Interleukin-1β Increases Baseline Expression and Secretion of Interleukin-6 by Human Uveal Melanocytes In Vitro via the p38 MAPK/NF-κB Pathway

Abstract

Purpose.: Melanocyte is the major cell component in the uvea. Interleukin (IL)-6 is a proinflammatory cytokine. The authors studied the expression of IL-6 in cultured human uveal melanocytes (UMs) and their modulation by IL-1β and explored the relevant signal pathways.

Methods.: Conditioned media and cells were collected from UMs cultured in medium with and without serum and were stimulated by IL-1β. IL-6 protein and transcript were measured using an ELISA kit and RT-PCR, respectively. NF-κB in nuclear extracts and phosphorylated p38 MAPK, ERK, and JNK in cells cultured with and without IL-1β were measured by ELISA kits. Inhibitors of p38 (SB203580), ERK (UO1026), JNK (SP600125), and NF-κB (BAY11–7082) were added to the cultures to evaluate their effects.

Results.: Low levels of IL-6 protein were detected in the conditioned medium in UMs cultured without serum. The addition of serum increased the secretion of IL-6. IL-1β (0.1–10 ng/mL) increased IL-6 transcript and protein levels in a dose- and time-dependent manner up to sixfold, accompanied by a significant increase of NF-κB in nuclear extracts and phosphorylated p38 MAPK in cell lysates. NF-κB and p38 inhibitors alone decreased, whereas a combination of these two inhibitors completely abolished the IL-1β–induced expression of IL-6.

Conclusions.: This is the first report on the expression and secretion of IL-6 by UMs. IL-1β augments IL-6 production from the melanocytes. The data suggest that UMs may play a role in the pathogenesis of ocular inflammatory and autoimmune diseases.

Uveal melanocytes (UMs) are the major cell population in the uveal tract. Little had been known about the function of UMs and their role in the pathogenesis of various ocular diseases. In the past few decades, our development of procedures for the isolation and culture of human UMs has allowed the establishment of pure cell lines of UMs from donor eyes and has provided an in vitro model for studying their cell biology. ^{1 –5} We have demonstrated that cultured human UMs are capable of the production of eumelanin and pheomelanin, matrix metalloproteinases, tissue plasminogen activator, and vascular endothelial growth factor in vitro. ^{6 –9} However, the production of cytokines by UMs has not been studied previously. 

Cytokines are regulatory proteins secreted by various cells, particularly cells involved in immune responses; they carry signals locally between cells and thus affect the behavior of other cells. Cytokines play an essential role in maintaining normal physiological states and in the modulation of immune responses and inflammatory processes. ^{10 –14} Cytokines have been classified as lymphokines (produced by lymphocytes), interleukins (secreted primarily by leukocytes), and chemokines (small cytokines of 8 to 10 kDa) based on their molecular mass, function, and cell of secretion or action. ^{10 –14}  

Interleukin (IL)-6, a pleiotropic cytokine, is produced by leukocytes and various cells, including several ocular cell types such as retinal pigment epithelial cells. ^{15 –20} It amplifies immune and inflammatory responses and plays a critical role in the occurrence of autoimmune diseases. ^{11 –14} Increased IL-6 levels in the aqueous humor, vitreous, and ocular tissues has been detected in certain ocular inflammatory and autoimmune diseases. ^{21 –28}  

IL-1 refers to a group of cytokines. The IL-1 superfamily comprises 11 members: IL-1α, IL-1β, IL-1R antagonist (IL-1Ra), IL-18 (IL-1F4), IL-1F5 to IL-1F10, and IL-33 (IL-1F11). The four primary members of the IL-1 superfamily are IL-1α, IL-1β, IL-18, and IL-33. ^10,29 IL-1β is a proinflammatory cytokine with the ability to trigger numerous inflammatory responses, one of which is the stimulation of expression of IL-6 in various cell types, including the retinal pigment epithelium (RPE). ^{15 –17,30} IL-1 has been detected in ocular fluids and tissues in ocular inflammatory diseases. ^27,28 IL-1β intravitreal injection caused a breakdown of the blood-retinal barrier and intraocular inflammation in Lewis rats. ^28,31  

The present study investigated the expression and secretion of IL-6 by UMs under various circumstances, including cultured with and without serum, and compared these with the expression of IL-6 by RPE. This study also investigated the effects of IL-1β on the expression and secretion of IL-6 by UMs and the relevant signal pathways involved in this process. 

