Ocular Immunology & Inflammation, 2015; 23(3): 194–204 ! Informa Healthcare USA, Inc. ISSN: 0927-3948 print / 1744-5078 online DOI: 10.3109/09273948.2014.891754
ORIGINAL ARTICLE
In Vitro Immunomodulatory Effects of Human Corneal Stromal Cells on Human Th17 Cells* Jong Joo Lee, MD1,2, Jin Suk Ryu, MS2, Hyun Jeong Jeong, MS2, Mee Kum Kim, MD, PhD1,2, and Won Ryang Wee, MD, PhD1,2 1
Department of Ophthalmology, Seoul National University College of Medicine, Seoul, Korea; and Laboratory of Ocular Regenerative Medicine and Immunology, Seoul Artificial Eye Center, Seoul National University Hospital Clinical Research Institute, Seoul, Korea and 2Laboratory of Ocular Regenerative Medicine and Immunology, Seoul Artificial Eye Center, Seoul National University Hospital Clinical Research Institute, Seoul, Korea
ABSTRACT Purpose: To investigate the immunomodulatory effects of human corneal stromal cells on the function of Th17 cells, in vitro. Methods: Th17 cells were co-cultured with corneal stromal cells with or without 1-methyltryptophan, 1400w dihydrochloride, interleukin (IL)-10 inhibitor, or transforming growth factor-beta (TGF-b) inhibitor. The secretion of IL-17A by Th17 cells and the upregulation of T regulatory (Treg) cell frequencies were evaluated. Transwell assays were conducted to evaluate whether the immunomodulatory effects of stromal cells were dependent on cell-to-cell contact. Results: Corneal stromal cells suppressed the secretive function of Th17 cells and upregulated Treg cell frequencies, enhancing TGF-b. Tranwell assays showed similar immunomodulatory effects of the stromal cells on the function of Th17 cells. TGF-b inhibitors attenuated the suppressive effects of the stromal cells on the function of Th17 cells, whereas other inhibitors didn’t. Conclusion: It which indicates that human corneal stromal cells suppress the function of Th17 cells through a TGF-b-dependent pathway. Keywords: Cornea, IL-17, stromal cells, T cells, T regulatory cells, TGF-b, Th17 cells
Immune privilege, a part of immune homeostasis, is a dynamic process involving multiple physiological, anatomical, and immunological properties that allow organs to perform their primary function with less inflammation. The brain, testis, and the eye have evolved into immune-privileged organs to save their critical functions, including control of homeostasis, reproduction, and vision, while the gut and lung have evolved into immune-modulatory organs to ensure their functionality in a foreign-antigen-enriched environment.1 The immune privilege of the eye was first recognized over 130 years ago2; the antigen
sequestration concept of the anterior chamber was first proposed 80 years later by Medawar.3 Thereon, extensive investigation has revealed the existence of anterior chamber associated immune deviation (ACAID).4 On the other hand, current studies have shed light on the possibility that the corneal stromal cells might also contribute to the immune privilege in the eye, and that other stromal cells residing in lymphoid organs affect immunohomeostasis.5–10 However, the immune modulatory role of the corneal stromal cells remains unclear. This study addresses whether corneal stromal cells suppress the function of Th17 cells. We also
*Critical comments were kindly provided by Su Kil Seo (Department of Ophthalmology, Busan Paik Hospital, Inje University Medical College, Busan, Korea) in measuring and interpreting the IDO-dependent pathway, and Won Woo Lee (Department of Microbiology and Immunology, Seoul National University College of Medicine, Seoul, Korea) kindly guided us to install Th17 cell-driving conditions. Received 7 June 2013; revised 8 January 2014; accepted 3 February 2014; published online 5 March 2014 Correspondence: Mee Kum Kim, MD, PhD, Department of Ophthalmology, Seoul National University College of Medicine, 103 Daehang-ro, Jongno-gu, Seoul, 110-799, Republic of Korea. E-mail: [email protected]
194
Human Corneal Stromal Cells and Human Th17 Cells 195 investigate which pathway is involved in the modulation of T cells.
MATERIALS AND METHODS The study was approved by the institutional review board of the Seoul National University Hospital (IRB approval number: H-1007-099-324) and was performed in accordance with the guidelines of the Code of Ethics of the World Medical Association (Declaration of Helsinki). Human blood was collected from healthy donor volunteers with written informed consent.
CD4+ T-cell Isolation and Th17 Differentiation CD4+ human peripheral blood T cells were isolated from freshly collected, heparinized peripheral blood using CD4 magnetic activated cell sorting (MACS) beads (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s protocols. Th17 cells were differentiated by stimulating 5 104 CD4+ T cells on a plate coated with 1 mg/mL anti-CD3 or anti-CD28 monoclonal antibody (mAb; eBioscience, San Diego, CA, USA) and subsequently culturing with 25 ng/mL IL-1b (R&D Systems, Minneapolis, USA) and 25 ng/mL IL-23 (R&D Systems) in RPMI 1640 medium (Welgen, Daegu, Korea) containing 10% fetal bovine serum (FBS), penicillin/streptomycin (100 U/mL), and 10 mM HEPES in a 48-well plate (NUNC, Roskilde, Denmark) for 7 days.
In vitro Suppression of Proliferation To examine the suppressive effects of corneal stromal cells on the proliferation of CD4+ T cells, isolated CD4+ T cells were stained with 5 mM carboxyfluorescein diacetate succinimidyl ester (CFSE; Invitrogen, Carlsbad, CA, USA). Anti-CD3/CD28 antibodies (1 mg/mL; eBioscience) were resuspended in the supplemented RPMI 1640 medium (Welgen) to stimulate cell proliferation. The T cells stimulated by antiCD3/CD28 antibodies served as positive control, while unstimulated T cells served as negative control. CD4+ T cells were placed in co-culture with stromal cells at a ratio of 10:1 (5 104 T cells: 5 103 stromal cells) and were co-cultured for 5 days in the supplemented RPMI 1640 medium. Suppression of T-cell proliferation was measured by reduction in the CFSE concentration by flow cytometry using FACS Vantage Flow Cytometer (BD Biosciences, San Jose, CA, USA). Reduction rate (%) was calculated by subtracting the proliferation percentage of the stromal cell-treated T cells (treated %) from the proliferation percentage of positive controlled T cells (PC %), with the value being divided by the proliferation percentage of positive controlled T cells and then by multiplying the resulting value by 100 (reduction rate % = [PC % – treated %] 7 PC% 100) A total of 250 mM 1-methyl-d,l-Trp; (1-D/L MT; Sigma Aldrich, St. Louis, MO, USA) was applied to assess whether suppression was diminished in an indoleamine 2,3-dioxygenase (IDO)-dependent manner. The experiment used eight sets of samples.
Corneal Stromal Cell Isolation and Culture
In vitro Functional Assays for the Suppression of Th17 Cells
Human corneas obtained from the Northwest Lions Eye Bank (Seattle, WA, USA) were used for experiments after the central corneal button had been transplanted to human subjects. The epithelial layers were removed using a surgical blade and the Descemet membranes were peeled off mechanically. The remaining corneal stroma was treated using 1.2 U/mL dispase II (Cat. No. 04 942 078 001; Roche Applied Science, Basel, Switzerland) and 200 U/mL type I collagenase (Worthington, Lakewood, NJ, USA). The cells were harvested by centrifugation at 1200 rpm for 5 min and inoculated into the culture plate with DMEM/F12 (Cat. No. 12-719F; Lonza, Valais, Switzerland) containing 10% fetal bovine serum (Welgen), penicillin/streptomycin (100 U/mL), 3.151 g/L glucose, L-glutamine, and 15 mM HEPES. The stromal cells were primarily cultured at 37 C in a carbon dioxide incubator for 1–2 weeks and were passaged 2–5 times. The cells were harvested and seeded (5 103/well) in a 48-well plate (NUNC) for co-culture with CD4+ T cells.
