- Email: [email protected]
Human Retinal Pigment Epithelial Cells Inhibit Proliferation and IL2R Expression of Activated T Cells
Exp. Eye Res. (2002) 74, 627±637 doi:10.1006/exer.2002.1183, available online at http://www.idealibrary.com on
Human Retinal Pigment Epithelial Cells Inhibit Proliferation and IL2R Expression of Activated T Cells C H A R LOT T E G . K á S T E L a, A N N E T T E Jé R G EN S E N a, M ETT E N IEL S EN b, K A R S TE N W. ER IKS EN b, N I E L S é D U M b, M O G EN S H O LS T N IS S EN a A N D CA R S T EN R OÈ P K E a* a
Institute of Medical Anatomy Section A, The Panum Institute, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen N, Denmark and bInstitute of Medical Microbiology and Immunology, The Panum Institute, University of Copenhagen, Denmark. (Received St. Louis 30 October 2001 and accepted in revised form 5 February 2002) The purpose of this study was to characterize the effects of human retinal pigment epithelial (RPE) cells on activated T cells. Activated T cells were cocultured with adult and foetal human RPE cells whereafter apoptosis and proliferation were determined by ¯ow cytometry and 3H-Thymidine incorporation assay, respectively. T cells and RPE cells were cultured directly together or in a transwell system for determination of the effect of cell contact. The importance of cell surface molecules was examined by application of a panel of blocking antibodies (CD2, CD18, CD40, CD40L, CD54, CD58) in addition to use of TCR negative T cell lines. The expression of IL2R-a -b and -g chains of activated T cells was analysed by ¯ow cytometry after incubation of T cells alone or with RPE cells. Human RPE cells were found to inhibit the proliferation of activated T cells by a cell contact-dependent mechanism. The RPE cells inhibitory abilities were not affected by blocking of any of the tested surface molecules. The inhibition of the T cells' proliferation correlates with a decreased expression of IL2R-b and -g chains. The T cells regain their ability to proliferate and increase their IL2R-b and -g chain expression within 24 hr after removal from the coculture. It is concluded that the cultured human adult and foetal RPE cells inhibit the proliferation of activated T cells by a process that does not involve apoptosis. It depends on cell contact but the involved surface molecules were not revealed. The proliferation inhibition correlates with a modulation of the T cells' expression of IL2R, and is reversible. # 2002 Elsevier Science Ltd. Key words: retinal pigment epithelial; T cells; proliferation; IL2R-g; inhibition.
1. Introduction The eye is known as an immune privileged organ and is characterized by allowing prolonged survival of foreign tissue grafts. Several mechanisms have been suggested to support this privilege, starting with the physical isolation of the eye from the systemic blood circulation due to the presence of blood: tissue barriers, as well as relative absence of lymphatic drainage from the eye (Streilein, 1993). Additionally, molecular features such as surface expression of FasL by ocular cells (cornea, retina, iris and ciliary body) (Grif®th et al., 1995; Kaplan et al., 1999; Kñstel et al., 2001) and membrane-bound inhibitors of complement activation (Morgan and Meri, 1994), as well as the presence of immune suppressive soluble factors in intraocular ¯uids, especially aqueous humor, have been proposed to be involved in the phenomenon (Kaiser et al., 1989; Cousins et al., 1991; Taylor, Streilein and Cousins, 1992, 1994; Taylor et al., 1998). There is, however, still many unanswered questions with regard to how the immune privilege of the eye is obtained (Streilein and Stein-Streilein, 2000). A number of recent studies have shown that retinal pigment epithelial (RPE) cells cocultured with activated T cells inhibit the T cell proliferation (Liversidge et al., * Author for correspondence. E-mail: [email protected]
0014-4835/02/ $35.00/0
1993; Jùrgensen et al., 1998; Farrokh-Siar et al., 1999a,b, 2000; Rezai et al., 1999; Willermain et al., 2000), but the mechanisms behind this inhibition are to this date still largely unknown. Human foetal RPE cellsÐnot expressing FasL themselves (Kñstel et al., 2001), have been found to induce apoptosis in both FasL independent and dependent manners (Rezai et al., 1999; Farrokh-Siar et al., 1999a, 2000; and Jùrgensen et al., 1998), whereas adult rat RPE cells as well as murine iris and ciliary body pigment epithelial cells, which resemble RPE cells, have been found to inhibit the proliferation via unknown membrane bound factors as well as soluble factors such as prostaglandin E2(PGE2) and TGF-b (Liversidge et al., 1993; Yoshida, Takeuchi and Streilein, 2000a; Yoshida et al., 2000b). In an attempt to understand the mechanisms involved in the RPE cell induced inhibition of T cells, the phenotype of the activated T cells has been analysed with regard to their expression of IL2R-a, -b and -g chain, and the long-term effect of the T cells' encounter with the RPE cells. 2. Materials and Methods Antibodies and Other Reagents Mouse IgG1 and IgG2a, mouse monoclonal antihuman IL2Rb-PE Ab and CD40L Ab as well as rat # 2002 Elsevier Science Ltd.
628
monoclonal anti-IL2Rg-PE Ab were from Pharmingen (San Diego, CA, U.S.A.). Mouse monoclonal antihuman CD3-FITC Ab, CD3-PE Ab and CD25-FITC Ab were from Dako (Glostrup, Denmark). Mouse antihuman CD40 Ab was from Mabtech (Nacka, Sweden). Mouse anti-human CD18, CD54 and CD58 were from Immunotech (Marseilles, France). Retinal Pigment Epithelial Cell Culture The human adult RPE cell line ARPE-19 was purchased from ATTC (Manassas, VA, U.S.A.) and human foetal RPE (HFRPE) cells were a kind gift from Dr Tongalp H. Tezel. The HFRPE cells were isolated from a 17 week old male fetus, harvested within 6 hr of an elective abortion. Method of securing human tissues complied with the Declaration of Helsinki. Harvesting and culturing technique is described in detail in Tezel et al., (1997). The RPE cells were thawed and seeded on tissueculture plates (Falcon; Becton Dickenson Labware, Bedford, MA, U.S.A.) at a density of 8000 cells cm 2. ARPE-19 cells were seeded directly on plastic where as the HFRPE cells were seeded on tissue-culture plates precoated with extracellular matrix prepared as described by Gospodarowicz and Ill (1980). After reaching con¯uence, they were transferred at a split ratio of 1 : 5 and seeded onto new tissue-culture plates, and repeatedly transferred at a split ratio of 1 : 8. When the cells had acquired a hexagonal shape and pigmentation after 2±4 weeks, they were applied in coculture experiments. ARPE-19 cells were maintained in Dulbecco's modi®ed Eagle's medium (H16; Gibco, Life Technologies, Paisley, Scotland, U.K.) supplemented with 15 % foetal calf serum (FCS; HyClone Laboratories, Logan, UT, U.S.A.) 300 mg ml 1 glutamine 50 mg ml 1 gentamycin and 2.5 mg ml 1 amphotericin B (Fungizone; Gibco, Life Technologies, Paisley, Scotland, U.K.). HFRPE cells were maintained in similar medium further supplemented with 1 ng ml 1 bFGF (Sigma St. Louis, MO, U.S.A.). Medium was changed every other day and the cells were regularly tested for mycoplasma. The ARPE19 cells used in this study were passage 23 or 26 and the HFRPE cells passage 3 or 4. The ARPE-19 cell line has been found to have comparable expression with human foetal RPE (p6) cells of the following surface molecules: MHC class I and II, CD40, CD48, CD54, CD58, CD68, CD80, CD86, CD95, CD95L (Jùrgensen et al., 2001; Kñstel et al., 2001). Peripheral Blood Lymphocytes Peripheral blood mononuclear cells (PBMC) were separated from buffycoats (The Blood Bank, National University Hospital, Copenhagen, Denmark) by lymphoprep (Nycomed, Oslo, Norway) density-gradient. The PBMCs (106 cells ml 1) were either stimulated with 4 mg ml 1 phytohemagglutinin (PHA, Welcome-
C. G. KáSTEL ET AL.
Difco, Detroit, MI, U.S.A.) in RPMI 1640 supplemented with 10 % FCS (Gibco, Life Technologies, Paisley, Scotland, U.K.), 200 U ml 1 soluble recombinant IL2 (Chiron, CA, U.S.A.), 2 mM L-glutamine, 100 mg ml 1 penicillin and 100 mg ml 1 streptomycin (The Panum Institute, Copenhagen, Denmark) and immediately cocultured with RPE cells or prestimulated with 4 mg ml 1 phytohemagglutinin, pelleted and resuspended in culture medium (1 106 cells ml 1) with 10 % FCS and 200 U ml 1 IL2 after 4 days and then after another 4 days applied to coculture experiments. A new blood donor was used for each repetitive experiment. T-cell Lines The T-cell lines, MF and Se-Ax, are lymphoid tumor cell lines established from a plaque biopsy specimen of patients with mycosis fungoides (MF) or the SeÂzary syndrome as described in detail elsewhere (Kaltoft et al., 1987, 1992). MF and Se-Ax cells were cultured in RPMI 1640 (Gibco, Life Technologies, Paisley, Scotland, U.K.) supplemented with 10 % FCS or 10 % human serum (HS), 2 mM L-glutamine, 100 mg ml 1 penicillin and 100 mg ml 1 streptomycin. The media of the Se-Ax cell line were additionaly supplemented with 1000 U ml 1 IL2 (Chiron B.V; Amsterdam, The Netherlands). Coculture of T Cells and RPE Cells RPE cells were trypsinised, resuspended in DMEM : H16 culture medium with or without bFGF and seeded onto 96 well (5 104 cells per well) or 24 well (2 105 cells per well) ¯atbottom culture plates (NUNC A/S, Kamstrup, Denmark) and incubated overnight in 10 % CO2, 378C, allowing the cells to stick to the bottom of the wells. The RPE cells were then washed in PBS (2) and T cells were added either directly or in inserts with a pore size of 0.2 mm (NUNC). All cocultures were maintained in RPMI 1640 supplemented with 10 % FCS with 200 U ml 1 soluble recombinant IL2, unless otherwise noted and incubated in 5 % CO2 at 378C. Cocultured cells were harvested for ¯ow cytometric analysis by pipetting both T cells and RPE cells with a 1 ml pipette. Trypsination was avoided due to the relatively short culturing of the RPE cells in the actual culture plate. The wells were checked after harvesting by light microscopy to make sure that all cells had been harvested. In order to investigate the long-term effect of ARPE19 and HFRPE cells on T cells, the T cells were carefully removed via pipette from coculture, avoiding the RPE cells. It was not possible to get all the T cells out of the coculture due to adhesion between the two types of cells, however, a recovery percentage was estimated to be no less than 75 % (number of T cells recovered from coculture in compared to number of T cells recovered from single culture).