Methods

Cell Culture

Human UMs were isolated from the iris, ciliary body, or choroid of adult donor eyes, as previously described. ^{1 –5} The isolated UMs were cultured with FIC medium, which is F12 medium supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 20 ng/mL bFGF, 0.1 mM isobutyl methylxanthine (IBMX), 10 ng/mL cholera toxin, and 50 μg/mL gentamicin (FBS and F12 medium; Gibco, Invitrogen, Carlsbad, CA; all others from Sigma, St. Louis, MO). ¹ After confluence was reached, the UMs were detached using trypsin-EDTA solution (Sigma), diluted 1:3 to 1:6, and were subcultured. The purity of the cell lines was demonstrated by immunocytochemical methods. UMs display S-100 antigen but not cytokeratin, whereas retinal pigment epithelial cells display both proteins; fibroblasts display neither of these proteins (all antibodies and related reagents were from Dako, Carpinteria, CA). ¹ Six UM cell lines (two from the iris, one from the ciliary body, and three from the choroid) were used in the present study. One primary culture of a human retinal pigment epithelial line with known IL-6 production activity was tested for comparison. Human RPE was isolated and cultured as described previously. ³² Cells were grown in Dulbecco's modified Eagle's medium (Gibco) supplemented with 10% FBS. Cell cultures in the third passage were used. 

Immunocytochemical Studies

Methods for immunocytochemical studies were described by us previously. ¹ The antibodies used consisted of a monoclonal mouse anti–human cytokeratin antibody (for keratins 6 and 18), a rabbit immunoglobulin to bovine S-100, a mouse anti–human α-smooth muscle actin, and a mouse anti–human IL-1R1 antibody. DAB (3,3′-diaminobenzidine) was used as the coloring reagent. All UM cell lines used in the present study were tested. All sera and reagents used for immunocytochemical staining were obtained form Dakopatts (Copenhagen, Demark) with the exception of IL-1R1 antibody, which was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). 

DOPA Test

The DOPA test was used for the detection of tyrosinase activity. The method for the DOPA test has been described by us previously. ³ Briefly, UMs cultured in chamber slides were fixed with 5% formalin, incubated with 0.1% L-DOPA (3,4-dihydroxyphenylalanine; Sigma) at 37°C for 3.5 hours with one change of solution, and then fixed with 10% formalin, coverslipped, and examined by light microscopy. Cells with tyrosinase activity were stained brown. All UM cell lines used in the present study were tested. 

Secretion of IL-6 in UMs with or without Serum

Early passages of six cell lines of cultured UMs were used in the present study. Ages of the donors who provided the six tested UM cell lines ranged from 30 to 65 years (mean ± SD, 53.7 ± 15.1). Cells were plated onto 24-well plates at a density of 1 × 10⁵ per well. After 24 hours, the culture medium was withdrawn. Cultures were washed with PBS twice, and serum-free culture medium or medium supplemented with 10% FBS was added. Conditioned media were collected 24 hours later and centrifuged at 800g for 5 minutes, and the supernatants were transferred to vials and stored at −70°C until analysis. Serum-free culture medium without cells was also tested. All experiments were performed in triplicate. 

Comparison of IL-6 Secretion between UMs and RPE

Early passages of one cell line of cultured UMs (from the choroid) and one cell line of RPE from the same donor eye were plated onto 24-well plates at a density of 1 × 10⁵ per well. After 24 hours, the culture medium was withdrawn, washed with PBS, and replaced with culture medium with or without FBS. After 24 hours, conditioned media were collected and stored as described. All experiments were performed in triplicate. 

IL-6 Secretion in UMs with and without IL-1β Stimulation

Early passages of cultured UMs (from the choroid) were plated onto 24-well plates at a density of 1 × 10⁵ per well. In a dose-effect study, after 24-hour culture, the cultured medium was withdrawn, washed, and replaced with serum-free medium. IL-1β at different concentrations (0, 0.1, 1.0, and 10 ng/mL) was added to the media. After 24 hours, conditioned media were collected and stored as described. In a time-effect study, after 24-hour culture, the cultured medium was withdrawn, washed, and replaced with serum-free medium. IL-1β (10 ng/mL) was added, and conditioned media (with and without IL-1) were collected at 6, 12, and 24 hours later and stored. All experiments were performed in triplicate. 

Measurement of IL-6 Levels

The amount of IL-6 protein in the conditioned media was determined using a human IL-6 ELISA kit (Quantikine; R&D Systems, Minneapolis, MN) according to the manufacturer's instructions. Optical density was read by using a microplate reader (Multiskan EX; Thermo, Vantaa, Finland) at 450 nm. The amount of IL-6 (pg/mL) was calculated from a standard curve. The sensitivity of this kit was 0.7 pg/mL. 

Measurement of IL-1β Levels

The amount of IL-1β protein in the cultured medium with 10% FBS but without cells was determined using the human IL-1β ELISA kit (Quantikine; R&D Systems) according to the manufacturer's instructions. Optical density was read by using a microplate reader (Multiskan EX; Thermo) at 450 nm. The amount of IL-1β (pg/mL) was calculated from a standard curve. The sensitivity of this kit was 1.0 pg/mL. 