CD4+ T cells that had been treated with IL-1b and IL-23 for Th17 differentiation were co-cultured with stromal cells at a ratio of 10:1 (5 104 T cells: 5 103 stromal cells) for 2 days in the supplemented RPMI 1640 medium after stimulation. Anti-CD3/CD28 antibodies (1 mg/mL; eBioscience) were resuspended in the supplemented RPMI 1640 medium (Welgen) to stimulate cell proliferation. Suspended Th17 cells were separated from the stromal cells, and isolated Th17 cells were subsequently cultured for an additional 5 days until 7 days (Supplementary Figure S1). Control Th17 cells were cultured for 7 days in the same condition without co-culture. A total of 250 mM 1-d/l MT or 1 mM/mL 1400w dihydrochloride was applied during the co-culture of IFN-g (1000U/mL)-treated stromal cells to assess whether the suppression was in a IDO- or nitric oxide synthase 2 (NOS2)-dependent manner. After 7 days, T cells were incubated with CD4-fluorescein isothiocynate (FITC) (eBioscience) and CD45RO-PerCP (eBioscience) for 30 min at 4 C, and the cells were
!
2015 Informa Healthcare USA, Inc.
196 J. J. Lee et al. washed twice. After treatment with phorbol 12-myristate 13-acetate (PMA; 50 ng/mL; Sigma-Aldrich) and ionomycin (1 mg/mL; Sigma-Aldrich), the protein transporter inhibitor monensin (Golgi Plug 1 mL/mL, BD Phamingen, San Diego, USA) was treated simultaneously. Intracellular staining was performed after 5 h of stimulation. The cells were fixed with a fixation/permeabilization solution for 20 min at 4 C. After washing with the Perm/Wash buffer, the cells were incubated with IL-17-PE (BD Phamingen) and IFN-g-APC (BD Phamingen) for 60 min at 4 C. Flow cytometric analysis was performed using FACS Canto (Becton Dickinson, San Diego, CA, USA), and data were analyzed by FlowJo Version 7.6.5 (Treestar, Ashland, OR, USA). The experiments were performed in 4 replicates. The levels of IL-17 and IFN-g in supernatants from the T-cell culture after removal of stromal cells were also measured. Commercial kits for enzyme-linked immunosorbent assay (ELISA) were used as per the manufacturers’ protocols (DuoSet ELISA; R&D Systems, Minneapolis, MN, USA). The experiments were conducted in 4 replicates. To explore whether the stromal cells suppress Th17 cells in a cell-to-cell contact-dependent manner or not, transwell assays of the co-cultures were conducted using the same protocol mentioned above with Millicell culture plate inserts (0.4 mm, Billerica, MA, USA). Cell surface staining with fluorescein isothiocyanate (FITC)-conjugated anti-CD4 (eBioscience) and PerCP-conjugated anti-CD45RO (eBioscience) antibodies, and intracellular staining with PE-conjugated anti-IL-17 (BD Pharmingen) antibodies was performed in the same manner as mentioned above. The experiments were repeated in 6 replicates. AS101 (2.5 mg/mL, IL-10 inhibitors; Tocris Bioscience, Ellisville, MO, USA) or 1 mM/mL SB 431542 (TGF-b inhibitors; Tocris Bioscience, Ellisville, MO, USA) was applied during the culture to assess whether the suppression was in an IL-10- or TGFb-dependent manner. Cell surface staining with fluorescein isothiocyanate (FITC)-conjugated anti-CD4 (eBioscience) and PerCP-conjugated anti-CD45RO (eBioscience) antibodies, and intracellular staining with PE-conjugated anti-IL-17 (BD Pharmingen) antibodies was performed in the same manner as mentioned above. The experiments were performed in 6 replicates.
The Change of TGF-b and IL-10 during Co-cultures The levels TGF-b and IL-10 were determined in supernatants from 3-day co-cultures of both CD4+ T cells and stromal cells using commercial DuoSet ELISA kits following the manufacturers’ protocols (R&D Systems). The concentration used for the
standard curve ranged from 1000 to 15.625 pg/mL, using human recombinant TGFb, and IL-10 (R&D Systems). The experiments were performed in 11 sets. To explore whether TGF-b and IL-10 were produced by either corneal stromal cells or T cells, harvested T cells in pellets after co-culture were analyzed by flow cytometry. Cells were stimulated for 6 h with PMA (10 ng/mL) and inomycin (1 mg/mL) and were treated simultaneously by the protein transporter inhibitor cocktail (Brefeldin A 5.3 mM + Monensin 1 mM, 2 mL/mL, eBioscience). Cell surface staining was performed with CD4 PE-cy7 (e-Bioscience) and the intracellular staining was performed using LAP (TGF-b) PerCP-cy5.5 (e-Bioscience) and IL-10–FITC (e-Bioscience).
In Vitro Change of CD4+CD25hiFOXP3hi T Regulatory Cells in Co-cultures CD4+ T cells stimulated with anti-CD3/CD28 antibodies (1 mg/mL; eBioscience) were co-cultured with stromal cells at a ratio of 10:1 (5 104 T cells: 5 103 stromal cells) for 3 days. To evaluate the changes of CD4+CD25hiFOXP3hi cells, flow cytometry was conducted after isolation of T cells. Harvested CD4+ T cells were incubated with CD4 PE-cy7 (e-bioscience) and CD25 APC (eBioscience) for 30 min at 4 C, and the cells were washed twice. The cells were fixed with a fixation/permeabilization solution for 60 min at 4 C. After washing with the Perm/Wash buffer, the cells were incubated with FOXP3-PE (BD Phamingen) for 60 min at 4 C. PE-conjugated isotype Ig G(eBioscience) was stained as isotype control. Flow cytometric analysis and data analysis were performed as described above. The experiments were performed using 6 replicates. To check IDO dependency, 1-d/l MT was applied (250 mM) to assess whether the changes of T regulatory cells would be in an IDO-dependent manner. To investigate whether stromal cells upregulate Treg cells in a cell-to-cell contact-dependent manner or not, transwell assays of the co-cultures were conducted using the same protocol mentioned above with Millicell culture plate inserts (0.4 mm). The procedures of the surface staining with CD4 PE-cy7 (e-Bioscience) and CD25-APC (eBioscience) and the intracellular staining with FOXP3-PE (BD Phamingen) were done as mentioned above. The experiments were conducted in 6 replicates.
Statistical Analysis SPSS 17.0 software (SPSS, Chicago, IL, USA) was used. The proportions of proliferating cells or IL-17/ IFN-g secreting cells, changes of T regulatory Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 197 cells, or the changes of the other cytokines were compared to those of the positive control using the nonparametric Wilcoxon signed rank test. The p values 50.05 were considered to be statistically significant.
RESULTS Corneal Stromal Cells Hinder CD4+ T-cell Proliferation Stromal cells were co-cultured with CD4+ T cells at a ratio of 1:1 to 1:100. Ratios of 1:1 to 1:10 produced CD4+ T-cell suppression (Figure 1a). Thereafter, experiments were done at a ratio of 1:10. The proliferation of CD4+ T cells was effectively reduced by 62.6% when co-cultured with corneal stromal cells (p = 0.012, vs. positive control, Wilcoxon signed rank test; Figure 1b). When stromal cells were treated with IFN-g to induce IDO,6 the inhibitory effect was similar to the effect of nontreated stromal cells (Figures 1b) on T cells. The suppressive effect of the stromal cells was not significantly different depending on the passage number (Figure 1c).
Corneal Stromal Cells Suppress IFN-c and IL-17 Secretion by Th17 Cells via the TGF-b Pathway Corneal stromal cells significantly suppressed the secretion of both IFN-g and IL-17 by Th17 cells, which were gated by CD4hi and CD45ROhi (p = 0.043, respectively, vs. positive control, Wilcoxon signed rank test; Figure 2a and b), which was further confirmed by the overall low levels of IFN-g and IL-17 in the supernatants from isolated Th17 cells after stromal cells had been removed (p = 0.000, respectively, vs. positive control, Wilcoxon signed rank test; Figure 2c). 1-MT or 1400w treatment did not abrogate the effect of suppression of stromal cells on both proliferation of CD4+ T cells and IL-17/IFN-g secretion by Th17 cells (Figure 3). On the other hand, the TGF-b inhibitor attenuated the effect of stromal cells on the secretion of IL-17 by Th17 cells (p = 0.028, Wilcoxon signed rank test), whereas the IL-10 inhibitor did not interfere with these effects (Figure 4).