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
Flow Cytometry The cell cycle of peripheral T cells after coculture with RPE cells was analysed by surface staining of CD3 followed by intracellular DNA staining with 7-amino-actinomycin D (7-AAD; Sigma St. Louis, MO, U.S.A.) (Rabinovitch, Torres and Engel, 1986), and ¯ow cytometry analysis. The cells (3±5 105 cells per sample) were washed in PBS stained for CD3 surface expression by adding anti-CD3-FITC Ab or isotype control Ab followed by incubation on ice for 30 min. Afterwards the cells were washed in PBS and then in PBS/0.05 % saponin (Sigma St. Louis, MO, U.S.A.), resuspended in 4 mg ml 1 7-AAD in PBS/ 0.05 % saponin and left to incubate in the dark at room temperature for 30 min before analysis on a FACSCalibur apparatus (Becton Dickinson, San Jose, CA, U.S.A.). The surface expression of IL2R-a, -b and -g chain of CD3 positive cells after culturing alone or with RPE cells was measured by use of anti-CD3-FITC or antiCD3-PE and anti-IL2R-a-FITC, anti-IL2R-b-PE or anti-IL2R-g-PE Ab as well as isotype control antibodies followed by ¯ow cytometry. When analyzing, a gate of live CD3 positive cells was applied, and the percentage of cells in the different cell cycle phases or the mean ¯uorescence intensity (MFI) of the IL2R chains of these cells was measured and used for statistical calculations. 3
H-Thymidine Incorporation Assay
The proliferation of T cells in culture alone or in coculture with RPE cells was measured by incorporation of 3H-Thymidine (1 mCi per well, 1 Ci 37 Gbp) (N.E.N., Boston, MA, U.S.A.) during the last 6 hr of culture. The incorporation of 3H-Thymidine in the RPE cells was measured as a control. It was found in all experiments that the RPE cells did not signi®cantly contribute to the incorporation of 3H-Thymidine in the coculture experiments. All assays were performed in 96 well culture plates or transwell systems of 96 well culture plates (NUNC). 3. Results RPE Cells Induce S and G2/M Phase Decreases but not Apoptosis in Cycling T cells To evaluate the effects of adult RPE cells on activated T cells, T cells were set up in IL2 containing cultures either alone or in the presence of ARPE-19 cells, which were added to the chambers 24 hr before coculture to allow the cells to adhere to the bottom. Fig. 1 shows analysis of cell cycle of T cells after culturing alone or with RPE cells via labelling with CD3 and 7-AAD for selection of T cells and for staining the DNA, respectively. It is seen that the percentages of CD3 positive T cells in S G2/M phases decrease signi®cantly in cocultures compared
629
with percentages in cultures of T cells alone. Similar results were obtained with cocultures of T cells and adult RPE (p0) cells as well as foetal RPE cells (p6) (data not shown). The ®gure further shows that the decrease in cycling cells is not accompanied by an increase in cells with subnormal DNA i.e. apoptosis. This was paralleled by ¯ow cytometric analysis of the early apoptosis marker Annexin V and the antiapoptotic proteins Bcl-2 and Bcl-xL as well as the proapoptotic protein Bax. No signi®cant difference was found in the expression of these proteins between T cells cultured alone or with ARPE-19 cells for 48 hr (data not shown). Since the ¯ow cytometric analysis by the use of 7-AAD at harvest depicts the collected events from the entire culture period, the authors used 3H-Thymidine labelling of cells to evaluate the DNA synthetic activity in selected time periods. Fig. 2(A) shows an experiment in which T cells were cultured for up to 72 hr with or without ARPE-19 cells. It is clearly seen that T cells maintain comparable DNA synthesis in both culture types at 18±24 hr, whereas a very pronounced decrease in DNA synthesis is seen in cocultures at the 42±48 and 66±72 hr time periods. Similar results were obtained with g-irradiated, nonproliferating RPE cells, data not shown. Fig. 2(B) illustrates the effect of ARPE-19 cells by the use of varying ratios of ARPE-19 cells and T cells from (1 : 1) to (0.25 : 1) for 48 hr. The ARPE-19 cell induced inhibition was ratio-dependent, and absent at the ratio (0.25 : 1) in these culture chambers. In addition, it was found that PHA and IL2 induced proliferation of freshly isolated peripheral T cells is inhibited in the presence of ARPE-19 cell (data not shown). The Inhibition of T Cell Proliferation is Cell Contact-dependent The role of cell contact in the ARPE-19 induced T cell inhibition was analysed in a transwell system. The ARPE-19 cells were placed in the lower chamber and separated by a membrane from the upper chamber containing T cells. By having the ARPE-19 cells in the lower chamber instead of in the upper chamber, `clotting' at the membrane by the epithelial cells adhering to the membrane was avoided. The membrane has a pore size of 0.2 mm and does not permit the transmigration of cells, but soluble factors can easily diffuse through the membrane. Fig. 3 shows both the proliferation of T cells cultured alone and the proliferation of cocultured T cells. T cells were either cocultured directly on top of ARPE-19 cells or in the upper chamber in a transwell system with ARPE-19 cells in the bottom chamber, or with T cells in the upper chamber and with both ARPE-19 cells and T cells in the bottom chamber. As can be seen from the ®gure, the ARPE-19 cell induced inhibition of proliferation of T cells is clearly reduced by the
630
C. G. KáSTEL ET AL.
F IG . 1. Flow cytometric analysis of the cell cycle of T cells after culturing alone (T) or with ARPE-19 cells (T R) for 24, 48 or 72 hr. The T cells were selected for 7-AAD analysis by gating on the CD3 labelled cells. Left bars indicate percentage of apoptosis and right bars the percentage of cells in S G2/M phase compared to total amount of cells. The depicted results are representative for three separate experiments.
absence of cell contact in the transwell culture system. Additionally there was no difference between the effect of T cells cultured in the upper chamber with ARPE-19 cells in the bottom chamber, compared to T cells cultured in the upper chamber with both ARPE19 cells and T cells in the bottom chamber. The Inhibition of T Cell Proliferation is Independent of the Surface Molecules CD2, CD18, CD40, CD40L, CD54, CD58 and of TCR and MHC Molecules The adhesion molecules CD54 and CD58 are known to in¯uence the interaction between T cells and other cell types including epithelial cells (Jùrgensen et al., 1996, 2001). The authors found both these molecules to be expressed on human cultured adult and foetal RPE cells (data not shown). Further, human RPE cells were recently found to also
express the costimulatory surface molecule CD40 in response to IFN-g stimulation (Willermain et al., 2000). Blocking antibodies against these molecules or their counterparts CD18 (LFA-1), CD2 (LFA-3) and CD40L were added to cocultures of T cells and ARPE19 cells. It was found that blocking of either of these molecules had no signi®cant effect on the ability of ARPE-19 cells to inhibit T-cells, see Fig. 4. Further, pre-treatment of ARPE-19 cells with IFN-g, which in addition to up-regulation of CD40 up-regulate the expression of CD54, MHCI and II (Liversidge et al., 1988; Zavazava et al., 1991; Elner et al., 1992; Platts et al., 1995), did not lead to changes in the effects of ARPE-19 cells on T cells (data not shown). This indicates that these molecules, including MHC molecules, are of minor importance in the ARPE-19 cell±T cell interaction. This was corroborated by the use of the T cell lines MF and Se-Ax that do not
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
631
F IG . 2. Proliferation of activated T cells after either culturing alone (T) or with ARPE-19 cells (T R) (1 : 1) for 24, 48 or 72 hr (A), or after incubation with a varying number of ARPE-19 cells for 48 hr (B). ARPE-19 cells were seeded onto a 96 well culture plate and activated T cells (5 105) were added the following day. Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of four replicates from a representative experiment of three. Statistical analysis with unpaired t-test showed that ARPE-19 cells signi®cantly inhibited the T cells at 48 and 72 hr (A) and at the ratios (1 : 1) and (0.5 : 1) (B) *P 5 0.05.
express functional TCR. ARPE-19 cells induced signi®cant decreases in DNA synthesis in both cell lines in the presence or absence of IL2 (see Fig. 5), supporting that TCR±MHC class I/II interaction is not essential for the observed suppression of the DNA synthesis in activated T cells. As can be seen from the ®gure, the presence of IL2 did not change the effect of ARPE-19 cells on these cell lines.
T Cell Proliferation: Dependence of IL2 Since the T cells are maintained in cell cycle by addition of IL2 to the cultures, the ®ndings of decreased DNA synthesis in T cells in cocultures with ARPE-19 cells may be caused by an inef®cient utilization of IL2 by the T cells in the cocultures. It was therefore tested if the proliferation inhibition could be overcome by the addition of increased amounts of IL2 in the coculture. As can be seen from Fig. 6, IL2 is necessary for T cell proliferation, but there is no signi®cant difference between the proliferation of T cells in cocultures with 100 U ml 1 compared to cocultures with up to 800 U ml 1 IL2.
RPE Cells Induce Down-regulation of IL2R-b and -g Chain Expression
F IG . 3. Effect of cell contact on ARPE-19 cell induced inhibition of T cells. The proliferation of activated T cells was measured after either incubation of T cells alone (T), of T cells directly together with ARPE-19 cells (T R) (1 : 1), of T cells grown in a transwell system (pore size 0.2 mm) separated from an ARPE-19 cell monolayer (T(in) R) (1 : 1) or T cells in transwell system separated from an ARPE-19 cell monolayer together with activated T cells (T(in) R T) (1 : 1 : 1) for 48 hr. ARPE-19 cells (5 105) were seeded onto a 96 well culture plate and activated T cells (5 105) were added the following day directly on top of the RPE cells or in an insert. Proliferation was measured by adding 3 H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of three replicates from a representative experiment of four. Statistical analysis with unpaired t-test showed that ARPE19 cells only inhibited the proliferation of T cells signi®cantly when the T cells were incubated directly on top of ARPE-19 cells *P 5 0.05.
Since T cell proliferation was maintained at comparable levels over a wide range of IL2 concentrations, the authors speculated that the RPE cell induced down-regulation of proliferation could be due to changes in IL2R signalling. Accordingly, the expression pattern of IL2R-a, -b and -g chains of T cells cultured either alone or cocultured with ARPE19 or HFRPE cells was investigated by ¯ow cytometry. Fig. 7 illustrates the mean expression of IL2R-a, -b and -g chains on CD3 positive cells after incubation of T cells alone or in coculture with ARPE-19 cells or HFRPE cells. Both foetal and adult derived RPE cells convert decreased IL2R-b and -g chain expression on T cells, which constitute the signal transducing segments of IL2R. The expression of IL2R chains is generally regulated by the presence of IL2. It was therefore tested if there was any effect of a lower concentration of IL2 in coculture media (50 U ml 1). There were no indications that the RPE cell induced
632
C. G. KáSTEL ET AL.
F IG . 4. Evaluation of the effect of the surface molecules CD2, CD18, CD40 and CD40L (A) and CD54 and CD58 (B) on ARPE19 cell induced inhibition of T cell proliferation. The proliferation of activated T cells was measured after either incubation alone or with ARPE-19 cells (1 : 1) with or without the presence of blocking antibodies as indicated. Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. Illustrated is the percent of proliferation of cocultured T cells compared to T cells cultured alone. The results depicted are the mean of three separate experiments each including three to four replicates. Statistical analysis with paired t-test showed that there was no signi®cant difference between the proliferation of T cells in coculture in the presence of blocking Ab or in the presence of the matching isotype control Ab.
F IG . 5. Effect of ARPE-19 cells (R) on the TCR negative T cell lines MF and Se-Ax. The proliferation of MF or Se-Ax cells was measured after either incubation alone or with ARPE-19 cells (1 : 1) for 48 hr in the absence (A) or presence (B) of IL2. ARPE19 cells (5 105) were seeded onto a 96 well culture plate and the following day MF or Se-Ax cells were added (5 105). Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of four replicates from a representative experiment of three. Statistical analysis with unpaired t-test showed that ARPE-19 cells inhibited both MF and Se-Ax signi®cantly in the absence as well as presence of IL2 *P 5 0.05.