RNA Isolation and RT-PCR

Early passages of cultured UMs were plated onto six-well plates at a density of 5 × 10⁵. After 24 hours, the culture medium was replaced with serum-free culture medium. In a time-effect study, IL-1β at 10 ng/mL was added to the culture medium, and cells were collected 0.5, 2, and 6 hours later. After the culture medium was withdrawn, the cultures were washed with cold PBS, and cells were harvested by scraping with a rubber policeman. Cells cultured without IL-1β were used as negative controls. After microcentrifuge at 800g for 5 minutes at 4°C, cell pellets were collected for mRNA extraction. Total RNA was isolated with a purification kit (RNeasy Mini Kit; Qiagen, Valencia, CA) according to the manufacturer's instructions. The first-strand synthesis system for RT-PCR kit (SuperScript; Invitrogen, Camarillo, CA) was used to perform cDNA synthesis. PCR primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were TGAACTGAAAGCTCTCCACC and CTGATGTACCAGTTGGGGAA. IL-6 primers were CACTCACCTCTTCAGAACGAAT and TTTGTACTCATCTGCACAGCTC. Both primers were obtained from Invitrogen. First-strand cDNAs were synthesized from 0.5 μg total RNA at 50°C for 50 minutes. PCR amplification was conducted in a PCR system (GeneAmp 9700; Applied Biosystems, Foster City, CA) using the following parameters: first denaturation at 94°C for 5 minutes followed by 35 cycles of reactions of denaturation at 94°C for 30 seconds, annealing at 58°C for 45 seconds, extension at 72°C for 45 seconds, and last extension for 5 minutes at 72°C. After amplification, samples were run on a 1% agarose gel (Invitrogen) in TBE (0.01 M Tris-borate), 0.001 M EDTA (Invitrogen) containing 2.0 μg/mL ethidium bromide (Invitrogen). Bands were visualized and photographed on a UV transilluminator (ChemiDoc XRS System; Bio-Rad, Hercules, CA). In a dose-effect study, IL-1β at different concentrations (0, 0.1, 1.0, 10 ng/mL) were added to the medium. After 6 hours, cells were collected and treated, and RT-PCR was performed as described. 

p38, ERK, and JNK MAPK Assay in UMs Cultured with and without IL-1β

UMs were seeded into six-well plates at a density of 1 × 10⁶. After 24 hours, IL-1β (10 ng/mL) was added. After 60 minutes, the cultures were washed with cold PBS, and the cells were harvested by scraping with a rubber policeman. Cells cultured without IL-1β were used as negative controls. After microcentrifuge for 5 minutes at 4°C, pellets were treated with ice-cold buffer (Cell Extraction Buffer; BioSource, Camarillo, CA) with protease inhibitors (Protease Inhibitor Cocktail; Sigma) and phenylmethylsulfonyl fluoride (BioSource) for 30 minutes, with subsequent vortexing at 10-minute intervals. Cell extractions were microcentrifuged for 30 minutes at 4°C. Supernatants were collected into vials and stored at −70°C until analysis. Phosphorylated p38 mitogen-activated protein kinase (MAPK), extracellular signal-regulated kinases 1 and 2 (ERK1/2), and c-Jun N-terminal kinase (JNK) measurement was performed in triplicate by using p38 MAPK, ERK, and JNK ELISA kits (BioSource) according to the protocol outlined by the manufacturer and were expressed as percentages of the control (cells not exposed to IL-1β). The sensitivity of these kits was 0.8 U/mL. 

NF-κB Assay in Nuclear Extracts in UMs Cultured with and without IL-1β

UMs were plated onto six-well plates at a density of 1 × 10⁶ cells per well. After 24-hour incubation, the medium was replaced, and IL-1β (10 ng/mL) was added to the medium, as described. Cells cultured without IL-1β were used as the negative controls. After 30 minutes, the culture medium was withdrawn. Cells were washed with cold PBS and then scraped from the well. Cells were treated with hypotonic buffer (BioSource) and centrifuged. The pellet (nuclear fraction) was collected and treated (Cell Extraction Buffer; BioSource), vortexed, and centrifuged. The supernatants (nuclear cell extracts) were stored at −70°C until analysis. The amount of NF-κB in nuclear cell extracts was measured by using NF-κB ELISA kits (KHO0371; Invitrogen) according to the manufacturer's instructions. NF-κB levels in nuclear extracts were calculated using a standard curve and expressed as percentages of the negative controls. The sensitivity of this kit was <50 pg/mL. All tests were performed in triplicate. 

Effects of MAPK and NF-κB Inhibitors on IL-1β–Induced Release of IL-6 by UMs

UMs were plated onto 24-well plates at a density of 1 × 10⁵ cells per well. After 24-hour incubation, the medium was changed, and various MAPK or NF-κB inhibitors were added to the medium separately, including 5 μM BAY11–7082 (NF-κB inhibitor), 10 μM UO1026 (ERK inhibitor), 10 μM SP600125 (JNK inhibitor), and 10 μM SB203580 (p38 MAPK inhibitor), all from Calbiochem (San Diego, CA). Thirty minutes later, IL-1β was added to the medium at a final concentration of 10 ng/mL. Cells cultured without IL-1β were used as negative controls. Cells cultured with IL-1β and without any inhibitors were used as positive controls. After 24-hour incubation, the conditioned media were collected and stored. The amount of IL-6 in the supernatants was determined using the human IL-6 ELISA kit (Quantikine; R&D Systems), as described. Tests were performed in triplicate. 