Anti-inflammatory Cytokines Increase in the Co-cultured Supernatants The anti-inflammatory cytokines TGF-b and IL-10 were also significantly elevated during the co-culture (p = 0.000–0.016, vs. control (CD4+ T cells or stromal cells alone, Mann-Whitney U test; Figure 5). As cell !
2015 Informa Healthcare USA, Inc.
sources of TGF-b and IL-10, a few CD4+ T cells secreted TGF-b. On the other hand, CD4+ T cells did not secrete IL-10 at all (Supplementary Figure S2), suggesting that most of the TGF-b and IL-10 might be produced by corneal stromal cells.
Stromal Cells Upregulate CD4+CD25hiFOXP3hi T Regulatory Cells Upregulation of T regulatory cells by co-cultured stromal cells was noted, which was statistically significant (p = 0.028, Wilcoxon signed rank test; Figure 6). IFN-g-treated stromal cells also enhanced T regulatory cells (p = 0.028, Wilcoxon signed rank test). 1-MT treatment did not attenuate the upregulation of CD4+CD25hiFOXP3hi T regulatory cells.
Stromal Cells Modulate Th17 Cells and Treg Cells through the Paracrine Pathway The suppressive effects of stromal cells on Th17 cell function were maintained when the cells were co-cultured using transwells, suggesting that the effect was not dependent on a cell-to-cell contact (Figure 7a and b; p = 0.028, Wilcoxon signed rank test). In addition, co-culturing stromal cells still resulted in an upregulation of Treg cell frequencies, suggesting that the upregulatory effect was not dependent on cell-to-cell contact (Figure 7c and d; p = 0.028, Wilcoxon signed rank test).
DISCUSSION This study demonstrated the immunomodulatory effects of human corneal stromal cells on the suppression of Th17 cells and upregulation of T regulatory cells along with the increase of TGF-b, in vitro. TGF-b may prefer to induce T regulatory cells by interfering with Th17 cell differentiation. Since mesenchymal stem cells have revealed their immune suppressive functions through various disease models in many research papers, the other stromal cells of mesenchymal origin have been currently drawing attention in their immune modulatory roles in homeostasis.8,11–13 Along with previous studies,5–10,12,13 our study suggests that tissue stromal cells of the eyes also seem to have an immune-modulatory capacity. However, unlike fibroblastic reticular cells (FRCs) in previous reports, we found that corneal stromal cells suppressed the effector function of CD4 T cells by non-IDO- or NOSdependent pathways. Besides, the effect of corneal stromal cells on the T cells appeared to be less potent than the effect of FRCs, which was reported by Veronika et al.11 FRCs suppressed T cells in a ratio of
198 J. J. Lee et al.
FIGURE 1. Suppression of CD4+ T-cell proliferation by stromal cells. (a) Suppressive function of corneal stromal cells on CD4+ T cells are attenuated using incrementally different ratios of stromal cells to CD4+ T cells. (b) The proliferation of CD4+ T cells was effectively reduced when they were in contact with stromal cells regardless of IFNg pretreatment in the stromal cells. *p = 0.012 (Wilcoxon signed rank test). (c) Reduction rate did not change according to the passage of stromal cells. All three passages of stromal cells caused significant reduction of CD4+ T-cell proliferation compared to positive control (p = 0.014, Mann-Whitney U test). All data represent the means ± standard error (n = 8).
1:100 (FRCs:T) while, our data showed that corneal stromal cells attenuated T cells in a ratio of 1:10 (stromal cells:T). It might be plausible that stromal cells would have preferential functions depending on residential area. Lymph nodes have a few FRCs and millions of T cells, on the contrary, the cornea has lots of stromal cells and a few T cells. The FRCs in the lymph nodes are supposed to encounter lots of T cells, which need to be tightly controlled for homeostasis in normal conditions. Meanwhile corneal stromal cells are expected to encounter some surveilling T cells that are occasionally found in the cornea and conjunctiva.14 It suggests that even less potent control of stromal cells might possibly be enough to maintain
homeostasis in the cornea rather than in lymph nodes in view of the ratio between the stromal cells and T cells. On the other hand, most of the reports revealed that IFN-g potentiated the immune modulatory functions of tissue stromal cells, presumably by the IFN-g-enhanced expression of IDO, NOS 2, or arginase.8,11 The reason we focused on the effect of IDO in stromal cells in the first place is that we found high expression of IDO in corneal stromal cells in RT-PCR (data not shown), as well as profound increase of IDO with IFN-g treatment. Meanwhile, our data presented similar suppressive effects of stromal cells, regardless of treatment of IFN-g and no Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 199
FIGURE 2. Suppression of Th17 cells by stromal cells. (a) Stromal cells suppressed IL-17/IFN-g secreting CD4 T cells. Results represent one of four separate experiments with Th17 cells derived by IL-1b/IL-23 and aCD3/aCD28, which were co-cultured with stromal cells by the protocol described by supplementary Figure S1. (b) The suppressive effect of stromal cells on IFN-g and IL-17 secretion by CD4hiCD45ROhi memory Th17 cells was statistically significant compared to positive control (p = 0.043, Wilcoxon signed rank test). (c) ELISA presented low levels of both IFN-g and IL-17 in the supernatants from isolated CD4 T cells after stromal cells had been removed. *p = 0.000 (Mann-Whitney U test). All data represent the means ± standard error (n = 4).
interference by IDO or NOS2 inhibitors. They suggested that the inhibitory effect of corneal stromal cells was less likely to be dependent on IFN-g. As flow cytometric analysis revealed that only a few T cells secreted TGF-b, stromal cells appeared !
2015 Informa Healthcare USA, Inc.
to produce most of the TGF-b, which seemed to be higher than the TGF-b production in mesenchymal stem cells (data not shown). Considering that the eye is composed of immune-privileged tissues with lots of immune-suppressive molecules, including TGF-b,
200 J. J. Lee et al.
FIGURE 3. The effect of IDO inhibitors and NOS2 inhibitors on the suppressive functions of stromal cells. 1-MT treatment did not attenuate for stromal cells to inhibit CD4+ T-cell proliferation (a, b) or the secretion of both IFN-g and IL-17 by Th17 cells (c, d). CD4 T means CD4+T cells, 1-MT stands for 1-methyl-d,l-Trp, and S(IFNg) refers to stromal cells pretreated with IFN-g. aCD3/aCD28 and IL-1b/IL-23 preapplied Th17 cells with treatment of 1-MT are used as a positive control to exclude the direct effect of 1-MT. (E, F) To analyze whether the suppressive effect of corneal stromal cells on T-cell function is blocked by inhibition of NOS2, 1 mM/mL 1400w dihydrochloride was treated during the co-culture. The proliferation of CD4+ T cells or the secretory function of IFN-g/IL-17 by Th17 cells was not largely affected by inhibition of NOS2. All data represent the means ± standard error (n = 4). 2d indicates 2 days of co-culture. Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 201
Figure 3. Continued.
FIGURE 4. The effect of TGF-b or IL-10 on the suppressive function of stromal cells. The TGF-b inhibitor (SB 431542) attenuated the effect of stromal cells on the secretion of IL-17 by Th17 cells (p = 0.028, Wilcoxon signed rank test), whereas IL-10 inhibitors (AS101) did not interfere with these effects. aCD3/aCD28 and IL-1b/IL-23 applied to the CD4+ T cells for the differentiation of Th17 cells. All data represent the means ± standard error (n = 6). 3d indicates 3 days of co-culture.
and the corneal stromal cells are one of the major cell sources of TGF-b secretion, TGF-b-associated immune suppression of corneal stromal cells on T cells would make sense. Furthermore, our data corresponded well with the fact that TGF-b could move the balance of the Th17–Treg cell axis toward the induction of T regulatory cells by diminishing Th17 cells. The initial co-culture showed not only a strong upregulation of !