IL2R chain modulation was sensitive to the lower IL2 concentration (data not shown). The T cells' expression of IL2R-a chain is signi®cantly up-regulated or unchanged after coincubation with ARPE-19 or HFRPE. Together these results suggest that the RPE cell induced inhibition of T cells may be achieved by the modulation of the T cells' normal expression pattern of IL2R-b and -g chain and possibly the IL2 signal. The Inhibition of T Cell Proliferation and IL2R Expression is Reversible The 7-AAD experiments (Fig. 1) showed that the depression of DNA synthesis was not accompanied by an increase in apoptosis of the cocultured T cells within the investigated time period. However, it is
possible that the effect of RPE cells is an induction of lasting G0 state or anergy in the T cells. To test this, the ability of the T cells to regain proliferation and capacity to achieve a `normal' expression pattern of IL2R-a, -b and -g chains was investigated. After 48 hr incubation alone or with ARPE-19 cells, the T cells were carefully removed from the cultures, and resuspended without addition of ARPE-19 cells in fresh medium containing IL2. Hereafter, the level of proliferation was analysed by 3H-Thymidine incorporation, and the expression of IL2R-a, -b and -g chains was measured by ¯ow cytometry. As can be seen from Fig. 8, the T cells from cocultures with ARPE-19 cells have a lower level of proliferation during the ®rst 6 hr in fresh media compared to T cells having been precultured alone, but after 24 hr comparable levels of proliferation is seen in the two culture types. Fig. 8(B) shows that a
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
633
macrophages on activated T cells (Strickland, Kees and Holt, 1996; Upham et al., 1997). Decrease or arrest of proliferation seems to be required for some types of apoptosis (Radvanyi, Mills and Miller, 1993; Lissy et al., 1998; Lee et al., 1999). Thus a total or partial stop in proliferation and induction of apoptosis may be part of the same sequence of effects, and subtle differences between
F IG . 6. Evaluation of the effect of various concentrations of IL2 on ARPE-19 induced proliferation inhibition of T cells. ARPE-19 cells (R) were seeded onto a 96 well culture plate and activated T cells (T) (5 105) were added the following day. In addition, T cells and ARPE-19 cells were set up separately in parallel cultures. The cells were incubated for 48 hr. Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of four replicates from a representative experiment of three. Statistical analysis with unpaired t-test showed that the T cells in cocultures (T R) proliferate at the same rate in the presence of 100±800 U ml 1 IL2.
`normal' expression pattern of IL2R-b and -g chains is also achieved after 24 hr incubation in fresh media. Taken together, these results indicate that the RPE cell induced inhibition of T cells is a reversible process allowing a normal T cell function after removal of the RPE cells from the coculture. There is no sign of the T cells becoming suppressed in prolonged time after coincubation with RPE cells.
4. Discussion This study con®rm earlier studies in showing that human RPE cells in¯uence the cell cycle of T cells in vitro. The most signi®cant new ®nding in the present study is a decrease in functional IL2R expression, which is likely to explain the decrease in T cell proliferation and increase in apoptosis found in previous studies (Jùrgensen et al., 1998; Rezai et al., 1999; Farrokh-Siar et al., 1999a, 2000), since IL2 is essential for survival of activated T cells. In the present study, the changes in IL2R expression was paralleled by a decrease T cell proliferation. Decreased proliferation was also found in the above mentioned studies, but in addition, several of these studies showed an increase in apoptosis (Jùrgensen et al., 1998; Rezai et al., 1999; Farrokh-Siar et al., 2000). In the present study, there were, however, no signs of apoptotic processes or changes in several apoptotic proteins. Furthermore, within 24 hr after removal of RPE cells from the coculture, the T cells regained normal proliferation and IL2R chain expression. Similar effects have been found of alveolar
F IG . 7. T cells' expression of IL2R-a, -b and -g chain after incubation alone (T) or with ARPE-19 or HFRPE cells for 72 hr. The T cells' expression of the IL2R-a, -b and -g chain was measured by use of anti-IL2R-a, -b and -g chain Abs and anti-CD3 Ab followed by ¯ow cytometry. Illustrated is the average mean expression (n 7±10) and SEM of the IL2R chains of CD3 positive cells. Statistical analysis with paired t-test indicated signi®cant difference (P 5 0.05) between expression of IL2R chain of T cells cultured alone compared to T cells cultured with RPE cells where marked with the symbol (*).
634
F IG . 8. Ability of T cells to regain proliferation (A) and `normal' expression pattern of IL2R-a, -b and -g chain (B) after preincubation with ARPE-19 cell followed by incubation alone in fresh culture medium. T cells were removed from coculture after 48 hr and incubated alone in fresh media. The proliferation was measured after 6 and 24 hr incubation by addition of 3H-thymidine (1 mCi per well 1) and the expression of the IL2R-a, -b and -g chain was measured by ¯ow cytometry. The proliferation results depicted are the mean and SEM of four replicates from a representative experiment of three. The ¯ow cytometric results depicted are means and SEM of mean ¯uorescence intensity of (MFI) of T cells precultured with RPE cells divided with MFI of T cells precultured alone (n 3±5). Statistical analysis with paired t-test indicated signi®cant difference (P 5 0.05) between level of proliferation and expression of IL2R chain of T cells precultured alone compared to T cells precultured with RPE cells where marked with the symbol (*).
types of epithelial cells, types of T cells and culture types may be decisive for the outcome of the interaction between T cells and RPE cells. Here it is also of interest that the apoptotic process may vary in different experiments. In a previous study, the authors found that apoptosis most likely was due to suicide or fratricide between T cells, induced by Fas±FasL interaction, paralleling for example melanoma speci®c T cell apoptosis after MHC restricted recognition of FasL negative tumor cells as reported by (Zaks et al.,
C. G. KáSTEL ET AL.
1999; Restifo, 2000). However, another recent study indicates that apoptosis may not be due to Fas±FasL interaction (Rezai et al., 1999). Of special interest are the results of several studies from one research group (Rezai et al., 1999; FarrokhSiar et al., 1999a, 2000) which show that the RPE cell induced inhibiting factor is an uncharacterized soluble one, whereas the present authors ®nd that cell to cell contact is required for signi®cant inactivation of the T cellsÐeven if some soluble factor may be of importance. Thus, a decreased proliferation of T cells was found cultured in inserts, but not signi®cant. Since the same culture types ± including transwell cultures ± are used in these studies, this essential difference is dif®cult to explain, but is not likely to be due to the type of RPE cells used, since the present authors have used both foetal and adult RPE cells (Jùrgensen et al., 1998) and (present study) with comparable results. However, besides differences in handling of cells and culture systems, the differences between the studies here and those of Farrokh-Siar et al. may be due to that the authors use `stable' activated T cells, whereas they investigate the induction period of T cell activation. It is found here that RPE cells induce decreased proliferation of MF and Se-Ax cell lines both in presence and in absence of exogenous IL2. Thus, exogenous IL2 is not required for the depression of proliferation. However, the use of exogenous IL2 is necessary to avoid apoptosis of activated periphera T cells grown alone in culture (RoÈpke et al., 1996) as also shown in Fig. 6, where essentially no 3H-TdR incorporation is seen in T cells grown in the absence of IL2. Accordingly the authors have used IL2 in the experiments, both when T cells are cultured alone and when they are cocultured with RPE cells. Presence of IL2 in the cultures is not the reason for the absence of apoptosis, since IL2 was used in all cultures in an earlier study in which high percentages of apoptosis were seen after coculture of T cells and RPE cells (Jùrgensen et al., 1998). IL2 is an essential growth factor for the T cells, and the IL2 signal is given through the IL2R, which is a complex composed by the three chains, IL2R-a, -b and -g. IL2R functions as low af®nity receptor, when constituted by the IL2R-b and -g chains (Kd 10 9 M), and as high af®nity receptor, when constituted by all three chains IL2R-a, -b and -g (Kd 10 11 M) (Minami et al., 1993). The authors analysed the state of the T cells when cultured alone or in the presence of ARPE-19 or HFRPE cells with regard to their expression of IL2R and their capacity to receive an IL2 signal. The results showed a change in expression of IL2R-b and g chains after coculturing with ARPE-19 as well as HFRPE cells. In addition, the ARPE-19 cells seemed to have a tendency to increase the T cells' IL2R-a chain expression, an effect that was not seen by the HFRPE cells. This effect is, therefore, not thought to be of signi®cant importance for a plausible IL2R linked inhibition.
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
The change, in expression of IL2R-g chain after coculture, was very obvious. This chain is shared by all the known T cell growth factor (TCGF) receptors i.e. IL2R, IL4R, IL7R, IL9R and IL15R and is therefore a key feature for the T cells' potential for proliferation and survival (Akbar et al., 1996; Karnitz and Abraham, 1996; Nakajima et al., 1997; Malek, Porter and He, 1999; and Van Parijs et al., 1999). The IL2R-g chain is also essential for T cell activation and a recent publication has shown that allograft survival is prolonged in the presence of blocking antiIL2R-g chain Ab (Li et al., 2000). Caspi et al. (1987) investigated the inhibitory effect of rat MuÈller cells on T cells and found a decreased IL2 production and through binding studies, a reduced expression of high af®nity IL2R of cocultured T cells. These data could be explained by the ®ndings of changed expression of IL2R chains published here and extent their ®nding to include the RPE cells. A reduced expression of IL2R-b and -g chains could limit the cells' ability to form high af®nity IL2R complexes. It is hypothesized therefore that RPE cells may induce their inhibitory effect on cycling T cells by modulation of the T cell IL2R and that this may be a non Ag-speci®c way of human RPE cells to confer immune privilege in the posterior segment of the eye. So far, no factor has yet been assigned to the RPE cell induced effect on IL2R, but recently it was shown that PGE2 is able to inhibit the expression of IL2R-g chain of human NK cells (Joshi et al., 2001). Additionally, PGE2 has been found to participate in the Rat RPE cell induced inhibition of T cells, in studies by Liversidge et al. (1993). The present authors investigated the effect of the PGE2 antagonist; indomethacin, in this model. However, no effect was found of indomethacin on the RPE cell induced inhibition or the IL2R modulation of T cells. This indicates that another mechanism is involved. The present studies show that a signi®cant part of the inhibition is dependent on direct contact between the two cell types. Additionally, no difference was found between T cells cultured in the upper chamber of a transwell system with ARPE-19 cells in the bottom chamber and T cells cultured in the upper chamber of a transwell system with both ARPE-19 cells and T cells in the bottom chamber. These data suggest that the ARPE-19 cell induced inhibition of T cells initially depend on a cell contact-dependent mechanism or a soluble factor present in a low concentration or vulnerable to quick degradation. The involvement of the surface molecules CD54, CD58, CD40 and TCR was tested and none of these molecules seemed to be signi®cantly involved in the inhibitory process. The supernatants from cocultures of human RPE and T cells have previously been analysed for a panel of soluble molecules (Liversidge et al., 1998; Rezai et al., 1999; Farrokh-Siar et al., 1999a; Yoshida et al., 2000b). So far, none of these soluble factors have been ascribed to the inhibitory process in activated T cells induced by human RPE cells.
635
In conclusion, several studies indicate that RPE cell±T cell interactions may be of importance in the maintenance of the immune privilege in the posterior part of the eye. However, these studies are all in vitro studies, and even in this ± compared to in vivo ± simple system, the essential factors in RPE cell induced inhibition of T cell proliferation is not pinpointed. Both cell surface molecules and soluble factors may be operative. Further, the inhibitory effect of RPE cells on T cells is limited to activation of these cells, since RPE cells are able to present antigen to ± and induce proliferation of ± resting T cells (Jùrgensen et al., 2001). Thus, more investigations are needed for clari®cation of the intracellular pathways through which inhibition of activated T cells is achieved. The present demonstration of RPE cell induced changes in T cells' IL2R expression may give clues where as to focus these investigations. Acknowledgements The authors thank Dr Tongalp H. Tezel for supplying the HFRPE cells and Ane Marie Rulykke, Helene Rasmussen and Evelyn Kury for skilful technical assistance. This work was supported by grants from The Danish Eye Research Foundation, Vñrn om Synet, The Danish Medical Council, Novo Nordisk and The Beckett Foundation.