Statistical Analysis

Statistical significance of differences in means throughout this study were calculated by one-way ANOVA in comparing data from more than two groups and Student's t-test in comparing data between two groups. P < 0.05 and P < 0.01 were considered to be statistically significant and very significant, respectively. 

Results

Immunocytochemical Studies

UM cell lines used in the present study stained positively with anti–S-100 antibody and were not stained by anti–keratin and anti–α-smooth muscle actin antibodies (Fig. 1). UMs were also stained by anti–IL-1R1 antibodies, indicating the presence of IL-1 receptors (Fig. 1). 

Figure 1.

 

Immunocytochemical staining of cultured human UMs. (A) Labeled with anti–S100 antibody. (B) Labeled with α-smooth muscle actin antibody. (C) Labeled with anti–cytokeratin antibody. (D) Labeled with anti–IL-1R1 antibody.

DOPA Test

DOPA tests in UMs cell revealed positive results, indicating the presence of tyrosinase activity in these cells (Fig. 2). 

Figure 2.

 

DOPA test of cultured human UMs. Cells revealed positive reaction.

Expression and Secretion of IL-6 in UMs Cultured with Serum-Free Culture Medium

Low levels of IL-6 protein was detected in the conditioned medium from all six UM cell lines cultured with serum-free culture medium (Table 1). The range of IL-6 levels was 24.9 to 31.0 pg/mL in these cell lines, with an average of 28.2 ± 3.7 pg/mL (mean ± SD), which might represent the basic (constitutive) secretion of IL-6 by normal UMs in vitro. The difference of IL-6 levels secreted from different cell lines was statistically nonsignificant (P > 0.05). IL-6 could not be detected in serum-free culture medium without cells. RT-PCR analysis demonstrated that IL-6 mRNA was expressed in normal UMs cultured with serum-free culture medium (Fig. 3). 

Table 1.

 

Secretion of IL-6 in Uveal Melanocytes Cultured with or without Serum

	Location	Iris Color	Age (y)	IL-6 (pg/mL)
				Serum-free	10% Serum
1	Iris	Blue	38	25.6 ± 2.9	69.3 ± 5.3
2	Iris	Brown	74	30.8 ± 3.5	76.3 ± 2.5
3	Ciliary body	Green	62	26.9 ± 4.1	69.2 ± 7.1
4	Choroid	Blue	38	31.0 ± 3.2	76.6 ± 5.6
5	Choroid	Green	62	24.9 ± 2.0	79.9 ± 5.3
6	Choroid	Brown	63	30.2 ± 2.7	73.5 ± 5.4
Mean ± SD				28.2 ± 3.7	74.1 ± 6.3

 Six uveal melanocyte cell lines isolated from different locations of the uveal tract from donor eyes with various colored irides were plated into 24-well plates at a density of 1× 10⁵ per well. After 24 hours, the culture medium was withdrawn. Cultures were washed by PBS and serum-free culture medium, or medium supplemented with 10% serum was added. Conditioned medium was collected 24 hours later. The amount of IL-6 protein in the conditioned medium was determined using the human IL-6 ELISA kit. The difference of IL-6 levels secreted from different cell lines (cultured with or without serum) was statistically nonsignificant (P > 0.05). The difference in IL-6 levels in cells cultured with and without serum was statistically significant (P < 0.01).

Figure 3.

 

Dose- and time-dependent effects of IL-1β on IL-6 mRNA expression by human UMs. Representative RT-PCR profiles from three experiments showed the mRNA expressions of IL-6 by cells exposed to IL-1β at different concentrations and time periods. GAPDH was used as an internal loading control. (A) Cells were plated onto six-well plates. IL-1β at 0, 0.1, 1.0, and 10 ng/mL was added to the culture. Six hours later, cells were collected, mRNA was extracted, and RT-PCR analysis was performed as described in the text. (B) Cells were plated onto six-well plates. IL-1β (10 ng/mL) was added, cells were collected 0.5, 2, and 6 hours later, mRNA was extracted, and RT-PCR analysis was performed as described in the text.

In UMs cultured with 10% FBS, IL-6 levels in the conditioned culture medium were significantly increased to 74.1 ± 6.3 pg/mL (Table 1) (2.64 times values of cells cultured without serum), which might represent higher levels of IL-6 production by serum-activated UMs. The difference of IL-6 levels in cells cultured with and without serum was statistically significant (P < 0.01). The differences of IL-6 levels in the culture medium from the six different UM cell lines were statistically nonsignificant (P > 0.05). 