2015 Informa Healthcare USA, Inc.
anti-inflammatory cytokines such as TGF-b and IL-10 but also an increase of proinflammatory cytokines like IFN-g and IL-6 (data not shown). However, the antiinflammatory effect of TGF-b appears to overcome the effect of proinflammatory cytokines by modulating the Th17–Treg axis over time, resulting in an overall decrease of IFN-g and IL-17 as the culture progressed. Although the main Th17-driving factors in mice are IL-6 and TGF-b, human Th17 cells are mainly induced by IL-1b and IL-23 and not by IL-6 and TGF-b. It is plausible that even high IL-6 and TGF-b combination did not increase percentages of Th17 cells in our human cell study. Our data implies that stromal cells may induce Treg cells by modulating the Th17–Treg balance. However, it should be further evaluated whether corneal stromal cells promote the function of Treg cells or induce Treg differentiation in future studies. IL-10 was not produced by CD4+ T cells in our study. Therefore, our data did not support that IL-10 might be involved in promoting T regulatory cells (data not shown) on contrast to previous reports.15 In addition, to rule out the argument about the confounding effect due to heterogenic cells mixed with the stromal cells (we found 1–10% of CD11b+ or CD11c+ cells were mixed in the cultured corneal stromal cells), we sorted out CD11b+ or CD11c+ cells through flow cytometry (Supplementary Figure S3). Nevertheless, we still observed almost the same inhibitory functions of the keratocytes on the secretion of IFN-g and IL-17A by Th17 cells. The data suggested
202 J. J. Lee et al.
FIGURE 5. Change of the anti-inflammatory cytokines during co-culture. ELISA revealed markedly increased secretion of TGF-b (a) and IL-10 (b) in 3 days co-cultured supernatants of both corneal stromal cells and CD4+ T cells. All data represent the means + standard error (n = 11). S indicates stromal cells, and CD4 T indicates CD4+ T cells. *p = 0.000, yp = 0.001, xp = 0.016 (Mann-Whitney U test).
FIGURE 6. Upregulation of T regulatory cells with co-culture of stromal cells. (a) Results represent one of six separate experiments. CD4 T indicates CD4+ T cells. 3d indicates 3 days of co-culture. (b) A significant increase of T regulatory cells was observed in the presence of stromal cells regardless of IFN-g treatment, which was not interfered with by 1-MT treatment. *p = 0.028 (Wilcoxon signed rank test). All data represent the means ± standard error (n = 6). 3d indicates 3 days of co-culture. Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 203
FIGURE 7. Transwell assays. Both suppressive effects of stromal cells on Th17 cells (a, b; p = 0.028, Wilcoxon signed rank test) and upregulatory effects on Treg cell frequencies (c, d; p = 0.028, Wilcoxon signed rank test) were maintained when the cells were co-cultured using transwells, suggesting the effect was not dependent on cell-to-cell contact. All data represent the means ± standard error (n = 6). 3d indicates 3 days of co-culture. S refers stromal cells.
!
2015 Informa Healthcare USA, Inc.
204 J. J. Lee et al. that keratocytes as well as macrophages/dendritic cells would be directly involved in this immunomodulatory function. Finally, we checked whether cell-to-cell competition had an effect on the suppressive function of the co-cultured cells. During the co-culture, corneal stromal cells could proliferate even in the RPMI media, and consequently might competitively affect CD4 T cells proliferation or apoptosis. We co-cultured skin fibroblasts as a control for 2 days on the same protocol that was described in Supplementary Figure S1 and evaluated the changes of Th17 cells (Supplementary Figure S4). Although, competitive cell co-culture showed a slight increase of T-cell death in both systems (the number of gated CD4 T cells was decreased in both system compared with that in Tcell-only culture), the suppressive effect on IL17 secretion was not affected by the skin fibroblast co-culture. This suggests that the immune modulatory effect of corneal stromal cells seems not to be caused by the cell-to-cell competition. The study was limited in that we could not directly observe intracellular staining of the stromal cells (keratocytes) for IL-10 or TGF-b, because the stimulation conditions of the keratocytes for intrastromal staining of IL-10 or TGF-b are not currently known. That’s why we looked into whether co-cultured T cells secreted IL-10 or TGF-b to reveal the cell source in Supplementary Figure S2. The indirect evidence that no significant intracellular staining of IL-10 or TGF-b in co-cultured T cells implies that a high concentration of IL-10 or TGF-b in co-cultured supernatants appears to be mainly secreted by co-cultured stromal cells. Although this study was conducted in vitro, this study is worthy of notice because it is the first report of which we are aware to present the immunomodulatory functions of human corneal stromal cells on Th17 cells. In conclusion, human corneal stromal cells show immunomodulatory properties by upregulating T regulatory cell frequency and by suppressing Th17 cell function via a TGFb-dependent pathway.
DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (MEST) (NRF 2010-0010629) and Grant from the Seoul National University Hospital Research Fund (number 04-2010-0850).
AUTHOR CONTRIBUTIONS Jong Joo Lee analyzed data and wrote the manuscript; Jin Suk Ryu and Hyun Jeong Jeong performed flow cytometry as well as ELISA; Mee Kum Kim designed the study, interpreted the data, provided grant funding, and made critical comments on the manuscript; and Won Ryang Wee supervised the project and provided critical comments to the manuscript.
REFERENCES 1. Niederkorn JY, Stein-Streilein J. History and physiology of immune privilege. Ocul Immunol Inflamm. 2010;18:19–23. 2. van Dooremaal JC. Die Entwicklung der in fremden Grund versetzten lebenden Geweba. A Graef Arch Ophthal. 1873; 19:358–373. 3. Medawar PB. Immunity to homologous grafted skin; the fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br J Exp Pathol. 1948;29:58–69. 4. Kaplan HJ, Streilein JW. Immune response to immunization via the anterior chamber of the eye, I: F. lymphocyteinduced immune deviation. J Immunol. 1977;118:809–814. 5. Holan V. Corneal stromal cells selectively inhibit the production of certain anti-inflammatory cytokines. Expert Rev Clin Immunol. 2006;2:101–108. 6. Ryu YH, Kim JC. Expression of indoleamine 2,3-dioxygenase in human corneal cells as a local immunosuppressive factor. Invest Ophthalmol Vis Sci. 2007;48:4148–4152. 7. Yang JW, Ham DS, Kim HW, et al. Expression of Stat3 and indoleamine 2,3-dioxygenase in cornea keratocytes as factor of ocular immune privilege. Graefe’s Arch Clin Exp Ophthalmol. 2012;250:25–31. 8. Randall TD. Stromal cells put the brakes on T-cell responses. Immunol Cell Biol. 2012;90:469–70. 9. Knickelbein JE, Divito S, Hendricks RL. Modulation of CD8+ CTL effector function by fibroblasts derived from the immunoprivileged cornea. Invest Ophthalmol Vis Sci. 2007; 48:2194–2202. 10. Donnelly JJ, Xi MS, Rockey JH. A soluble product of human corneal fibroblasts inhibits lymphocyte activation: enhancement by interferon-gamma. Exp Eye Res. 1993;56: 157–165. 11. Lukacs-Kornek V, Malhotra D, Fletcher AL, et al. Regulated release of nitric oxide by nonhematopoietic stroma controls expansion of the activated T cell pool in lymph nodes. Nature Immunol. 2011;12:1096–1104. 12. Bouffi C, Bony C, Jorgensen C, Noel D. Skin fibroblasts are potent suppressors of inflammation in experimental arthritis. Ann Rheum Dis. 2011;70:1671–1676. 13. Buettner M, Pabst R, Bode U. Lymph node stromal cells strongly influence immune response suppression. Eur J Immunol. 2011;41:624–633. 14. Zhang X, Volpe EA, Gandhi NB, et al. NK cells promote Th-17 mediated corneal barrier disruption in dry eye. PloS One. 2012;7:e36822. 15. Cao Q, Wang Y, Zheng D, et al. IL-10/TGF-beta-modified macrophages induce regulatory T cells and protect against adriamycin nephrosis. J Am Soc Nephrol. 2010;21:933–942.