References Akbar, A. N., Borthwick, N. J., Wickremasinghe, R. G., Panayoitidis, P., Pilling, D., Bo®ll, M., Krajewski, S., Reed, J. C. and Salmon, M. (1996). Interleukin-2 receptor common gamma-chain signaling cytokines regulate activated T cell apoptosis in response to growth factor withdrawal: selective induction of anti-apoptotic (bcl-2, bcl-xL) but not pro- apoptotic (bax, bcl-xS) gene expression. Eur. J. Immunol. 26, 294±9. Caspi, R. R., Roberge, F. G. and Nussenblatt, R. B. (1987). Organ-resident, nonlymphoid cells suppress proliferation of autoimmune T-helper lymphocytes. Science 237, 1029±32. Cousins, S. W., McCabe, M. M., Danielpour, D. and Streilein, J. W. (1991). Identi®cation of transforming growth factor-beta as an immunosuppressive factor in aqueous humor. Invest. Ophthalmol. Vis. Sci. 32, 2201±11. Elner, S. G., Elner, V. M., Pavilack, M. A., Todd, R. F., III, Mayo-Bond, L., Franklin, W. A., Strieter, R. M., Kunkel, S. L. and Huber, A. R. (1992). Modulation and function of intercellular adhesion molecule-1 (CD54) on human retinal pigment epithelial cells. Lab. Invest. 66, 200±11. Farrokh-Siar, L., Rezai, K. A., Semnani, R. T., Patel, S. C., Ernest, J. T., Peterson, E. J., Koretzky, G. A. and van Seventer, G. A. (1999a). Human fetal retinal pigment epithelial cells induce apoptosis in the T-cell line Jurkat. Invest. Ophthalmol. Vis. Sci. 40, 1503±11. Farrokh-Siar, L., Rezai, K. A., Semnani, R. T., Patel, S. C., Ernest, J. T. and van Seventer, G. A. (1999b). Human fetal retinal epithelium suppresses the activation of CD4() and CD8() T-cells. Graefes Arch. Clin. Exp. Ophthalmol. 237, 934±9. Farrokh-Siar, L., Rezai, K. A., Palmer, E. M., Patel, S. C., Ernest, J. T. and van Seventer, G. A. (2000). Human fetal retinal pigment epithelium-induced cell cycle arrest, loss of mitochondrial membrane potential and apoptosis. Invest. Ophthalmol. Vis. Sci. 41, 3991±8.
636
Gospodarowicz, D. and Ill, C. (1980). Extracellular matrix and control of proliferation of vascular endothelial cells. J. Clin. Invest. 65, 1351±64. Grif®th, T. S., Brunner, T., Fletcher, S. M., Green, D. R. and Ferguson, T. A. (1995). Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270, 1189±92. Joshi, P. C., Zhou, X., Cuchens, M. and Jones, Q. (2001). Prostaglandin E2 suppressed IL-15-mediated human NK cell function through down-regulation of common gamma-chain. J. Immunol. 166, 885±91. Jùrgensen, A., Ropke, C., Nielsen, M., Madsen, H., Svejgaard, A. and Odum, N. (1996). Human thymic epithelial cells express functional HLA-DP molecules. Tissue Antigens 47, 300±6. Jùrgensen, A., Wiencke, A. K., la Cour, M., Kñstel, C. G., Madsen, H. O., Hamann, S., Lui, G. M., Scher®g, E., Prause, J. U., Svejgaard, A., Odum, N., Nissen, M. H. and Ropke, C. (1998). Human retinal pigment epithelial cell-induced apoptosis in activated T cells. Invest. Ophthalmol. Vis. Sci. 39, 1590±9. Jùrgensen, A., Junker, N., Kñstel, C. G., Liang, Y., Wiencke, A., Lui, G. M., Odum, N., Nissen, M. H. and Ropke, C. (2001). Superantigen presentation by human retinal pigment epithelial cells to T cells is dependent on CD2CD58 and CD18-CD54 molecule interaction. Exp. Eye Res. 73, 723±33. Kaiser, C. J., Ksander, B. R. and Streilein, J. W. (1989). Inhibition of lymphocyte proliferation by aqueous humor. Reg. Immunol. 2, 42±9. Kaltoft, K., Bisballe, S., Rasmussen, H. F., ThestrupPedersen, K., Thomsen, K. and Sterry, W. (1987). A continuous T-cell line from a patient with Sezary syndrome. Arch. Dermatol. Res. 279, 293±8. Kaltoft, K., Bisballe, S., Dyrberg, T., Boel, E., Rasmussen, P. B. and Thestrup-Pedersen, K. (1992). Establishment of two continuous T-cell strains from a single plaque of a patient with mycosis fungoides. In Vitro Cell Dev. Biol. 28A, 161±7. Kaplan, H. J., Leibole, M. A., Tezel, T. and Ferguson, T. A. (1999). Fas ligand (CD95 ligand) controls angiogenesis beneath the retina. Nat. Med. 5, 292±7. Karnitz, L. M. and Abraham, R. T. (1996). Interleukin-2 receptor signaling mechanisms. Adv. Immunol. 61, 147±99. Kñstel, C. G., Madsen, H., Prause, J. U., Jùrgensen, A., la Cour, M., Lui, G. M., Odum, N., Nissen, M. H. and Ropke, C. (2001). Lack of FasL Expression in Cultured Human Retinal Pigment Epithelial Cells. Exp. Clin. Immunogenet. 18, 34±41. Lee, I. H., Li, W. P., Hisert, K. B. and Ivashkiv, L. B. (1999). Inhibition of interleukin 2 signaling and signal transducer and activator of transcription (STAT)5 activation during T cell receptor-mediated feedback inhibition of T cell expansion. J. Exp. Med. 190, 1263±74. Li, X. C., Ima, A., Li, Y., Zheng, X. X., Malek, T. R. and Strom, T. B. (2000). Blocking the common gammachain of cytokine receptors induces T cell apoptosis and long-term islet allograft survival. J. Immunol. 164, 1193±9. Lissy, N. A., VanDyk, L. F., Becker-Hapak, M., VoceroAkbani, A., Mendler, J. H. and Dowdy, S. F. (1998). TCR antigen-induced cell death occurs from a late G1 phase cell cycle check point. Immunity 8, 57±65. Liversidge, J., Sewell, H. F., Thomson, A. W. and Forrester, J. V. (1988). Lymphokine-induced MHC class II antigen expression on cultured retinal pigment epithelial cells and the in¯uence of cyclosporin A. Immunology 63, 313±7.
C. G. KáSTEL ET AL.
Liversidge, J., McKay, D., Mullen, G. and Forrester, J. V. (1993). Retinal pigment epithelial cells modulate lymphocyte function at the blood-retina barrier by autocrine PGE2 and membrane-bound mechanisms. Cell Immunol. 149, 315±30. Malek, T. R., Porter, B. O. and He, Y. W. (1999). Multiple gamma c-dependent cytokines regulate T cell development. Immunol. Today 20, 71±6. Minami, Y., Kono, T., Miyazaki, T. and Taniguchi, T. (1993). The IL-2 receptor complex: its structure, function, and target genes. Ann. Rev. Immunol. 11, 245±68. Morgan, B. P. and Meri, S. (1994). Membrane proteins that protect against complement lysis. Springer Semin. Immunopathol. 15, 369±96. Nakajima, H., Shores, E. W., Noguchi, M. and Leonard, W. J. (1997). The common cytokine receptor gamma chain plays an essential role in regulating lymphoid homeostasis. J. Exp. Med. 185, 189±95. Platts, K. E., Benson, M. T., Rennie, I. G., Sharrard, R. M. and Rees, R. C. (1995). Cytokine modulation of adhesion molecule expression on human retinal pigment epithelial cells. Invest. Ophthalmol. Vis. Sci. 36, 2262±9. Rabinovitch, P. S., Torres, R. M. and Engel, D. (1986). Simultaneous cell cycle analysis and two-color surface immuno¯uorescence using 7-amino-actnimycin D and single laser excitation: applications to study of cell activation and cell cycle of murine LT-1 B cells. J. Immunol. 136, 2769±73. Radvanyi, L. G., Mills, G. B. and Miller, R. G. (1993). Religation of the T cell receptor after primary activation of mature T cells inhibits proliferation and induces apoptotic cell death. J. Immunol. 150, 5704±15. Restifo, N. P. (2000). Not so Fas: Re-evaluating the mechanisms of immune privilege and tumor escape. Nat. Med. 6, 493±5. Rezai, K. A., Semnani, R. T., Farrokh-Siar, L., Hamann, K. J., Patel, S. C., Ernest, J. T. and vanSeventer, G. A. (1999). Human fetal retinal pigment epithelial cells induce apoptosis in allogenic T-cells in a Fas ligand and PGE2 independent pathway. Curr. Eye Res. 18, 430±9. RoÈpke, C., Gladstone, P., Nielsen, M., Borregaard, N., Ledbetter, J. A., Svejgaard, A. and édum, N. (1996). Apoptosis following interleukin-2 withdrawal from T cells: Evidence for a regulatory role of CD18 (b2-integrin) molecules. Tissue Antigens 48, 127±35. Streilein, J. W. (1993). Immune privilege as the result of local tissue barriers and immunosuppressive microenvironments. Curr. Opin. Immunol. 5, 428±32. Streilein, J. W. and Stein-Streilein, J. (2000). Does innate immune privilege exist?. J. Leukoc Biol. 67, 479±87. Strickland, D., Kees, U. R. and Holt, P. G. (1996). Regulation of T-cell activation in the lung: alveolar macrophages induce reversible T-cell anergy in vitro associated with inhibition of interleukin-2 receptor signal transduction. Immunology 87, 250±8. Taylor, A. W., Streilein, J. W. and Cousins, S. W. (1992). Identi®cation of alpha-melanocyte stimulating hormone as a potential immunosuppressive factor in aqueous humor. Curr. Eye Res. 11, 1199±206. Taylor, A. W., Streilein, J. W. and Cousins, S. W. (1994). Immunoreactive vasoactive intestinal peptide contributes to the immunosuppressive activity of normal aqueous humor. J. Immunol. 153, 1080±6. Taylor, A. W., Yee, D. G. and Streilein, J. W. (1998). Suppression of nitric oxide generated by in¯ammatory macrophages by calcitonin gene-related peptide in aqueous humor. Invest. Ophthalmol. Vis. Sci. 39, 1372±8.
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
Tezel, T. H., Del Priore, L. V. and Kaplan, H. J. (1997). Harvest and storage of adult human retinal pigment epithelial sheets. Curr. Eye Res. 16, 802±9. Upham, J. W., Strickland, D. H., Robinson, B. W. and Holt, P. G. (1997). Selective inhibition of T cell proliferation but not expression of effector function by human alveolar macrophages. Thorax 52, 786±95. Van Parijs, L., Refaeli, Y., Lord, J. D., Nelson, B. H., Abbas, A. K. and Baltimore, D. (1999). Uncoupling IL-2 signals that regulate T cell proliferation, survival, and Fasmediated activation induced cell death. Immunity 11, 281±8. Willermain, F., Caspers-Velu, L., Baudson, N., Dubois, C., Hamdane, M., Willems, F., Velu, T. and Bruyns, C. (2000). Role and expression of CD40 on human retinal pigment epithelial cells. Invest. Ophthalmol. Vis. Sci. 41, 3485±91. Yoshida, M., Takeuchi, M. and Streilein, J. W. (2000a). Participation of pigment epithelium of iris and ciliary
637
body in ocular immune privilege. I. Inhibition of T-cell activation in vitro by direct cell-to-cell contact. Invest. Ophthalmol. Vis. Sci. 41, 811±21. Yoshida, M., Kezuka, T. and Streilein, J. W. (2000b). Participation of pigment epithelium of iris and ciliary body in ocular immune privilege. 2. Generation of TGFbeta-producing regulatory T cells. Invest. Ophthalmol. Vis. Sci. 41, 3862±70. Zaks, T. Z., Chappell, D. B., Rosenberg, S. A. and Restifo, N. P. (1999). Fas-mediated suicide of tumor-reactive T cells following activation by speci®c tumor: selective rescue by caspase inhibition. J. Immunol. 162, 3273±9. Zavazava, N., Halene, M., Westphal, E., Nolle, B., Duncker, G., Eckstein, E., Harpprecht, J. and Muller-Ruchholtz, W. (1991). Expression of MHC class I and II molecules by cadaver retinal pigment epithelium cells: optimization of post-mortem HLA typing. Clin. Exp. Immunol. 84, 163±6.