IL-1β Amount in Culture Medium with Serum

IL-1β could not be detected in culture medium supplemented with 10% FBS but without cells, indicating that IL-1β was not the molecule in the serum that stimulated the expression of IL-6 in the present study. 

Comparison of IL-6 Secretion between UMs and RPE

IL-6 levels in the conditioned culture media from the UMs and RPE isolated from the same healthy donor eye were measured and compared. In cells cultured with serum-free medium, IL-6 levels were 29.4 ± 3.7 pg/mL and 32.2 ± 3.0 pg/mL in the conditioned media of the UMs and RPE, respectively. The secretion of IL-6 from the UMs was slightly less than that from the RPE; however, the difference between these two cell lines was statistically nonsignificant (P > 0.05). In cells cultured with 10% FBS, IL-6 levels were 85.5 ± 12.4 and 111.1 ± 12.5 in the conditioned media from the UMs and RPE, respectively. IL-6 levels in UMs and RPE cultured with 10% FBS were 2.91 and 3.45 times control values, respectively. The secretion of IL-6 from the UMs was also slightly less than that from the RPE, but this was statistically nonsignificant (P > 0.05). 

Effects of Exogenous IL-1β on IL-6 Production by UMs

When IL-1β was added to the culture medium of UMs at different concentrations (0.1–10 ng/mL) and cultured for 24 hours, there were significantly increased levels of IL-6 protein in the conditioned medium in a dose-dependent manner (Fig. 4). IL-6 levels in conditioned medium from cells cultured without IL-1β was 30.6 ± 3.9 pg/mL (mean ± SD). IL-6 levels in conditioned medium from cells cultured with IL-1β at 0.1, 1.0, and 10 ng/mL were 1.36, 4.01, and 6.24 times control values (cell cultured without IL-1β), respectively (Fig. 4). The difference of IL-6 levels between IL-1β–treated cells and controls was statistically significant at 0.1 ng/mL IL-1β (0.05 > P > 0.01) and very significant at 1 to 10 ng/mL IL-1β (P < 0.01). 

Figure 4.

 

Dose-dependent effects of IL-1β on the secretion of IL-6 by cultured human UMs. Cells were plated onto 24-well plates. After 24-hour incubation, IL-1β at 0, 0.1, 1.0, and 10 ng/mL was added to the culture and incubated for 24 hours. Conditioned culture media were collected, and the amount of IL-6 protein in the conditioned media was determined. IL-6 levels in conditioned culture media were expressed as pg/mL (mean ± SD in triplicate tests). *P < 0.05 and **P < 0.01, compared with controls (cells cultured without IL-1β).

The IL-1β–induced increase of secretion of IL-6 by UMs was also time dependent (Fig. 5). The IL-6 level in conditioned media from cells cultured without IL-1β was 29.5 ± 3.5 pg/mL. IL-6 levels in conditioned media from cells cultured with IL-1β (10 ng/mL) for 6, 12, and 24 hours were 2.12, 3.75, and 6.90 times control values, respectively (Fig. 5). The difference in IL-6 levels between IL-1β–treated cells and controls was statistically very significant in cells treated for 6, 12, and 24 hours (P < 0.01). 

Figure 5.

 

Time-dependent effects of IL-1β on the secretion of IL-6 by cultured human UMs. Cells were plated onto 24-well plates. After 24-hour incubation, IL-1β (10 ng/mL) was added to the culture and cultured for 6, 12, and 24 hours. Conditioned culture media were collected, and the amount of IL-6 protein in the conditioned media was determined kit. IL-6 levels in conditioned culture medium were expressed as pg/mL (mean ± SD in triplicate tests). *P < 0.05, **P < 0.01, compared with controls (cells cultured without IL-1β).

Effects of IL-1β on Expression of IL-6 Transcript by UMs

The RT- PCR experiment demonstrated that IL-6 mRNA was expressed in UMs cultured without IL-1β (Fig. 3). IL-1β upregulated IL-6 mRNA expression in the UMs. In the time-dependent study, the expression of IL-6 in IL-1β (10 ng/mL)–treated cells increased to 1.09, 2.80, and 6.64 times control values (cells cultured without IL-1β) after 0.5-, 2-, and 6-hour treatments, respectively. The difference in the IL-6 mRNA levels between IL-1β–treated cells and controls was not significant in cells treated with IL-1β for 0.5 hour (P > 0.05) and was statistically very significant in cells treated for 2 and 6 hours (P < 0.01). 

In the dose-dependent study, the IL-6 mRNA level in cells treated with 0.1, 1.0, and 10 ng/mL IL-1β increased to 1.87, 4.00, and 6.07 times control values after 6-hour treatment, respectively (Fig. 3). The difference in the IL-6 mRNA levels between the IL-1β–treated cells and the controls was statistically very significant in cells cultured with 0.1, 1.0, and 10 ng/mL IL-1β (P < 0.01). 

