstet Gynecol. 2003;188:719– 726. 38. Silk JD, Lakhal S, Laynes R, et al. IDO induces expression of a novel tryptophan transporter in mouse and human tumor cells. J Immunol. 2011;187:1617–1625. 39. Reeve J, Einecke G, Mengel M, et al. Diagnosing rejection in renal transplants: a comparison of molecular- and histopathology-based approaches. Am J Transplant. 2009;9:1802–1810. 40. Rosenthal AS, Shevach EM. Function of macrophages in antigen recognition by guinea pig T lymphocytes. I. Requirement for histocompatible macrophages and lymphocytes. J Exp Med. 1973;138:1194–1212. 41. Hall BM, Bishop GA, Duggin GG, Horvath JS, Philips J, Tiller DJ. Increased expression of HLA-DR antigens on renal tubular cells in renal transplants: relevance to the rejection response. Lancet. 1984;2:247–251. 42. Volc-Platzer B, Majdic O, Knapp W, et al. Evidence of HLA-DR antigen biosynthesis by human keratinocytes in disease. J Exp Med. 1984;159:1784–1789. 43. Iwata M, Yagihashi A, Roat MI, Zeevi A, Iwaki Y, Thoft RA. Human leukocyte antigen-class II-positive human corneal epithelial cells activate allogeneic T cells. Invest Ophthalmol Vis Sci. 1994;35:3991–4000. 44. Moen T, Albrechtsen D, Flatmark A, et al. Importance of HLADR matching in cadaveric renal transplantation: a prospective one-center study of 170 transplants. N Engl J Med. 1980;303: 850–854. 45. Young E, Stark WJ, Prendergast RA. Immunology of corneal allograft rejection: HLA-DR antigens on human corneal cells. Invest Ophthalmol Vis Sci. 1985;26:571–574. 46. Opelz G, Wujciak T, Dohler B, Scherer S, Mytilineos J. HLA compatibility and organ transplant survival. Collaborative Transplant Study. Rev Immunogenet. 1999;1:334–342. 47. Niederkorn JY, Mayhew E, Mellon J, Hegde S. Role of tumor necrosis factor receptor expression in anterior chamberassociated immune deviation (ACAID) and corneal allograft survival. Invest Ophthalmol Vis Sci. 2004;45:2674–2681.