Human Retinal Pigment Epithelial Cells Inhibit Proliferation and IL2R Expression of Activated T Cells C H A R LOT T E G . K á S T E L a, A N N E T T E Jé R G EN S E N a, M ETT E N IEL S EN b, K A R S TE N W. ER IKS EN b, N I E L S é D U M b, M O G EN S H O LS T N IS S EN a A N D CA R S T EN R OÈ P K E a* a
Institute of Medical Anatomy Section A, The Panum Institute, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen N, Denmark and bInstitute of Medical Microbiology and Immunology, The Panum Institute, University of Copenhagen, Denmark. (Received St. Louis 30 October 2001 and accepted in revised form 5 February 2002) The purpose of this study was to characterize the effects of human retinal pigment epithelial (RPE) cells on activated T cells. Activated T cells were cocultured with adult and foetal human RPE cells whereafter apoptosis and proliferation were determined by ¯ow cytometry and 3H-Thymidine incorporation assay, respectively. T cells and RPE cells were cultured directly together or in a transwell system for determination of the effect of cell contact. The importance of cell surface molecules was examined by application of a panel of blocking antibodies (CD2, CD18, CD40, CD40L, CD54, CD58) in addition to use of TCR negative T cell lines. The expression of IL2R-a -b and -g chains of activated T cells was analysed by ¯ow cytometry after incubation of T cells alone or with RPE cells. Human RPE cells were found to inhibit the proliferation of activated T cells by a cell contact-dependent mechanism. The RPE cells inhibitory abilities were not affected by blocking of any of the tested surface molecules. The inhibition of the T cells' proliferation correlates with a decreased expression of IL2R-b and -g chains. The T cells regain their ability to proliferate and increase their IL2R-b and -g chain expression within 24 hr after removal from the coculture. It is concluded that the cultured human adult and foetal RPE cells inhibit the proliferation of activated T cells by a process that does not involve apoptosis. It depends on cell contact but the involved surface molecules were not revealed. The proliferation inhibition correlates with a modulation of the T cells' expression of IL2R, and is reversible. # 2002 Elsevier Science Ltd. Key words: retinal pigment epithelial; T cells; proliferation; IL2R-g; inhibition.
1. Introduction The eye is known as an immune privileged organ and is characterized by allowing prolonged survival of foreign tissue grafts. Several mechanisms have been suggested to support this privilege, starting with the physical isolation of the eye from the systemic blood circulation due to the presence of blood: tissue barriers, as well as relative absence of lymphatic drainage from the eye (Streilein, 1993). Additionally, molecular features such as surface expression of FasL by ocular cells (cornea, retina, iris and ciliary body) (Grif®th et al., 1995; Kaplan et al., 1999; Kñstel et al., 2001) and membrane-bound inhibitors of complement activation (Morgan and Meri, 1994), as well as the presence of immune suppressive soluble factors in intraocular ¯uids, especially aqueous humor, have been proposed to be involved in the phenomenon (Kaiser et al., 1989; Cousins et al., 1991; Taylor, Streilein and Cousins, 1992, 1994; Taylor et al., 1998). There is, however, still many unanswered questions with regard to how the immune privilege of the eye is obtained (Streilein and Stein-Streilein, 2000). A number of recent studies have shown that retinal pigment epithelial (RPE) cells cocultured with activated T cells inhibit the T cell proliferation (Liversidge et al., * Author for correspondence. E-mail: [email protected]
0014-4835/02/ $35.00/0
1993; Jùrgensen et al., 1998; Farrokh-Siar et al., 1999a,b, 2000; Rezai et al., 1999; Willermain et al., 2000), but the mechanisms behind this inhibition are to this date still largely unknown. Human foetal RPE cellsÐnot expressing FasL themselves (Kñstel et al., 2001), have been found to induce apoptosis in both FasL independent and dependent manners (Rezai et al., 1999; Farrokh-Siar et al., 1999a, 2000; and Jùrgensen et al., 1998), whereas adult rat RPE cells as well as murine iris and ciliary body pigment epithelial cells, which resemble RPE cells, have been found to inhibit the proliferation via unknown membrane bound factors as well as soluble factors such as prostaglandin E2(PGE2) and TGF-b (Liversidge et al., 1993; Yoshida, Takeuchi and Streilein, 2000a; Yoshida et al., 2000b). In an attempt to understand the mechanisms involved in the RPE cell induced inhibition of T cells, the phenotype of the activated T cells has been analysed with regard to their expression of IL2R-a, -b and -g chain, and the long-term effect of the T cells' encounter with the RPE cells. 2. Materials and Methods Antibodies and Other Reagents Mouse IgG1 and IgG2a, mouse monoclonal antihuman IL2Rb-PE Ab and CD40L Ab as well as rat # 2002 Elsevier Science Ltd.
628
monoclonal anti-IL2Rg-PE Ab were from Pharmingen (San Diego, CA, U.S.A.). Mouse monoclonal antihuman CD3-FITC Ab, CD3-PE Ab and CD25-FITC Ab were from Dako (Glostrup, Denmark). Mouse antihuman CD40 Ab was from Mabtech (Nacka, Sweden). Mouse anti-human CD18, CD54 and CD58 were from Immunotech (Marseilles, France). Retinal Pigment Epithelial Cell Culture The human adult RPE cell line ARPE-19 was purchased from ATTC (Manassas, VA, U.S.A.) and human foetal RPE (HFRPE) cells were a kind gift from Dr Tongalp H. Tezel. The HFRPE cells were isolated from a 17 week old male fetus, harvested within 6 hr of an elective abortion. Method of securing human tissues complied with the Declaration of Helsinki. Harvesting and culturing technique is described in detail in Tezel et al., (1997). The RPE cells were thawed and seeded on tissueculture plates (Falcon; Becton Dickenson Labware, Bedford, MA, U.S.A.) at a density of 8000 cells cm 2. ARPE-19 cells were seeded directly on plastic where as the HFRPE cells were seeded on tissue-culture plates precoated with extracellular matrix prepared as described by Gospodarowicz and Ill (1980). After reaching con¯uence, they were transferred at a split ratio of 1 : 5 and seeded onto new tissue-culture plates, and repeatedly transferred at a split ratio of 1 : 8. When the cells had acquired a hexagonal shape and pigmentation after 2±4 weeks, they were applied in coculture experiments. ARPE-19 cells were maintained in Dulbecco's modi®ed Eagle's medium (H16; Gibco, Life Technologies, Paisley, Scotland, U.K.) supplemented with 15 % foetal calf serum (FCS; HyClone Laboratories, Logan, UT, U.S.A.) 300 mg ml 1 glutamine 50 mg ml 1 gentamycin and 2.5 mg ml 1 amphotericin B (Fungizone; Gibco, Life Technologies, Paisley, Scotland, U.K.). HFRPE cells were maintained in similar medium further supplemented with 1 ng ml 1 bFGF (Sigma St. Louis, MO, U.S.A.). Medium was changed every other day and the cells were regularly tested for mycoplasma. The ARPE19 cells used in this study were passage 23 or 26 and the HFRPE cells passage 3 or 4. The ARPE-19 cell line has been found to have comparable expression with human foetal RPE (p6) cells of the following surface molecules: MHC class I and II, CD40, CD48, CD54, CD58, CD68, CD80, CD86, CD95, CD95L (Jùrgensen et al., 2001; Kñstel et al., 2001). Peripheral Blood Lymphocytes Peripheral blood mononuclear cells (PBMC) were separated from buffycoats (The Blood Bank, National University Hospital, Copenhagen, Denmark) by lymphoprep (Nycomed, Oslo, Norway) density-gradient. The PBMCs (106 cells ml 1) were either stimulated with 4 mg ml 1 phytohemagglutinin (PHA, Welcome-
C. G. KáSTEL ET AL.
Difco, Detroit, MI, U.S.A.) in RPMI 1640 supplemented with 10 % FCS (Gibco, Life Technologies, Paisley, Scotland, U.K.), 200 U ml 1 soluble recombinant IL2 (Chiron, CA, U.S.A.), 2 mM L-glutamine, 100 mg ml 1 penicillin and 100 mg ml 1 streptomycin (The Panum Institute, Copenhagen, Denmark) and immediately cocultured with RPE cells or prestimulated with 4 mg ml 1 phytohemagglutinin, pelleted and resuspended in culture medium (1 106 cells ml 1) with 10 % FCS and 200 U ml 1 IL2 after 4 days and then after another 4 days applied to coculture experiments. A new blood donor was used for each repetitive experiment. T-cell Lines The T-cell lines, MF and Se-Ax, are lymphoid tumor cell lines established from a plaque biopsy specimen of patients with mycosis fungoides (MF) or the SeÂzary syndrome as described in detail elsewhere (Kaltoft et al., 1987, 1992). MF and Se-Ax cells were cultured in RPMI 1640 (Gibco, Life Technologies, Paisley, Scotland, U.K.) supplemented with 10 % FCS or 10 % human serum (HS), 2 mM L-glutamine, 100 mg ml 1 penicillin and 100 mg ml 1 streptomycin. The media of the Se-Ax cell line were additionaly supplemented with 1000 U ml 1 IL2 (Chiron B.V; Amsterdam, The Netherlands). Coculture of T Cells and RPE Cells RPE cells were trypsinised, resuspended in DMEM : H16 culture medium with or without bFGF and seeded onto 96 well (5 104 cells per well) or 24 well (2 105 cells per well) ¯atbottom culture plates (NUNC A/S, Kamstrup, Denmark) and incubated overnight in 10 % CO2, 378C, allowing the cells to stick to the bottom of the wells. The RPE cells were then washed in PBS (2) and T cells were added either directly or in inserts with a pore size of 0.2 mm (NUNC). All cocultures were maintained in RPMI 1640 supplemented with 10 % FCS with 200 U ml 1 soluble recombinant IL2, unless otherwise noted and incubated in 5 % CO2 at 378C. Cocultured cells were harvested for ¯ow cytometric analysis by pipetting both T cells and RPE cells with a 1 ml pipette. Trypsination was avoided due to the relatively short culturing of the RPE cells in the actual culture plate. The wells were checked after harvesting by light microscopy to make sure that all cells had been harvested. In order to investigate the long-term effect of ARPE19 and HFRPE cells on T cells, the T cells were carefully removed via pipette from coculture, avoiding the RPE cells. It was not possible to get all the T cells out of the coculture due to adhesion between the two types of cells, however, a recovery percentage was estimated to be no less than 75 % (number of T cells recovered from coculture in compared to number of T cells recovered from single culture).