Effects of IL-1β on NF-κB and Phosphorylated MAPK Levels in UMs

IL-1β treatment (10 ng/mL with 30-minute incubation) increased phosphorylated p38 MAPK levels but not phosphorylated ERK1/2 and JNK1/2 levels in UMs (Fig. 6). The levels of phosphorylated p38 MAPK in UMs cultured with IL-1β increased to 7.45 ± 0.77 (mean ± SD) times control values (UMs cultured without IL-1β). The difference in phosphorylated p38 MAPK levels between cells treated with and without IL-1β was statistically very significant (P < 0.01). The level of phosphorylated JNK1/2 in cells cultured with IL-1β increased to 1.19 ± 0.16 (mean ± SD) times control values. However, the difference in phosphorylated JNK1/2 levels between cells treated with and without IL-1β was statistically nonsignificant (P > 0.05). The level of phosphorylated ERK1/2 in cells cultured with IL-1β was 0.89 ± 0.08 (mean ± SD) times control values. The difference in phosphorylated ERK1/2 levels between cells treated with and without IL-1β was also statistically nonsignificant (P > 0.05). 

Figure 6.

 

Effects of IL-1β on NF-κB in nuclear extracts and phosphorylated ERK, JNK, and p38 MAPK in cultured UMs. Cells were plated onto 24-well plates. After 24-hour incubation, IL-1β (10 ng/mL) was added to the medium. Cells were collected 30 minutes later. The amount of NF-κB levels in nuclear extracts and phosphorylated p38 MAPK, JNK, and ERK1/2 in cell lysates were measured using the relevant NF-κB ELISA kit and phosphorylated MAPK ELISA kits. The levels of NF-κB in nuclear extracts and phosphorylated p38, ERK and JNK in cell lysates were expressed as the percentages of the controls (cells cultured without IL-1β). * P < 0.05, **P < 0.01, compared with controls.

IL-1β treatment increased NF-κB levels in nuclear extracts of the UMs (Fig. 6). The levels of NF-κB in nuclear extracts in cells cultured with IL-1β were 3.35 ± 0.30 times control values. The difference of NF-κB levels between cells treated with and without IL-1β was statistically very significant (P < 0.01). 

Effects of MAPK and NF-κB Inhibitors on IL-1β–Induced Release of IL-6 by UMs

IL-6 protein levels in conditioned media from cells cultured with and without IL-1β were 155.7 ± 21.5 (positive control) and 24.2 ± 2.3 pg/mL (negative control), respectively (Fig. 7). 

Figure 7.

 

Effects of NF-κB and MAPK inhibitors on IL-1β–induced production of IL-6 by UMs. Cells were plated onto 24-well plates. After 24-hour incubation, various MAPK and NF-κB inhibitors were added to the medium separately, including BAY11–7082 (NF-κB inhibitor), UO1026 (ERK inhibitor), SP600125 (JNK inhibitor), and SB203580 (p38 MAPK inhibitor), at a final concentration of 10 μM with the exception of BAY11–7082 (5 μM). Thirty minutes later, IL-1β (10 ng/mL) was added to the medium. Cells cultured without IL-1β were used as negative controls. Cells cultured with IL-1β but without inhibitors were used as positive controls. After 24-hour incubation, culture medium was collected, and the IL-6 levels were measured with human IL-6 Quantikine ELISA kit and expressed as pg/mL (mean ± SD in triplicate tests).*P < 0.05, **P < 0.01, compared with the positive controls (cells cultured with IL-1 β but without inhibitors).

Treatment of cells with UO1026 (ERK inhibitor) or SP600125 (JNK inhibitor) before the addition of IL-1β did not cause significant changes in IL-6 levels of conditioned medium compared with cells cultured with IL-1β alone (positive control) (P > 0.05) (Fig. 7). 

Treatment of cells with SB203580 (p38 MAPK inhibitor) 30 minutes before the addition of IL-1β decreased the secretion of IL-6 to 49% of that in cells treated with IL-1β alone (Fig. 7). The difference in the amount of IL-6 in medium between cells treated with and without SB203580 (positive control) was very significant (P < 0.01). However, IL-6 levels in cells treated with SB203580 and IL-1β were still significantly greater than those of the negative controls (P < 0.01), indicating that p38 inhibition partially inhibited but did not completely abolish IL-1β–induced secretion of IL-6 in UMs. 

Treatment of BAY11–7082 (NF-κB inhibitor) decreased the release of IL-6 by the UMs to 61% of that in cells treated by IL-1β alone (Fig. 7). The difference in the amount of IL-6 in the medium between cells treated with and without BAY11–7082 (positive control) was very significant (P < 0.01). The levels of IL-6 in medium from BAY11–7082 and IL-1β–treated cells were still significantly greater than those of negative controls (P < 0.01), indicating that NF-κB inhibition partially inhibited but did not completely abolish IL-1β–induced expression of IL-6 in UMs. 