Ocular Immunology & Inflammation
Copyright of Ocular Immunology & Inflammation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
In Vitro Immunomodulatory Effects of Human Corneal Stromal Cells on Human Th17 Cells* Jong Joo Lee, MD1,2, Jin Suk Ryu, MS2, Hyun Jeong Jeong, MS2, Mee Kum Kim, MD, PhD1,2, and Won Ryang Wee, MD, PhD1,2 1
Department of Ophthalmology, Seoul National University College of Medicine, Seoul, Korea; and Laboratory of Ocular Regenerative Medicine and Immunology, Seoul Artificial Eye Center, Seoul National University Hospital Clinical Research Institute, Seoul, Korea and 2Laboratory of Ocular Regenerative Medicine and Immunology, Seoul Artificial Eye Center, Seoul National University Hospital Clinical Research Institute, Seoul, Korea
ABSTRACT Purpose: To investigate the immunomodulatory effects of human corneal stromal cells on the function of Th17 cells, in vitro. Methods: Th17 cells were co-cultured with corneal stromal cells with or without 1-methyltryptophan, 1400w dihydrochloride, interleukin (IL)-10 inhibitor, or transforming growth factor-beta (TGF-b) inhibitor. The secretion of IL-17A by Th17 cells and the upregulation of T regulatory (Treg) cell frequencies were evaluated. Transwell assays were conducted to evaluate whether the immunomodulatory effects of stromal cells were dependent on cell-to-cell contact. Results: Corneal stromal cells suppressed the secretive function of Th17 cells and upregulated Treg cell frequencies, enhancing TGF-b. Tranwell assays showed similar immunomodulatory effects of the stromal cells on the function of Th17 cells. TGF-b inhibitors attenuated the suppressive effects of the stromal cells on the function of Th17 cells, whereas other inhibitors didn’t. Conclusion: It which indicates that human corneal stromal cells suppress the function of Th17 cells through a TGF-b-dependent pathway. Keywords: Cornea, IL-17, stromal cells, T cells, T regulatory cells, TGF-b, Th17 cells
Immune privilege, a part of immune homeostasis, is a dynamic process involving multiple physiological, anatomical, and immunological properties that allow organs to perform their primary function with less inflammation. The brain, testis, and the eye have evolved into immune-privileged organs to save their critical functions, including control of homeostasis, reproduction, and vision, while the gut and lung have evolved into immune-modulatory organs to ensure their functionality in a foreign-antigen-enriched environment.1 The immune privilege of the eye was first recognized over 130 years ago2; the antigen
sequestration concept of the anterior chamber was first proposed 80 years later by Medawar.3 Thereon, extensive investigation has revealed the existence of anterior chamber associated immune deviation (ACAID).4 On the other hand, current studies have shed light on the possibility that the corneal stromal cells might also contribute to the immune privilege in the eye, and that other stromal cells residing in lymphoid organs affect immunohomeostasis.5–10 However, the immune modulatory role of the corneal stromal cells remains unclear. This study addresses whether corneal stromal cells suppress the function of Th17 cells. We also
*Critical comments were kindly provided by Su Kil Seo (Department of Ophthalmology, Busan Paik Hospital, Inje University Medical College, Busan, Korea) in measuring and interpreting the IDO-dependent pathway, and Won Woo Lee (Department of Microbiology and Immunology, Seoul National University College of Medicine, Seoul, Korea) kindly guided us to install Th17 cell-driving conditions. Received 7 June 2013; revised 8 January 2014; accepted 3 February 2014; published online 5 March 2014 Correspondence: Mee Kum Kim, MD, PhD, Department of Ophthalmology, Seoul National University College of Medicine, 103 Daehang-ro, Jongno-gu, Seoul, 110-799, Republic of Korea. E-mail: [email protected]
194
Human Corneal Stromal Cells and Human Th17 Cells 195 investigate which pathway is involved in the modulation of T cells.
MATERIALS AND METHODS The study was approved by the institutional review board of the Seoul National University Hospital (IRB approval number: H-1007-099-324) and was performed in accordance with the guidelines of the Code of Ethics of the World Medical Association (Declaration of Helsinki). Human blood was collected from healthy donor volunteers with written informed consent.
CD4+ T-cell Isolation and Th17 Differentiation CD4+ human peripheral blood T cells were isolated from freshly collected, heparinized peripheral blood using CD4 magnetic activated cell sorting (MACS) beads (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s protocols. Th17 cells were differentiated by stimulating 5 104 CD4+ T cells on a plate coated with 1 mg/mL anti-CD3 or anti-CD28 monoclonal antibody (mAb; eBioscience, San Diego, CA, USA) and subsequently culturing with 25 ng/mL IL-1b (R&D Systems, Minneapolis, USA) and 25 ng/mL IL-23 (R&D Systems) in RPMI 1640 medium (Welgen, Daegu, Korea) containing 10% fetal bovine serum (FBS), penicillin/streptomycin (100 U/mL), and 10 mM HEPES in a 48-well plate (NUNC, Roskilde, Denmark) for 7 days.
In vitro Suppression of Proliferation To examine the suppressive effects of corneal stromal cells on the proliferation of CD4+ T cells, isolated CD4+ T cells were stained with 5 mM carboxyfluorescein diacetate succinimidyl ester (CFSE; Invitrogen, Carlsbad, CA, USA). Anti-CD3/CD28 antibodies (1 mg/mL; eBioscience) were resuspended in the supplemented RPMI 1640 medium (Welgen) to stimulate cell proliferation. The T cells stimulated by antiCD3/CD28 antibodies served as positive control, while unstimulated T cells served as negative control. CD4+ T cells were placed in co-culture with stromal cells at a ratio of 10:1 (5 104 T cells: 5 103 stromal cells) and were co-cultured for 5 days in the supplemented RPMI 1640 medium. Suppression of T-cell proliferation was measured by reduction in the CFSE concentration by flow cytometry using FACS Vantage Flow Cytometer (BD Biosciences, San Jose, CA, USA). Reduction rate (%) was calculated by subtracting the proliferation percentage of the stromal cell-treated T cells (treated %) from the proliferation percentage of positive controlled T cells (PC %), with the value being divided by the proliferation percentage of positive controlled T cells and then by multiplying the resulting value by 100 (reduction rate % = [PC % – treated %] 7 PC% 100) A total of 250 mM 1-methyl-d,l-Trp; (1-D/L MT; Sigma Aldrich, St. Louis, MO, USA) was applied to assess whether suppression was diminished in an indoleamine 2,3-dioxygenase (IDO)-dependent manner. The experiment used eight sets of samples.
Corneal Stromal Cell Isolation and Culture
In vitro Functional Assays for the Suppression of Th17 Cells
Human corneas obtained from the Northwest Lions Eye Bank (Seattle, WA, USA) were used for experiments after the central corneal button had been transplanted to human subjects. The epithelial layers were removed using a surgical blade and the Descemet membranes were peeled off mechanically. The remaining corneal stroma was treated using 1.2 U/mL dispase II (Cat. No. 04 942 078 001; Roche Applied Science, Basel, Switzerland) and 200 U/mL type I collagenase (Worthington, Lakewood, NJ, USA). The cells were harvested by centrifugation at 1200 rpm for 5 min and inoculated into the culture plate with DMEM/F12 (Cat. No. 12-719F; Lonza, Valais, Switzerland) containing 10% fetal bovine serum (Welgen), penicillin/streptomycin (100 U/mL), 3.151 g/L glucose, L-glutamine, and 15 mM HEPES. The stromal cells were primarily cultured at 37 C in a carbon dioxide incubator for 1–2 weeks and were passaged 2–5 times. The cells were harvested and seeded (5 103/well) in a 48-well plate (NUNC) for co-culture with CD4+ T cells.