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
Flow Cytometry The cell cycle of peripheral T cells after coculture with RPE cells was analysed by surface staining of CD3 followed by intracellular DNA staining with 7-amino-actinomycin D (7-AAD; Sigma St. Louis, MO, U.S.A.) (Rabinovitch, Torres and Engel, 1986), and ¯ow cytometry analysis. The cells (3±5 105 cells per sample) were washed in PBS stained for CD3 surface expression by adding anti-CD3-FITC Ab or isotype control Ab followed by incubation on ice for 30 min. Afterwards the cells were washed in PBS and then in PBS/0.05 % saponin (Sigma St. Louis, MO, U.S.A.), resuspended in 4 mg ml 1 7-AAD in PBS/ 0.05 % saponin and left to incubate in the dark at room temperature for 30 min before analysis on a FACSCalibur apparatus (Becton Dickinson, San Jose, CA, U.S.A.). The surface expression of IL2R-a, -b and -g chain of CD3 positive cells after culturing alone or with RPE cells was measured by use of anti-CD3-FITC or antiCD3-PE and anti-IL2R-a-FITC, anti-IL2R-b-PE or anti-IL2R-g-PE Ab as well as isotype control antibodies followed by ¯ow cytometry. When analyzing, a gate of live CD3 positive cells was applied, and the percentage of cells in the different cell cycle phases or the mean ¯uorescence intensity (MFI) of the IL2R chains of these cells was measured and used for statistical calculations. 3
H-Thymidine Incorporation Assay
The proliferation of T cells in culture alone or in coculture with RPE cells was measured by incorporation of 3H-Thymidine (1 mCi per well, 1 Ci 37 Gbp) (N.E.N., Boston, MA, U.S.A.) during the last 6 hr of culture. The incorporation of 3H-Thymidine in the RPE cells was measured as a control. It was found in all experiments that the RPE cells did not signi®cantly contribute to the incorporation of 3H-Thymidine in the coculture experiments. All assays were performed in 96 well culture plates or transwell systems of 96 well culture plates (NUNC). 3. Results RPE Cells Induce S and G2/M Phase Decreases but not Apoptosis in Cycling T cells To evaluate the effects of adult RPE cells on activated T cells, T cells were set up in IL2 containing cultures either alone or in the presence of ARPE-19 cells, which were added to the chambers 24 hr before coculture to allow the cells to adhere to the bottom. Fig. 1 shows analysis of cell cycle of T cells after culturing alone or with RPE cells via labelling with CD3 and 7-AAD for selection of T cells and for staining the DNA, respectively. It is seen that the percentages of CD3 positive T cells in S G2/M phases decrease signi®cantly in cocultures compared
629
with percentages in cultures of T cells alone. Similar results were obtained with cocultures of T cells and adult RPE (p0) cells as well as foetal RPE cells (p6) (data not shown). The ®gure further shows that the decrease in cycling cells is not accompanied by an increase in cells with subnormal DNA i.e. apoptosis. This was paralleled by ¯ow cytometric analysis of the early apoptosis marker Annexin V and the antiapoptotic proteins Bcl-2 and Bcl-xL as well as the proapoptotic protein Bax. No signi®cant difference was found in the expression of these proteins between T cells cultured alone or with ARPE-19 cells for 48 hr (data not shown). Since the ¯ow cytometric analysis by the use of 7-AAD at harvest depicts the collected events from the entire culture period, the authors used 3H-Thymidine labelling of cells to evaluate the DNA synthetic activity in selected time periods. Fig. 2(A) shows an experiment in which T cells were cultured for up to 72 hr with or without ARPE-19 cells. It is clearly seen that T cells maintain comparable DNA synthesis in both culture types at 18±24 hr, whereas a very pronounced decrease in DNA synthesis is seen in cocultures at the 42±48 and 66±72 hr time periods. Similar results were obtained with g-irradiated, nonproliferating RPE cells, data not shown. Fig. 2(B) illustrates the effect of ARPE-19 cells by the use of varying ratios of ARPE-19 cells and T cells from (1 : 1) to (0.25 : 1) for 48 hr. The ARPE-19 cell induced inhibition was ratio-dependent, and absent at the ratio (0.25 : 1) in these culture chambers. In addition, it was found that PHA and IL2 induced proliferation of freshly isolated peripheral T cells is inhibited in the presence of ARPE-19 cell (data not shown). The Inhibition of T Cell Proliferation is Cell Contact-dependent The role of cell contact in the ARPE-19 induced T cell inhibition was analysed in a transwell system. The ARPE-19 cells were placed in the lower chamber and separated by a membrane from the upper chamber containing T cells. By having the ARPE-19 cells in the lower chamber instead of in the upper chamber, `clotting' at the membrane by the epithelial cells adhering to the membrane was avoided. The membrane has a pore size of 0.2 mm and does not permit the transmigration of cells, but soluble factors can easily diffuse through the membrane. Fig. 3 shows both the proliferation of T cells cultured alone and the proliferation of cocultured T cells. T cells were either cocultured directly on top of ARPE-19 cells or in the upper chamber in a transwell system with ARPE-19 cells in the bottom chamber, or with T cells in the upper chamber and with both ARPE-19 cells and T cells in the bottom chamber. As can be seen from the ®gure, the ARPE-19 cell induced inhibition of proliferation of T cells is clearly reduced by the
630
C. G. KáSTEL ET AL.
F IG . 1. Flow cytometric analysis of the cell cycle of T cells after culturing alone (T) or with ARPE-19 cells (T R) for 24, 48 or 72 hr. The T cells were selected for 7-AAD analysis by gating on the CD3 labelled cells. Left bars indicate percentage of apoptosis and right bars the percentage of cells in S G2/M phase compared to total amount of cells. The depicted results are representative for three separate experiments.
absence of cell contact in the transwell culture system. Additionally there was no difference between the effect of T cells cultured in the upper chamber with ARPE-19 cells in the bottom chamber, compared to T cells cultured in the upper chamber with both ARPE19 cells and T cells in the bottom chamber. The Inhibition of T Cell Proliferation is Independent of the Surface Molecules CD2, CD18, CD40, CD40L, CD54, CD58 and of TCR and MHC Molecules The adhesion molecules CD54 and CD58 are known to in¯uence the interaction between T cells and other cell types including epithelial cells (Jùrgensen et al., 1996, 2001). The authors found both these molecules to be expressed on human cultured adult and foetal RPE cells (data not shown). Further, human RPE cells were recently found to also
express the costimulatory surface molecule CD40 in response to IFN-g stimulation (Willermain et al., 2000). Blocking antibodies against these molecules or their counterparts CD18 (LFA-1), CD2 (LFA-3) and CD40L were added to cocultures of T cells and ARPE19 cells. It was found that blocking of either of these molecules had no signi®cant effect on the ability of ARPE-19 cells to inhibit T-cells, see Fig. 4. Further, pre-treatment of ARPE-19 cells with IFN-g, which in addition to up-regulation of CD40 up-regulate the expression of CD54, MHCI and II (Liversidge et al., 1988; Zavazava et al., 1991; Elner et al., 1992; Platts et al., 1995), did not lead to changes in the effects of ARPE-19 cells on T cells (data not shown). This indicates that these molecules, including MHC molecules, are of minor importance in the ARPE-19 cell±T cell interaction. This was corroborated by the use of the T cell lines MF and Se-Ax that do not
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
631
F IG . 2. Proliferation of activated T cells after either culturing alone (T) or with ARPE-19 cells (T R) (1 : 1) for 24, 48 or 72 hr (A), or after incubation with a varying number of ARPE-19 cells for 48 hr (B). ARPE-19 cells were seeded onto a 96 well culture plate and activated T cells (5 105) were added the following day. Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of four replicates from a representative experiment of three. Statistical analysis with unpaired t-test showed that ARPE-19 cells signi®cantly inhibited the T cells at 48 and 72 hr (A) and at the ratios (1 : 1) and (0.5 : 1) (B) *P 5 0.05.
express functional TCR. ARPE-19 cells induced signi®cant decreases in DNA synthesis in both cell lines in the presence or absence of IL2 (see Fig. 5), supporting that TCR±MHC class I/II interaction is not essential for the observed suppression of the DNA synthesis in activated T cells. As can be seen from the ®gure, the presence of IL2 did not change the effect of ARPE-19 cells on these cell lines.
T Cell Proliferation: Dependence of IL2 Since the T cells are maintained in cell cycle by addition of IL2 to the cultures, the ®ndings of decreased DNA synthesis in T cells in cocultures with ARPE-19 cells may be caused by an inef®cient utilization of IL2 by the T cells in the cocultures. It was therefore tested if the proliferation inhibition could be overcome by the addition of increased amounts of IL2 in the coculture. As can be seen from Fig. 6, IL2 is necessary for T cell proliferation, but there is no signi®cant difference between the proliferation of T cells in cocultures with 100 U ml 1 compared to cocultures with up to 800 U ml 1 IL2.
RPE Cells Induce Down-regulation of IL2R-b and -g Chain Expression
F IG . 3. Effect of cell contact on ARPE-19 cell induced inhibition of T cells. The proliferation of activated T cells was measured after either incubation of T cells alone (T), of T cells directly together with ARPE-19 cells (T R) (1 : 1), of T cells grown in a transwell system (pore size 0.2 mm) separated from an ARPE-19 cell monolayer (T(in) R) (1 : 1) or T cells in transwell system separated from an ARPE-19 cell monolayer together with activated T cells (T(in) R T) (1 : 1 : 1) for 48 hr. ARPE-19 cells (5 105) were seeded onto a 96 well culture plate and activated T cells (5 105) were added the following day directly on top of the RPE cells or in an insert. Proliferation was measured by adding 3 H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of three replicates from a representative experiment of four. Statistical analysis with unpaired t-test showed that ARPE19 cells only inhibited the proliferation of T cells signi®cantly when the T cells were incubated directly on top of ARPE-19 cells *P 5 0.05.
Since T cell proliferation was maintained at comparable levels over a wide range of IL2 concentrations, the authors speculated that the RPE cell induced down-regulation of proliferation could be due to changes in IL2R signalling. Accordingly, the expression pattern of IL2R-a, -b and -g chains of T cells cultured either alone or cocultured with ARPE19 or HFRPE cells was investigated by ¯ow cytometry. Fig. 7 illustrates the mean expression of IL2R-a, -b and -g chains on CD3 positive cells after incubation of T cells alone or in coculture with ARPE-19 cells or HFRPE cells. Both foetal and adult derived RPE cells convert decreased IL2R-b and -g chain expression on T cells, which constitute the signal transducing segments of IL2R. The expression of IL2R chains is generally regulated by the presence of IL2. It was therefore tested if there was any effect of a lower concentration of IL2 in coculture media (50 U ml 1). There were no indications that the RPE cell induced
632
C. G. KáSTEL ET AL.
F IG . 4. Evaluation of the effect of the surface molecules CD2, CD18, CD40 and CD40L (A) and CD54 and CD58 (B) on ARPE19 cell induced inhibition of T cell proliferation. The proliferation of activated T cells was measured after either incubation alone or with ARPE-19 cells (1 : 1) with or without the presence of blocking antibodies as indicated. Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. Illustrated is the percent of proliferation of cocultured T cells compared to T cells cultured alone. The results depicted are the mean of three separate experiments each including three to four replicates. Statistical analysis with paired t-test showed that there was no signi®cant difference between the proliferation of T cells in coculture in the presence of blocking Ab or in the presence of the matching isotype control Ab.
F IG . 5. Effect of ARPE-19 cells (R) on the TCR negative T cell lines MF and Se-Ax. The proliferation of MF or Se-Ax cells was measured after either incubation alone or with ARPE-19 cells (1 : 1) for 48 hr in the absence (A) or presence (B) of IL2. ARPE19 cells (5 105) were seeded onto a 96 well culture plate and the following day MF or Se-Ax cells were added (5 105). Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of four replicates from a representative experiment of three. Statistical analysis with unpaired t-test showed that ARPE-19 cells inhibited both MF and Se-Ax signi®cantly in the absence as well as presence of IL2 *P 5 0.05.