Treatment of SB203580 combined with BAY11–7082 decreased the release of IL-6 by UMs to 80% of that in cells treated by IL-1β alone (Fig. 7). The difference in the amount of IL-6 in medium between cells treated with and without SB203580 and BAY11–7082 (positive control) was statistically very significant (P < 0.01). The difference in IL-6 levels between cells treated with IL-1β, SB203580, and BAY11–7082 and the negative controls was statistically nonsignificant (P > 0.05), indicating that blocking both p38 and NF-κB pathways could completely abolish IL-1β–induced IL-6 secretion in UMs. 

Discussion

UMs are the major resident cells in the uveal tract, and they play an important role in the pathogenesis of various common uveal diseases, including uveitis and uveal melanoma. UMs are considered antigen-presenting cells in human Vogt-Koyanagi-Harada disease ^33,34 and experimental animal melanin-induced uveitis. ^35,36 However, investigation of UMs in vitro has been hampered by an inability to obtain a sufficient number of pure UMs for study. Since 1990, we have developed methods for the isolation, cultivation, purification, and investigation of human UMs. This in vitro model of UMs has been used for studying the growth, function, and melanogenesis of these cells for the past 20 years. ^{1 –9} In the present study, tested cells expressed S-100 but did not express cytokeratin and α-smooth muscle actin. These results documented that the cells used were pure melanocytes (negative results for cytokeratin and α-smooth muscle actin distinguished melanocytes from retinal pigment epithelial cells, and positive results of S-100 excluded the possibility of fibroblasts). ¹ Furthermore, the DOPA test, which can determine the presence of tyrosinase activity, has been used to study melanogenesis of the UMs used here. All tested cells showed a positive DOPA reaction, indicating the presence of tyrosinase activity in these cells. ³ Cultured adult human retinal pigment epithelial cells never express tyrosinase activity. ³⁷ These results provide further evidence that the cells used in this study were pure UMs. 

IL-6 is a pleiotropic cytokine that regulates multiple biological processes, including the development of the nervous and hematopoietic systems, acute-phase responses, inflammation, and immune responses. ^{10 –14,38} IL-6 is produced by a wide variety of cell types, including inflammatory cells (lymphocytes, macrophages, monocytes, neutrophils) and noninflammatory cells, such as fibroblasts, myocytes, osteoblasts, endothelial cells, and epithelial cells (among them retinal pigment epithelial cells). ^{15 –20,38 –45} IL-6 expression in other uveal cells, such as fibroblasts, macrophages, and T lymphocytes, have been reported. ²⁷ However, to the best of our knowledge, IL-6 production by UMs and its modulation by IL-1 have not been previously reported. In the present study, cultured human UMs consistently express IL-6 mRNA and secrete low levels of IL-6 at basal conditions (cultured with serum-free culture medium), indicating a low constitutive expression of IL-6 in normal UMs. 

The secretion of IL-6 by UMs is significantly increased with additional serum in the culture medium. Serum contains low levels of cytokines, growth factors, and other factors that may activate cells for the production of various biological substances. ⁴⁴ IL-1β could not be detected in culture medium with 10% serum. Therefore, it seems that IL-1β is not the molecule in the serum that stimulated the expression of IL-6 in the present study. Of the many molecules we studied previously, several could modulate the growth and melanogenesis of UMs in vitro (e.g., hepatocyte growth factor, bFGF, transforming growth factor-β1, epinephrine, prostaglandin E, and endothelin1), and these are present in the serum. ⁵ These may be the molecules in serum that induced the expression of IL-6 in UMs. Furthermore, there are numerous other molecules and likely more unknown factors in the serum. The exact factor responsible for serum-induced expression of IL-6 by UMs requires further investigation. Retinal pigment epithelial cells are known to produce and secrete IL-6. ^{15 –20} Compared with the RPE, cultured UMs produce lower levels of IL-6 and are able to increase secretion in a culture medium containing serum, a finding similar to that in cultured RPE. ^{15 –20}  

Although UMs are derived from neural crest and retinal pigment epithelial cells are derived from neuroectoderm, these two cell types play relatively similar roles in low-grade inflammation induced by chronic tissue stress (para-inflammation). ⁴⁶ In aging eyes, a remarkable histologic feature is the appearance of many round, enlarged melanocytes or “activated melanocytes.” ⁴⁷ These activated melanocytes are considered the consequence of para-inflammatory responses. Normal age-related para-inflammatory responses are beneficial for tissue repair and remodeling and for the restoration of tissue physiological functions. ⁴⁶ Therefore, constitutive expression of low-level IL-6 from normal UMs may help maintain uveal tissue homeostasis during the aging process. 