CD4+ T cells that had been treated with IL-1b and IL-23 for Th17 differentiation were co-cultured with stromal cells at a ratio of 10:1 (5 104 T cells: 5 103 stromal cells) for 2 days in the supplemented RPMI 1640 medium after stimulation. Anti-CD3/CD28 antibodies (1 mg/mL; eBioscience) were resuspended in the supplemented RPMI 1640 medium (Welgen) to stimulate cell proliferation. Suspended Th17 cells were separated from the stromal cells, and isolated Th17 cells were subsequently cultured for an additional 5 days until 7 days (Supplementary Figure S1). Control Th17 cells were cultured for 7 days in the same condition without co-culture. A total of 250 mM 1-d/l MT or 1 mM/mL 1400w dihydrochloride was applied during the co-culture of IFN-g (1000U/mL)-treated stromal cells to assess whether the suppression was in a IDO- or nitric oxide synthase 2 (NOS2)-dependent manner. After 7 days, T cells were incubated with CD4-fluorescein isothiocynate (FITC) (eBioscience) and CD45RO-PerCP (eBioscience) for 30 min at 4 C, and the cells were
!
2015 Informa Healthcare USA, Inc.
196 J. J. Lee et al. washed twice. After treatment with phorbol 12-myristate 13-acetate (PMA; 50 ng/mL; Sigma-Aldrich) and ionomycin (1 mg/mL; Sigma-Aldrich), the protein transporter inhibitor monensin (Golgi Plug 1 mL/mL, BD Phamingen, San Diego, USA) was treated simultaneously. Intracellular staining was performed after 5 h of stimulation. The cells were fixed with a fixation/permeabilization solution for 20 min at 4 C. After washing with the Perm/Wash buffer, the cells were incubated with IL-17-PE (BD Phamingen) and IFN-g-APC (BD Phamingen) for 60 min at 4 C. Flow cytometric analysis was performed using FACS Canto (Becton Dickinson, San Diego, CA, USA), and data were analyzed by FlowJo Version 7.6.5 (Treestar, Ashland, OR, USA). The experiments were performed in 4 replicates. The levels of IL-17 and IFN-g in supernatants from the T-cell culture after removal of stromal cells were also measured. Commercial kits for enzyme-linked immunosorbent assay (ELISA) were used as per the manufacturers’ protocols (DuoSet ELISA; R&D Systems, Minneapolis, MN, USA). The experiments were conducted in 4 replicates. To explore whether the stromal cells suppress Th17 cells in a cell-to-cell contact-dependent manner or not, transwell assays of the co-cultures were conducted using the same protocol mentioned above with Millicell culture plate inserts (0.4 mm, Billerica, MA, USA). Cell surface staining with fluorescein isothiocyanate (FITC)-conjugated anti-CD4 (eBioscience) and PerCP-conjugated anti-CD45RO (eBioscience) antibodies, and intracellular staining with PE-conjugated anti-IL-17 (BD Pharmingen) antibodies was performed in the same manner as mentioned above. The experiments were repeated in 6 replicates. AS101 (2.5 mg/mL, IL-10 inhibitors; Tocris Bioscience, Ellisville, MO, USA) or 1 mM/mL SB 431542 (TGF-b inhibitors; Tocris Bioscience, Ellisville, MO, USA) was applied during the culture to assess whether the suppression was in an IL-10- or TGFb-dependent manner. Cell surface staining with fluorescein isothiocyanate (FITC)-conjugated anti-CD4 (eBioscience) and PerCP-conjugated anti-CD45RO (eBioscience) antibodies, and intracellular staining with PE-conjugated anti-IL-17 (BD Pharmingen) antibodies was performed in the same manner as mentioned above. The experiments were performed in 6 replicates.
The Change of TGF-b and IL-10 during Co-cultures The levels TGF-b and IL-10 were determined in supernatants from 3-day co-cultures of both CD4+ T cells and stromal cells using commercial DuoSet ELISA kits following the manufacturers’ protocols (R&D Systems). The concentration used for the
standard curve ranged from 1000 to 15.625 pg/mL, using human recombinant TGFb, and IL-10 (R&D Systems). The experiments were performed in 11 sets. To explore whether TGF-b and IL-10 were produced by either corneal stromal cells or T cells, harvested T cells in pellets after co-culture were analyzed by flow cytometry. Cells were stimulated for 6 h with PMA (10 ng/mL) and inomycin (1 mg/mL) and were treated simultaneously by the protein transporter inhibitor cocktail (Brefeldin A 5.3 mM + Monensin 1 mM, 2 mL/mL, eBioscience). Cell surface staining was performed with CD4 PE-cy7 (e-Bioscience) and the intracellular staining was performed using LAP (TGF-b) PerCP-cy5.5 (e-Bioscience) and IL-10–FITC (e-Bioscience).
In Vitro Change of CD4+CD25hiFOXP3hi T Regulatory Cells in Co-cultures CD4+ T cells stimulated with anti-CD3/CD28 antibodies (1 mg/mL; eBioscience) were co-cultured with stromal cells at a ratio of 10:1 (5 104 T cells: 5 103 stromal cells) for 3 days. To evaluate the changes of CD4+CD25hiFOXP3hi cells, flow cytometry was conducted after isolation of T cells. Harvested CD4+ T cells were incubated with CD4 PE-cy7 (e-bioscience) and CD25 APC (eBioscience) for 30 min at 4 C, and the cells were washed twice. The cells were fixed with a fixation/permeabilization solution for 60 min at 4 C. After washing with the Perm/Wash buffer, the cells were incubated with FOXP3-PE (BD Phamingen) for 60 min at 4 C. PE-conjugated isotype Ig G(eBioscience) was stained as isotype control. Flow cytometric analysis and data analysis were performed as described above. The experiments were performed using 6 replicates. To check IDO dependency, 1-d/l MT was applied (250 mM) to assess whether the changes of T regulatory cells would be in an IDO-dependent manner. To investigate whether stromal cells upregulate Treg cells in a cell-to-cell contact-dependent manner or not, transwell assays of the co-cultures were conducted using the same protocol mentioned above with Millicell culture plate inserts (0.4 mm). The procedures of the surface staining with CD4 PE-cy7 (e-Bioscience) and CD25-APC (eBioscience) and the intracellular staining with FOXP3-PE (BD Phamingen) were done as mentioned above. The experiments were conducted in 6 replicates.
Statistical Analysis SPSS 17.0 software (SPSS, Chicago, IL, USA) was used. The proportions of proliferating cells or IL-17/ IFN-g secreting cells, changes of T regulatory Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 197 cells, or the changes of the other cytokines were compared to those of the positive control using the nonparametric Wilcoxon signed rank test. The p values 50.05 were considered to be statistically significant.
RESULTS Corneal Stromal Cells Hinder CD4+ T-cell Proliferation Stromal cells were co-cultured with CD4+ T cells at a ratio of 1:1 to 1:100. Ratios of 1:1 to 1:10 produced CD4+ T-cell suppression (Figure 1a). Thereafter, experiments were done at a ratio of 1:10. The proliferation of CD4+ T cells was effectively reduced by 62.6% when co-cultured with corneal stromal cells (p = 0.012, vs. positive control, Wilcoxon signed rank test; Figure 1b). When stromal cells were treated with IFN-g to induce IDO,6 the inhibitory effect was similar to the effect of nontreated stromal cells (Figures 1b) on T cells. The suppressive effect of the stromal cells was not significantly different depending on the passage number (Figure 1c).
Corneal Stromal Cells Suppress IFN-c and IL-17 Secretion by Th17 Cells via the TGF-b Pathway Corneal stromal cells significantly suppressed the secretion of both IFN-g and IL-17 by Th17 cells, which were gated by CD4hi and CD45ROhi (p = 0.043, respectively, vs. positive control, Wilcoxon signed rank test; Figure 2a and b), which was further confirmed by the overall low levels of IFN-g and IL-17 in the supernatants from isolated Th17 cells after stromal cells had been removed (p = 0.000, respectively, vs. positive control, Wilcoxon signed rank test; Figure 2c). 1-MT or 1400w treatment did not abrogate the effect of suppression of stromal cells on both proliferation of CD4+ T cells and IL-17/IFN-g secretion by Th17 cells (Figure 3). On the other hand, the TGF-b inhibitor attenuated the effect of stromal cells on the secretion of IL-17 by Th17 cells (p = 0.028, Wilcoxon signed rank test), whereas the IL-10 inhibitor did not interfere with these effects (Figure 4).