IL2R chain modulation was sensitive to the lower IL2 concentration (data not shown). The T cells' expression of IL2R-a chain is signi®cantly up-regulated or unchanged after coincubation with ARPE-19 or HFRPE. Together these results suggest that the RPE cell induced inhibition of T cells may be achieved by the modulation of the T cells' normal expression pattern of IL2R-b and -g chain and possibly the IL2 signal. The Inhibition of T Cell Proliferation and IL2R Expression is Reversible The 7-AAD experiments (Fig. 1) showed that the depression of DNA synthesis was not accompanied by an increase in apoptosis of the cocultured T cells within the investigated time period. However, it is
possible that the effect of RPE cells is an induction of lasting G0 state or anergy in the T cells. To test this, the ability of the T cells to regain proliferation and capacity to achieve a `normal' expression pattern of IL2R-a, -b and -g chains was investigated. After 48 hr incubation alone or with ARPE-19 cells, the T cells were carefully removed from the cultures, and resuspended without addition of ARPE-19 cells in fresh medium containing IL2. Hereafter, the level of proliferation was analysed by 3H-Thymidine incorporation, and the expression of IL2R-a, -b and -g chains was measured by ¯ow cytometry. As can be seen from Fig. 8, the T cells from cocultures with ARPE-19 cells have a lower level of proliferation during the ®rst 6 hr in fresh media compared to T cells having been precultured alone, but after 24 hr comparable levels of proliferation is seen in the two culture types. Fig. 8(B) shows that a
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
633
macrophages on activated T cells (Strickland, Kees and Holt, 1996; Upham et al., 1997). Decrease or arrest of proliferation seems to be required for some types of apoptosis (Radvanyi, Mills and Miller, 1993; Lissy et al., 1998; Lee et al., 1999). Thus a total or partial stop in proliferation and induction of apoptosis may be part of the same sequence of effects, and subtle differences between
F IG . 6. Evaluation of the effect of various concentrations of IL2 on ARPE-19 induced proliferation inhibition of T cells. ARPE-19 cells (R) were seeded onto a 96 well culture plate and activated T cells (T) (5 105) were added the following day. In addition, T cells and ARPE-19 cells were set up separately in parallel cultures. The cells were incubated for 48 hr. Proliferation was measured by adding 3H-thymidine (1 mCi per well 1) 6 hr before the end of incubation time. The results depicted are the mean and SEM of four replicates from a representative experiment of three. Statistical analysis with unpaired t-test showed that the T cells in cocultures (T R) proliferate at the same rate in the presence of 100±800 U ml 1 IL2.
`normal' expression pattern of IL2R-b and -g chains is also achieved after 24 hr incubation in fresh media. Taken together, these results indicate that the RPE cell induced inhibition of T cells is a reversible process allowing a normal T cell function after removal of the RPE cells from the coculture. There is no sign of the T cells becoming suppressed in prolonged time after coincubation with RPE cells.
4. Discussion This study con®rm earlier studies in showing that human RPE cells in¯uence the cell cycle of T cells in vitro. The most signi®cant new ®nding in the present study is a decrease in functional IL2R expression, which is likely to explain the decrease in T cell proliferation and increase in apoptosis found in previous studies (Jùrgensen et al., 1998; Rezai et al., 1999; Farrokh-Siar et al., 1999a, 2000), since IL2 is essential for survival of activated T cells. In the present study, the changes in IL2R expression was paralleled by a decrease T cell proliferation. Decreased proliferation was also found in the above mentioned studies, but in addition, several of these studies showed an increase in apoptosis (Jùrgensen et al., 1998; Rezai et al., 1999; Farrokh-Siar et al., 2000). In the present study, there were, however, no signs of apoptotic processes or changes in several apoptotic proteins. Furthermore, within 24 hr after removal of RPE cells from the coculture, the T cells regained normal proliferation and IL2R chain expression. Similar effects have been found of alveolar
F IG . 7. T cells' expression of IL2R-a, -b and -g chain after incubation alone (T) or with ARPE-19 or HFRPE cells for 72 hr. The T cells' expression of the IL2R-a, -b and -g chain was measured by use of anti-IL2R-a, -b and -g chain Abs and anti-CD3 Ab followed by ¯ow cytometry. Illustrated is the average mean expression (n 7±10) and SEM of the IL2R chains of CD3 positive cells. Statistical analysis with paired t-test indicated signi®cant difference (P 5 0.05) between expression of IL2R chain of T cells cultured alone compared to T cells cultured with RPE cells where marked with the symbol (*).
634
F IG . 8. Ability of T cells to regain proliferation (A) and `normal' expression pattern of IL2R-a, -b and -g chain (B) after preincubation with ARPE-19 cell followed by incubation alone in fresh culture medium. T cells were removed from coculture after 48 hr and incubated alone in fresh media. The proliferation was measured after 6 and 24 hr incubation by addition of 3H-thymidine (1 mCi per well 1) and the expression of the IL2R-a, -b and -g chain was measured by ¯ow cytometry. The proliferation results depicted are the mean and SEM of four replicates from a representative experiment of three. The ¯ow cytometric results depicted are means and SEM of mean ¯uorescence intensity of (MFI) of T cells precultured with RPE cells divided with MFI of T cells precultured alone (n 3±5). Statistical analysis with paired t-test indicated signi®cant difference (P 5 0.05) between level of proliferation and expression of IL2R chain of T cells precultured alone compared to T cells precultured with RPE cells where marked with the symbol (*).
types of epithelial cells, types of T cells and culture types may be decisive for the outcome of the interaction between T cells and RPE cells. Here it is also of interest that the apoptotic process may vary in different experiments. In a previous study, the authors found that apoptosis most likely was due to suicide or fratricide between T cells, induced by Fas±FasL interaction, paralleling for example melanoma speci®c T cell apoptosis after MHC restricted recognition of FasL negative tumor cells as reported by (Zaks et al.,
C. G. KáSTEL ET AL.
1999; Restifo, 2000). However, another recent study indicates that apoptosis may not be due to Fas±FasL interaction (Rezai et al., 1999). Of special interest are the results of several studies from one research group (Rezai et al., 1999; FarrokhSiar et al., 1999a, 2000) which show that the RPE cell induced inhibiting factor is an uncharacterized soluble one, whereas the present authors ®nd that cell to cell contact is required for signi®cant inactivation of the T cellsÐeven if some soluble factor may be of importance. Thus, a decreased proliferation of T cells was found cultured in inserts, but not signi®cant. Since the same culture types ± including transwell cultures ± are used in these studies, this essential difference is dif®cult to explain, but is not likely to be due to the type of RPE cells used, since the present authors have used both foetal and adult RPE cells (Jùrgensen et al., 1998) and (present study) with comparable results. However, besides differences in handling of cells and culture systems, the differences between the studies here and those of Farrokh-Siar et al. may be due to that the authors use `stable' activated T cells, whereas they investigate the induction period of T cell activation. It is found here that RPE cells induce decreased proliferation of MF and Se-Ax cell lines both in presence and in absence of exogenous IL2. Thus, exogenous IL2 is not required for the depression of proliferation. However, the use of exogenous IL2 is necessary to avoid apoptosis of activated periphera T cells grown alone in culture (RoÈpke et al., 1996) as also shown in Fig. 6, where essentially no 3H-TdR incorporation is seen in T cells grown in the absence of IL2. Accordingly the authors have used IL2 in the experiments, both when T cells are cultured alone and when they are cocultured with RPE cells. Presence of IL2 in the cultures is not the reason for the absence of apoptosis, since IL2 was used in all cultures in an earlier study in which high percentages of apoptosis were seen after coculture of T cells and RPE cells (Jùrgensen et al., 1998). IL2 is an essential growth factor for the T cells, and the IL2 signal is given through the IL2R, which is a complex composed by the three chains, IL2R-a, -b and -g. IL2R functions as low af®nity receptor, when constituted by the IL2R-b and -g chains (Kd 10 9 M), and as high af®nity receptor, when constituted by all three chains IL2R-a, -b and -g (Kd 10 11 M) (Minami et al., 1993). The authors analysed the state of the T cells when cultured alone or in the presence of ARPE-19 or HFRPE cells with regard to their expression of IL2R and their capacity to receive an IL2 signal. The results showed a change in expression of IL2R-b and g chains after coculturing with ARPE-19 as well as HFRPE cells. In addition, the ARPE-19 cells seemed to have a tendency to increase the T cells' IL2R-a chain expression, an effect that was not seen by the HFRPE cells. This effect is, therefore, not thought to be of signi®cant importance for a plausible IL2R linked inhibition.
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
The change, in expression of IL2R-g chain after coculture, was very obvious. This chain is shared by all the known T cell growth factor (TCGF) receptors i.e. IL2R, IL4R, IL7R, IL9R and IL15R and is therefore a key feature for the T cells' potential for proliferation and survival (Akbar et al., 1996; Karnitz and Abraham, 1996; Nakajima et al., 1997; Malek, Porter and He, 1999; and Van Parijs et al., 1999). The IL2R-g chain is also essential for T cell activation and a recent publication has shown that allograft survival is prolonged in the presence of blocking antiIL2R-g chain Ab (Li et al., 2000). Caspi et al. (1987) investigated the inhibitory effect of rat MuÈller cells on T cells and found a decreased IL2 production and through binding studies, a reduced expression of high af®nity IL2R of cocultured T cells. These data could be explained by the ®ndings of changed expression of IL2R chains published here and extent their ®nding to include the RPE cells. A reduced expression of IL2R-b and -g chains could limit the cells' ability to form high af®nity IL2R complexes. It is hypothesized therefore that RPE cells may induce their inhibitory effect on cycling T cells by modulation of the T cell IL2R and that this may be a non Ag-speci®c way of human RPE cells to confer immune privilege in the posterior segment of the eye. So far, no factor has yet been assigned to the RPE cell induced effect on IL2R, but recently it was shown that PGE2 is able to inhibit the expression of IL2R-g chain of human NK cells (Joshi et al., 2001). Additionally, PGE2 has been found to participate in the Rat RPE cell induced inhibition of T cells, in studies by Liversidge et al. (1993). The present authors investigated the effect of the PGE2 antagonist; indomethacin, in this model. However, no effect was found of indomethacin on the RPE cell induced inhibition or the IL2R modulation of T cells. This indicates that another mechanism is involved. The present studies show that a signi®cant part of the inhibition is dependent on direct contact between the two cell types. Additionally, no difference was found between T cells cultured in the upper chamber of a transwell system with ARPE-19 cells in the bottom chamber and T cells cultured in the upper chamber of a transwell system with both ARPE-19 cells and T cells in the bottom chamber. These data suggest that the ARPE-19 cell induced inhibition of T cells initially depend on a cell contact-dependent mechanism or a soluble factor present in a low concentration or vulnerable to quick degradation. The involvement of the surface molecules CD54, CD58, CD40 and TCR was tested and none of these molecules seemed to be signi®cantly involved in the inhibitory process. The supernatants from cocultures of human RPE and T cells have previously been analysed for a panel of soluble molecules (Liversidge et al., 1998; Rezai et al., 1999; Farrokh-Siar et al., 1999a; Yoshida et al., 2000b). So far, none of these soluble factors have been ascribed to the inhibitory process in activated T cells induced by human RPE cells.
635
In conclusion, several studies indicate that RPE cell±T cell interactions may be of importance in the maintenance of the immune privilege in the posterior part of the eye. However, these studies are all in vitro studies, and even in this ± compared to in vivo ± simple system, the essential factors in RPE cell induced inhibition of T cell proliferation is not pinpointed. Both cell surface molecules and soluble factors may be operative. Further, the inhibitory effect of RPE cells on T cells is limited to activation of these cells, since RPE cells are able to present antigen to ± and induce proliferation of ± resting T cells (Jùrgensen et al., 2001). Thus, more investigations are needed for clari®cation of the intracellular pathways through which inhibition of activated T cells is achieved. The present demonstration of RPE cell induced changes in T cells' IL2R expression may give clues where as to focus these investigations. Acknowledgements The authors thank Dr Tongalp H. Tezel for supplying the HFRPE cells and Ane Marie Rulykke, Helene Rasmussen and Evelyn Kury for skilful technical assistance. This work was supported by grants from The Danish Eye Research Foundation, Vñrn om Synet, The Danish Medical Council, Novo Nordisk and The Beckett Foundation.