However, uncontrolled inflammation resulting from prolonged tissue damage or altered immune responses can become harmful and contribute to aging diseases in the eye such as age-related macular degeneration (AMD). ⁴⁸ In uncontrolled inflammation, many proinflammatory cytokines and mediators are released. One of the highly active proinflammatory cytokines is IL-1. ⁴⁹ IL-1 induces the expression of several other proinflammatory genes, such as cytokines (including IL-6), chemokines, growth factors, matrix proteases, prostaglandins, and adhesion molecules, in different cell types. ^{10,15 –17,29,39 –43} In the present study, IL-1 receptors were detected in cultured UMs. IL-1β dramatically increased the expression and secretion of IL-6 by UMs in a time- and dose-dependent manner. This is consistent with previous reports that the production of IL-6 was stimulated by IL-1 in various cell types, including RPE. ^{15 –18,38 –43}  

Two main signal pathways involve the IL-1β–induced expression of IL-6 by other cell types: MAPK and NF-κB signal pathways. The MAPK pathways seem to preferentially enhance cytokine induction downstream from NF-κB activation. These pathways are some of the most important pathways influenced by IL-1β. This family consists primarily of ERK1/2, JNK1/2, and p38 MAPK pathways. The ERK1/2 pathway responds to growth factors and has an important role in modulating the survival and growth of cells, whereas p38 MAPK and JNK pathways response to various stresses, including proinflammatory cytokines, and modulate the differentiation and apoptosis of cells. IL-1 is reported to stimulate the production of IL-6 in fibroblasts, myocytes, endometrial stroma cells, and leukemic cells through the activation of p38 MAPK, JNK1/2, or ERK1/2 signaling pathways (mainly through the p38 MAPK pathway). ^{30,39 –41}  

NF-κB is a major transcription factor that promotes the expression of more than 150 genes involved in a variety of cellular processes. ⁵⁰ NF-κB is present in the cytoplasm in an inactive NF-κB complex through their noncovalent interaction with inhibitory proteins known as IκBs. In response to various stimuli, including cytokines, the latent cytoplasmic NF-κB/IκBα complex is activated by phosphorylation on conserved serine residues in the N-terminal portion of IκB. Activated NF-κB translocates to the nucleus and binds to its cognate DNA-binding site in the promoter or enhancer regions of specific genes and then induces the expression of relevant genes, including various cytokines. ⁵⁰ It has been reported that the IL-1β–induced expression of IL-6 occurs through the activation and translocation of NF-κB to the nucleus. ^42,43  

The involvement of the MAPK or NF-κB pathway and the selection of three different MAPK pathways in IL-1β–induced IL-6 expression are cell type specific. ^{30,39 –44} Therefore, the study of effects of IL-1β on these signal pathways in UMs is essential for determining the exact signal pathways responsible for IL-1β–induced expression of IL-6 in these cell types. 

The present study found that IL-1β causes a significant increase of p38 MAPK and NF-κB but not of ERK1/2 or JNK1/2. p38 MAPK or NF-κB inhibitors alone inhibit, but do not abolish, the IL-1β–induced production of IL-6 by UMs, whereas ERK1/2 and JUN1/2 inhibitors have no significant influence on the IL-1β–induced production of IL-6. Combination of p38 MAPK and NF-κB inhibitors before the addition of IL-1β reduced the levels of IL-6 secretion closely to that of the negative controls (cells cultured without IL-1β), indicating that simultaneously blocking both p38 MAPK and NF-κB pathways may completely abolish the IL-1β–induced secretion of IL-6 by UMs. Therefore, it seems that the IL-1β–induced secretion of IL-6 by UMs occurs through the activation of the p38 MAPK and NF-κB signal pathways. This is consistent with previous reports regarding the signal pathways involved in IL-1β–induced secretion of IL-6 by other cell types. ^{30,39 –44}  

The present study provides evidence of the production of the inflammatory cytokine IL-6 by cultured UMs, supporting the hypothesis that these cells may play a role in the pathogenesis of uveitis. Furthermore, recently, subtoxic levels of hydrogen peroxide were found to stimulate the production of IL-6 by RPE, which suggests a molecular linkage between oxidative stress and inflammatory and autoimmune reactions in the pathogenesis of AMD. ⁵¹ Increased IL-6 levels in the aqueous humor and vitreous have been measured in various ocular diseases, including uveitis, proliferative vitreoretinopathy, proliferative diabetic retinopathy, retinal vein occlusion, retinopathy of prematurity, and AMD. ^{21 –28,52} This indicates that IL-6 may be involved in the pathogenesis of ocular inflammatory and autoimmune diseases and in retinal degenerative diseases such as AMD in which inflammation is a major contributor to pathologic changes. 

In summary, the present studies demonstrated that UMs constitutively express and secrete IL-6 at a relatively low level. The secretion of IL-6 is increased by activation with serum and is dramatically increased by the stimulation of the proinflammatory cytokine IL-1β. UMs may play a role in ocular inflammatory processes and in the development of autoimmune and retinal degenerative diseases in the eye. 

Footnotes

 Supported by the Pathology Research Fund of New York Eye and Ear Infirmary and the Bendheim-Lowenstein Family Foundation, New York.

Footnotes

 Disclosure: D.-N. Hu, None; M. Chen, None; D.Y. Zhang, None; F. Ye, None; S.A. McCormick, None; C.-C. Chan, None