Anti-inflammatory Cytokines Increase in the Co-cultured Supernatants The anti-inflammatory cytokines TGF-b and IL-10 were also significantly elevated during the co-culture (p = 0.000–0.016, vs. control (CD4+ T cells or stromal cells alone, Mann-Whitney U test; Figure 5). As cell !
2015 Informa Healthcare USA, Inc.
sources of TGF-b and IL-10, a few CD4+ T cells secreted TGF-b. On the other hand, CD4+ T cells did not secrete IL-10 at all (Supplementary Figure S2), suggesting that most of the TGF-b and IL-10 might be produced by corneal stromal cells.
Stromal Cells Upregulate CD4+CD25hiFOXP3hi T Regulatory Cells Upregulation of T regulatory cells by co-cultured stromal cells was noted, which was statistically significant (p = 0.028, Wilcoxon signed rank test; Figure 6). IFN-g-treated stromal cells also enhanced T regulatory cells (p = 0.028, Wilcoxon signed rank test). 1-MT treatment did not attenuate the upregulation of CD4+CD25hiFOXP3hi T regulatory cells.
Stromal Cells Modulate Th17 Cells and Treg Cells through the Paracrine Pathway The suppressive effects of stromal cells on Th17 cell function were maintained when the cells were co-cultured using transwells, suggesting that the effect was not dependent on a cell-to-cell contact (Figure 7a and b; p = 0.028, Wilcoxon signed rank test). In addition, co-culturing stromal cells still resulted in an upregulation of Treg cell frequencies, suggesting that the upregulatory effect was not dependent on cell-to-cell contact (Figure 7c and d; p = 0.028, Wilcoxon signed rank test).
DISCUSSION This study demonstrated the immunomodulatory effects of human corneal stromal cells on the suppression of Th17 cells and upregulation of T regulatory cells along with the increase of TGF-b, in vitro. TGF-b may prefer to induce T regulatory cells by interfering with Th17 cell differentiation. Since mesenchymal stem cells have revealed their immune suppressive functions through various disease models in many research papers, the other stromal cells of mesenchymal origin have been currently drawing attention in their immune modulatory roles in homeostasis.8,11–13 Along with previous studies,5–10,12,13 our study suggests that tissue stromal cells of the eyes also seem to have an immune-modulatory capacity. However, unlike fibroblastic reticular cells (FRCs) in previous reports, we found that corneal stromal cells suppressed the effector function of CD4 T cells by non-IDO- or NOSdependent pathways. Besides, the effect of corneal stromal cells on the T cells appeared to be less potent than the effect of FRCs, which was reported by Veronika et al.11 FRCs suppressed T cells in a ratio of
198 J. J. Lee et al.
FIGURE 1. Suppression of CD4+ T-cell proliferation by stromal cells. (a) Suppressive function of corneal stromal cells on CD4+ T cells are attenuated using incrementally different ratios of stromal cells to CD4+ T cells. (b) The proliferation of CD4+ T cells was effectively reduced when they were in contact with stromal cells regardless of IFNg pretreatment in the stromal cells. *p = 0.012 (Wilcoxon signed rank test). (c) Reduction rate did not change according to the passage of stromal cells. All three passages of stromal cells caused significant reduction of CD4+ T-cell proliferation compared to positive control (p = 0.014, Mann-Whitney U test). All data represent the means ± standard error (n = 8).
1:100 (FRCs:T) while, our data showed that corneal stromal cells attenuated T cells in a ratio of 1:10 (stromal cells:T). It might be plausible that stromal cells would have preferential functions depending on residential area. Lymph nodes have a few FRCs and millions of T cells, on the contrary, the cornea has lots of stromal cells and a few T cells. The FRCs in the lymph nodes are supposed to encounter lots of T cells, which need to be tightly controlled for homeostasis in normal conditions. Meanwhile corneal stromal cells are expected to encounter some surveilling T cells that are occasionally found in the cornea and conjunctiva.14 It suggests that even less potent control of stromal cells might possibly be enough to maintain
homeostasis in the cornea rather than in lymph nodes in view of the ratio between the stromal cells and T cells. On the other hand, most of the reports revealed that IFN-g potentiated the immune modulatory functions of tissue stromal cells, presumably by the IFN-g-enhanced expression of IDO, NOS 2, or arginase.8,11 The reason we focused on the effect of IDO in stromal cells in the first place is that we found high expression of IDO in corneal stromal cells in RT-PCR (data not shown), as well as profound increase of IDO with IFN-g treatment. Meanwhile, our data presented similar suppressive effects of stromal cells, regardless of treatment of IFN-g and no Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 199
FIGURE 2. Suppression of Th17 cells by stromal cells. (a) Stromal cells suppressed IL-17/IFN-g secreting CD4 T cells. Results represent one of four separate experiments with Th17 cells derived by IL-1b/IL-23 and aCD3/aCD28, which were co-cultured with stromal cells by the protocol described by supplementary Figure S1. (b) The suppressive effect of stromal cells on IFN-g and IL-17 secretion by CD4hiCD45ROhi memory Th17 cells was statistically significant compared to positive control (p = 0.043, Wilcoxon signed rank test). (c) ELISA presented low levels of both IFN-g and IL-17 in the supernatants from isolated CD4 T cells after stromal cells had been removed. *p = 0.000 (Mann-Whitney U test). All data represent the means ± standard error (n = 4).
interference by IDO or NOS2 inhibitors. They suggested that the inhibitory effect of corneal stromal cells was less likely to be dependent on IFN-g. As flow cytometric analysis revealed that only a few T cells secreted TGF-b, stromal cells appeared !
2015 Informa Healthcare USA, Inc.
to produce most of the TGF-b, which seemed to be higher than the TGF-b production in mesenchymal stem cells (data not shown). Considering that the eye is composed of immune-privileged tissues with lots of immune-suppressive molecules, including TGF-b,
200 J. J. Lee et al.
FIGURE 3. The effect of IDO inhibitors and NOS2 inhibitors on the suppressive functions of stromal cells. 1-MT treatment did not attenuate for stromal cells to inhibit CD4+ T-cell proliferation (a, b) or the secretion of both IFN-g and IL-17 by Th17 cells (c, d). CD4 T means CD4+T cells, 1-MT stands for 1-methyl-d,l-Trp, and S(IFNg) refers to stromal cells pretreated with IFN-g. aCD3/aCD28 and IL-1b/IL-23 preapplied Th17 cells with treatment of 1-MT are used as a positive control to exclude the direct effect of 1-MT. (E, F) To analyze whether the suppressive effect of corneal stromal cells on T-cell function is blocked by inhibition of NOS2, 1 mM/mL 1400w dihydrochloride was treated during the co-culture. The proliferation of CD4+ T cells or the secretory function of IFN-g/IL-17 by Th17 cells was not largely affected by inhibition of NOS2. All data represent the means ± standard error (n = 4). 2d indicates 2 days of co-culture. Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 201
Figure 3. Continued.
FIGURE 4. The effect of TGF-b or IL-10 on the suppressive function of stromal cells. The TGF-b inhibitor (SB 431542) attenuated the effect of stromal cells on the secretion of IL-17 by Th17 cells (p = 0.028, Wilcoxon signed rank test), whereas IL-10 inhibitors (AS101) did not interfere with these effects. aCD3/aCD28 and IL-1b/IL-23 applied to the CD4+ T cells for the differentiation of Th17 cells. All data represent the means ± standard error (n = 6). 3d indicates 3 days of co-culture.
and the corneal stromal cells are one of the major cell sources of TGF-b secretion, TGF-b-associated immune suppression of corneal stromal cells on T cells would make sense. Furthermore, our data corresponded well with the fact that TGF-b could move the balance of the Th17–Treg cell axis toward the induction of T regulatory cells by diminishing Th17 cells. The initial co-culture showed not only a strong upregulation of !