References Akbar, A. N., Borthwick, N. J., Wickremasinghe, R. G., Panayoitidis, P., Pilling, D., Bo®ll, M., Krajewski, S., Reed, J. C. and Salmon, M. (1996). Interleukin-2 receptor common gamma-chain signaling cytokines regulate activated T cell apoptosis in response to growth factor withdrawal: selective induction of anti-apoptotic (bcl-2, bcl-xL) but not pro- apoptotic (bax, bcl-xS) gene expression. Eur. J. Immunol. 26, 294±9. Caspi, R. R., Roberge, F. G. and Nussenblatt, R. B. (1987). Organ-resident, nonlymphoid cells suppress proliferation of autoimmune T-helper lymphocytes. Science 237, 1029±32. Cousins, S. W., McCabe, M. M., Danielpour, D. and Streilein, J. W. (1991). Identi®cation of transforming growth factor-beta as an immunosuppressive factor in aqueous humor. Invest. Ophthalmol. Vis. Sci. 32, 2201±11. Elner, S. G., Elner, V. M., Pavilack, M. A., Todd, R. F., III, Mayo-Bond, L., Franklin, W. A., Strieter, R. M., Kunkel, S. L. and Huber, A. R. (1992). Modulation and function of intercellular adhesion molecule-1 (CD54) on human retinal pigment epithelial cells. Lab. Invest. 66, 200±11. Farrokh-Siar, L., Rezai, K. A., Semnani, R. T., Patel, S. C., Ernest, J. T., Peterson, E. J., Koretzky, G. A. and van Seventer, G. A. (1999a). Human fetal retinal pigment epithelial cells induce apoptosis in the T-cell line Jurkat. Invest. Ophthalmol. Vis. Sci. 40, 1503±11. Farrokh-Siar, L., Rezai, K. A., Semnani, R. T., Patel, S. C., Ernest, J. T. and van Seventer, G. A. (1999b). Human fetal retinal epithelium suppresses the activation of CD4() and CD8() T-cells. Graefes Arch. Clin. Exp. Ophthalmol. 237, 934±9. Farrokh-Siar, L., Rezai, K. A., Palmer, E. M., Patel, S. C., Ernest, J. T. and van Seventer, G. A. (2000). Human fetal retinal pigment epithelium-induced cell cycle arrest, loss of mitochondrial membrane potential and apoptosis. Invest. Ophthalmol. Vis. Sci. 41, 3991±8.
636
Gospodarowicz, D. and Ill, C. (1980). Extracellular matrix and control of proliferation of vascular endothelial cells. J. Clin. Invest. 65, 1351±64. Grif®th, T. S., Brunner, T., Fletcher, S. M., Green, D. R. and Ferguson, T. A. (1995). Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270, 1189±92. Joshi, P. C., Zhou, X., Cuchens, M. and Jones, Q. (2001). Prostaglandin E2 suppressed IL-15-mediated human NK cell function through down-regulation of common gamma-chain. J. Immunol. 166, 885±91. Jùrgensen, A., Ropke, C., Nielsen, M., Madsen, H., Svejgaard, A. and Odum, N. (1996). Human thymic epithelial cells express functional HLA-DP molecules. Tissue Antigens 47, 300±6. Jùrgensen, A., Wiencke, A. K., la Cour, M., Kñstel, C. G., Madsen, H. O., Hamann, S., Lui, G. M., Scher®g, E., Prause, J. U., Svejgaard, A., Odum, N., Nissen, M. H. and Ropke, C. (1998). Human retinal pigment epithelial cell-induced apoptosis in activated T cells. Invest. Ophthalmol. Vis. Sci. 39, 1590±9. Jùrgensen, A., Junker, N., Kñstel, C. G., Liang, Y., Wiencke, A., Lui, G. M., Odum, N., Nissen, M. H. and Ropke, C. (2001). Superantigen presentation by human retinal pigment epithelial cells to T cells is dependent on CD2CD58 and CD18-CD54 molecule interaction. Exp. Eye Res. 73, 723±33. Kaiser, C. J., Ksander, B. R. and Streilein, J. W. (1989). Inhibition of lymphocyte proliferation by aqueous humor. Reg. Immunol. 2, 42±9. Kaltoft, K., Bisballe, S., Rasmussen, H. F., ThestrupPedersen, K., Thomsen, K. and Sterry, W. (1987). A continuous T-cell line from a patient with Sezary syndrome. Arch. Dermatol. Res. 279, 293±8. Kaltoft, K., Bisballe, S., Dyrberg, T., Boel, E., Rasmussen, P. B. and Thestrup-Pedersen, K. (1992). Establishment of two continuous T-cell strains from a single plaque of a patient with mycosis fungoides. In Vitro Cell Dev. Biol. 28A, 161±7. Kaplan, H. J., Leibole, M. A., Tezel, T. and Ferguson, T. A. (1999). Fas ligand (CD95 ligand) controls angiogenesis beneath the retina. Nat. Med. 5, 292±7. Karnitz, L. M. and Abraham, R. T. (1996). Interleukin-2 receptor signaling mechanisms. Adv. Immunol. 61, 147±99. Kñstel, C. G., Madsen, H., Prause, J. U., Jùrgensen, A., la Cour, M., Lui, G. M., Odum, N., Nissen, M. H. and Ropke, C. (2001). Lack of FasL Expression in Cultured Human Retinal Pigment Epithelial Cells. Exp. Clin. Immunogenet. 18, 34±41. Lee, I. H., Li, W. P., Hisert, K. B. and Ivashkiv, L. B. (1999). Inhibition of interleukin 2 signaling and signal transducer and activator of transcription (STAT)5 activation during T cell receptor-mediated feedback inhibition of T cell expansion. J. Exp. Med. 190, 1263±74. Li, X. C., Ima, A., Li, Y., Zheng, X. X., Malek, T. R. and Strom, T. B. (2000). Blocking the common gammachain of cytokine receptors induces T cell apoptosis and long-term islet allograft survival. J. Immunol. 164, 1193±9. Lissy, N. A., VanDyk, L. F., Becker-Hapak, M., VoceroAkbani, A., Mendler, J. H. and Dowdy, S. F. (1998). TCR antigen-induced cell death occurs from a late G1 phase cell cycle check point. Immunity 8, 57±65. Liversidge, J., Sewell, H. F., Thomson, A. W. and Forrester, J. V. (1988). Lymphokine-induced MHC class II antigen expression on cultured retinal pigment epithelial cells and the in¯uence of cyclosporin A. Immunology 63, 313±7.
C. G. KáSTEL ET AL.
Liversidge, J., McKay, D., Mullen, G. and Forrester, J. V. (1993). Retinal pigment epithelial cells modulate lymphocyte function at the blood-retina barrier by autocrine PGE2 and membrane-bound mechanisms. Cell Immunol. 149, 315±30. Malek, T. R., Porter, B. O. and He, Y. W. (1999). Multiple gamma c-dependent cytokines regulate T cell development. Immunol. Today 20, 71±6. Minami, Y., Kono, T., Miyazaki, T. and Taniguchi, T. (1993). The IL-2 receptor complex: its structure, function, and target genes. Ann. Rev. Immunol. 11, 245±68. Morgan, B. P. and Meri, S. (1994). Membrane proteins that protect against complement lysis. Springer Semin. Immunopathol. 15, 369±96. Nakajima, H., Shores, E. W., Noguchi, M. and Leonard, W. J. (1997). The common cytokine receptor gamma chain plays an essential role in regulating lymphoid homeostasis. J. Exp. Med. 185, 189±95. Platts, K. E., Benson, M. T., Rennie, I. G., Sharrard, R. M. and Rees, R. C. (1995). Cytokine modulation of adhesion molecule expression on human retinal pigment epithelial cells. Invest. Ophthalmol. Vis. Sci. 36, 2262±9. Rabinovitch, P. S., Torres, R. M. and Engel, D. (1986). Simultaneous cell cycle analysis and two-color surface immuno¯uorescence using 7-amino-actnimycin D and single laser excitation: applications to study of cell activation and cell cycle of murine LT-1 B cells. J. Immunol. 136, 2769±73. Radvanyi, L. G., Mills, G. B. and Miller, R. G. (1993). Religation of the T cell receptor after primary activation of mature T cells inhibits proliferation and induces apoptotic cell death. J. Immunol. 150, 5704±15. Restifo, N. P. (2000). Not so Fas: Re-evaluating the mechanisms of immune privilege and tumor escape. Nat. Med. 6, 493±5. Rezai, K. A., Semnani, R. T., Farrokh-Siar, L., Hamann, K. J., Patel, S. C., Ernest, J. T. and vanSeventer, G. A. (1999). Human fetal retinal pigment epithelial cells induce apoptosis in allogenic T-cells in a Fas ligand and PGE2 independent pathway. Curr. Eye Res. 18, 430±9. RoÈpke, C., Gladstone, P., Nielsen, M., Borregaard, N., Ledbetter, J. A., Svejgaard, A. and édum, N. (1996). Apoptosis following interleukin-2 withdrawal from T cells: Evidence for a regulatory role of CD18 (b2-integrin) molecules. Tissue Antigens 48, 127±35. Streilein, J. W. (1993). Immune privilege as the result of local tissue barriers and immunosuppressive microenvironments. Curr. Opin. Immunol. 5, 428±32. Streilein, J. W. and Stein-Streilein, J. (2000). Does innate immune privilege exist?. J. Leukoc Biol. 67, 479±87. Strickland, D., Kees, U. R. and Holt, P. G. (1996). Regulation of T-cell activation in the lung: alveolar macrophages induce reversible T-cell anergy in vitro associated with inhibition of interleukin-2 receptor signal transduction. Immunology 87, 250±8. Taylor, A. W., Streilein, J. W. and Cousins, S. W. (1992). Identi®cation of alpha-melanocyte stimulating hormone as a potential immunosuppressive factor in aqueous humor. Curr. Eye Res. 11, 1199±206. Taylor, A. W., Streilein, J. W. and Cousins, S. W. (1994). Immunoreactive vasoactive intestinal peptide contributes to the immunosuppressive activity of normal aqueous humor. J. Immunol. 153, 1080±6. Taylor, A. W., Yee, D. G. and Streilein, J. W. (1998). Suppression of nitric oxide generated by in¯ammatory macrophages by calcitonin gene-related peptide in aqueous humor. Invest. Ophthalmol. Vis. Sci. 39, 1372±8.
I N H I B I T I O N O F AC T I VAT E D T C E L L S B Y H R P E
Tezel, T. H., Del Priore, L. V. and Kaplan, H. J. (1997). Harvest and storage of adult human retinal pigment epithelial sheets. Curr. Eye Res. 16, 802±9. Upham, J. W., Strickland, D. H., Robinson, B. W. and Holt, P. G. (1997). Selective inhibition of T cell proliferation but not expression of effector function by human alveolar macrophages. Thorax 52, 786±95. Van Parijs, L., Refaeli, Y., Lord, J. D., Nelson, B. H., Abbas, A. K. and Baltimore, D. (1999). Uncoupling IL-2 signals that regulate T cell proliferation, survival, and Fasmediated activation induced cell death. Immunity 11, 281±8. Willermain, F., Caspers-Velu, L., Baudson, N., Dubois, C., Hamdane, M., Willems, F., Velu, T. and Bruyns, C. (2000). Role and expression of CD40 on human retinal pigment epithelial cells. Invest. Ophthalmol. Vis. Sci. 41, 3485±91. Yoshida, M., Takeuchi, M. and Streilein, J. W. (2000a). Participation of pigment epithelium of iris and ciliary
637
body in ocular immune privilege. I. Inhibition of T-cell activation in vitro by direct cell-to-cell contact. Invest. Ophthalmol. Vis. Sci. 41, 811±21. Yoshida, M., Kezuka, T. and Streilein, J. W. (2000b). Participation of pigment epithelium of iris and ciliary body in ocular immune privilege. 2. Generation of TGFbeta-producing regulatory T cells. Invest. Ophthalmol. Vis. Sci. 41, 3862±70. Zaks, T. Z., Chappell, D. B., Rosenberg, S. A. and Restifo, N. P. (1999). Fas-mediated suicide of tumor-reactive T cells following activation by speci®c tumor: selective rescue by caspase inhibition. J. Immunol. 162, 3273±9. Zavazava, N., Halene, M., Westphal, E., Nolle, B., Duncker, G., Eckstein, E., Harpprecht, J. and Muller-Ruchholtz, W. (1991). Expression of MHC class I and II molecules by cadaver retinal pigment epithelium cells: optimization of post-mortem HLA typing. Clin. Exp. Immunol. 84, 163±6.