2015 Informa Healthcare USA, Inc.
anti-inflammatory cytokines such as TGF-b and IL-10 but also an increase of proinflammatory cytokines like IFN-g and IL-6 (data not shown). However, the antiinflammatory effect of TGF-b appears to overcome the effect of proinflammatory cytokines by modulating the Th17–Treg axis over time, resulting in an overall decrease of IFN-g and IL-17 as the culture progressed. Although the main Th17-driving factors in mice are IL-6 and TGF-b, human Th17 cells are mainly induced by IL-1b and IL-23 and not by IL-6 and TGF-b. It is plausible that even high IL-6 and TGF-b combination did not increase percentages of Th17 cells in our human cell study. Our data implies that stromal cells may induce Treg cells by modulating the Th17–Treg balance. However, it should be further evaluated whether corneal stromal cells promote the function of Treg cells or induce Treg differentiation in future studies. IL-10 was not produced by CD4+ T cells in our study. Therefore, our data did not support that IL-10 might be involved in promoting T regulatory cells (data not shown) on contrast to previous reports.15 In addition, to rule out the argument about the confounding effect due to heterogenic cells mixed with the stromal cells (we found 1–10% of CD11b+ or CD11c+ cells were mixed in the cultured corneal stromal cells), we sorted out CD11b+ or CD11c+ cells through flow cytometry (Supplementary Figure S3). Nevertheless, we still observed almost the same inhibitory functions of the keratocytes on the secretion of IFN-g and IL-17A by Th17 cells. The data suggested
202 J. J. Lee et al.
FIGURE 5. Change of the anti-inflammatory cytokines during co-culture. ELISA revealed markedly increased secretion of TGF-b (a) and IL-10 (b) in 3 days co-cultured supernatants of both corneal stromal cells and CD4+ T cells. All data represent the means + standard error (n = 11). S indicates stromal cells, and CD4 T indicates CD4+ T cells. *p = 0.000, yp = 0.001, xp = 0.016 (Mann-Whitney U test).
FIGURE 6. Upregulation of T regulatory cells with co-culture of stromal cells. (a) Results represent one of six separate experiments. CD4 T indicates CD4+ T cells. 3d indicates 3 days of co-culture. (b) A significant increase of T regulatory cells was observed in the presence of stromal cells regardless of IFN-g treatment, which was not interfered with by 1-MT treatment. *p = 0.028 (Wilcoxon signed rank test). All data represent the means ± standard error (n = 6). 3d indicates 3 days of co-culture. Ocular Immunology & Inflammation
Human Corneal Stromal Cells and Human Th17 Cells 203
FIGURE 7. Transwell assays. Both suppressive effects of stromal cells on Th17 cells (a, b; p = 0.028, Wilcoxon signed rank test) and upregulatory effects on Treg cell frequencies (c, d; p = 0.028, Wilcoxon signed rank test) were maintained when the cells were co-cultured using transwells, suggesting the effect was not dependent on cell-to-cell contact. All data represent the means ± standard error (n = 6). 3d indicates 3 days of co-culture. S refers stromal cells.
!
2015 Informa Healthcare USA, Inc.
204 J. J. Lee et al. that keratocytes as well as macrophages/dendritic cells would be directly involved in this immunomodulatory function. Finally, we checked whether cell-to-cell competition had an effect on the suppressive function of the co-cultured cells. During the co-culture, corneal stromal cells could proliferate even in the RPMI media, and consequently might competitively affect CD4 T cells proliferation or apoptosis. We co-cultured skin fibroblasts as a control for 2 days on the same protocol that was described in Supplementary Figure S1 and evaluated the changes of Th17 cells (Supplementary Figure S4). Although, competitive cell co-culture showed a slight increase of T-cell death in both systems (the number of gated CD4 T cells was decreased in both system compared with that in Tcell-only culture), the suppressive effect on IL17 secretion was not affected by the skin fibroblast co-culture. This suggests that the immune modulatory effect of corneal stromal cells seems not to be caused by the cell-to-cell competition. The study was limited in that we could not directly observe intracellular staining of the stromal cells (keratocytes) for IL-10 or TGF-b, because the stimulation conditions of the keratocytes for intrastromal staining of IL-10 or TGF-b are not currently known. That’s why we looked into whether co-cultured T cells secreted IL-10 or TGF-b to reveal the cell source in Supplementary Figure S2. The indirect evidence that no significant intracellular staining of IL-10 or TGF-b in co-cultured T cells implies that a high concentration of IL-10 or TGF-b in co-cultured supernatants appears to be mainly secreted by co-cultured stromal cells. Although this study was conducted in vitro, this study is worthy of notice because it is the first report of which we are aware to present the immunomodulatory functions of human corneal stromal cells on Th17 cells. In conclusion, human corneal stromal cells show immunomodulatory properties by upregulating T regulatory cell frequency and by suppressing Th17 cell function via a TGFb-dependent pathway.
DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (MEST) (NRF 2010-0010629) and Grant from the Seoul National University Hospital Research Fund (number 04-2010-0850).
AUTHOR CONTRIBUTIONS Jong Joo Lee analyzed data and wrote the manuscript; Jin Suk Ryu and Hyun Jeong Jeong performed flow cytometry as well as ELISA; Mee Kum Kim designed the study, interpreted the data, provided grant funding, and made critical comments on the manuscript; and Won Ryang Wee supervised the project and provided critical comments to the manuscript.
REFERENCES 1. Niederkorn JY, Stein-Streilein J. History and physiology of immune privilege. Ocul Immunol Inflamm. 2010;18:19–23. 2. van Dooremaal JC. Die Entwicklung der in fremden Grund versetzten lebenden Geweba. A Graef Arch Ophthal. 1873; 19:358–373. 3. Medawar PB. Immunity to homologous grafted skin; the fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br J Exp Pathol. 1948;29:58–69. 4. Kaplan HJ, Streilein JW. Immune response to immunization via the anterior chamber of the eye, I: F. lymphocyteinduced immune deviation. J Immunol. 1977;118:809–814. 5. Holan V. Corneal stromal cells selectively inhibit the production of certain anti-inflammatory cytokines. Expert Rev Clin Immunol. 2006;2:101–108. 6. Ryu YH, Kim JC. Expression of indoleamine 2,3-dioxygenase in human corneal cells as a local immunosuppressive factor. Invest Ophthalmol Vis Sci. 2007;48:4148–4152. 7. Yang JW, Ham DS, Kim HW, et al. Expression of Stat3 and indoleamine 2,3-dioxygenase in cornea keratocytes as factor of ocular immune privilege. Graefe’s Arch Clin Exp Ophthalmol. 2012;250:25–31. 8. Randall TD. Stromal cells put the brakes on T-cell responses. Immunol Cell Biol. 2012;90:469–70. 9. Knickelbein JE, Divito S, Hendricks RL. Modulation of CD8+ CTL effector function by fibroblasts derived from the immunoprivileged cornea. Invest Ophthalmol Vis Sci. 2007; 48:2194–2202. 10. Donnelly JJ, Xi MS, Rockey JH. A soluble product of human corneal fibroblasts inhibits lymphocyte activation: enhancement by interferon-gamma. Exp Eye Res. 1993;56: 157–165. 11. Lukacs-Kornek V, Malhotra D, Fletcher AL, et al. Regulated release of nitric oxide by nonhematopoietic stroma controls expansion of the activated T cell pool in lymph nodes. Nature Immunol. 2011;12:1096–1104. 12. Bouffi C, Bony C, Jorgensen C, Noel D. Skin fibroblasts are potent suppressors of inflammation in experimental arthritis. Ann Rheum Dis. 2011;70:1671–1676. 13. Buettner M, Pabst R, Bode U. Lymph node stromal cells strongly influence immune response suppression. Eur J Immunol. 2011;41:624–633. 14. Zhang X, Volpe EA, Gandhi NB, et al. NK cells promote Th-17 mediated corneal barrier disruption in dry eye. PloS One. 2012;7:e36822. 15. Cao Q, Wang Y, Zheng D, et al. IL-10/TGF-beta-modified macrophages induce regulatory T cells and protect against adriamycin nephrosis. J Am Soc Nephrol. 2010;21:933–942.
Ocular Immunology & Inflammation
Copyright of Ocular Immunology & Inflammation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
