Effect of Interferon alfa-2a Treatment on Adaptive and Innate Immune Systems in Patients With Behçet Disease Uveitis

Abstract

Purpose: To investigate the effect of interferon alfa-2a on T regulatory (Treg) cells, T helper 17 (Th17) cells, and expression of Toll-like receptors (TLRs) in Behçet disease (BD) patients with uveitis.

Methods: Twenty-seven patients who received interferon alfa-2a for active BD uveitis despite conventional immunomodulatory therapies and healthy controls were enrolled. Peripheral blood Treg and Th17 cell frequencies were determined by flow cytometry as gated cells for CD3⁺CD4⁺Foxp3⁺ and CD3⁺CD4⁺IL17A⁺, respectively. Th17 RAR-related orphan receptor (ROR)γt mRNA expression was verified by real-time PCR (RT-PCR). Treg and Th17 cell cytokines were detected by ELISA in the supernatant of short-term cell cultures. RT-PCR was used to assess expression of TLR-2, TLR-3, TLR-4, TLR-8, and TLR-9 using cDNA prepared from CD4⁺ T cells and monocytes.

Results: Treg and Th17 cell frequencies and Th17 RORγt expression were significantly elevated, and IL-10 concentration in Treg cell supernatants was significantly lower in BD patients than in controls. Th17 IL-17, IL-6, IL-21, IL-22, IL-23, IFN-γ, and TNF-α concentrations were significantly higher and all TLR expressions were significantly elevated in patients. Interferon alfa-2a led to a significant reversal in Treg and Th17 cell frequencies, Th17 RORγt expression, Treg and Th17 cell cytokine production, and TLR expression by CD4⁺ T cells and monocytes.

Conclusions: Despite a relative increase in Treg cells, impaired IL-10 production suggests that Treg dysfunction may play a role in induction of BD uveitis. Favorable effects of interferon alfa-2a may be associated with recovery of Treg cell function, suppression of Th17 cells, and reduced expression of TLRs on CD4⁺ T cells and monocytes.

Uveitis of Behçet disease (BD) is characterized by bilateral nongranulomatous panuveitis and retinal vasculitis.¹ Interferon alfa-2a has been shown to be efficacious in the treatment of BD uveitis refractory to conventional immunomodulatory therapies.^2–4 Interferons are innate cytokines with pleiotropic effects including control of viral disease, cell growth, apoptosis, and innate immunity. Their role in adaptive immunity is complex and less well characterized.⁵ 

The etiopathogenesis of BD is still not well known. Both innate and adaptive immune systems are implicated in the pathogenesis.⁶ CD4⁺CD25⁺ T regulatory (Treg) cells, a subset of CD4⁺ T cells, are recognized for their capability to downregulate immune responses against self- or non-self antigens and thus maintain immune tolerance.⁷ Transcription factor forkhead box protein 3 (Foxp3) has been shown to play a critical role in the induction of Treg cells and has been used as a marker to define this cell subset.⁷ Although scarce, studies focusing on the frequency of Treg cells in the peripheral blood show controversial results in patients with BD uveitis.^8–11 

Recent studies documented the relative abundance and central role of T helper (Th) 17 cells and their plasticity by increased expression of IL-17, a key proinflammatory cytokine of T helper 17 (Th17) pathway, and IFN-γ.^12–18 Neutrophil-dominant inflammatory lesions in BD are also possibly driven by Th17 immune responses.¹² Patients with active BD, especially those with uveitis or folliculitis, were reported to have a significantly higher Th17/Th1 cell ratio in peripheral blood than healthy controls.¹⁹ A study showed that Th17 cells from patients with active BD uveitis exposed to the TNF-α inhibitor infliximab in vitro failed to produce IL-17 and showed diminished expression of the Th17 transcription factor, RAR-related orphan receptor (ROR)γt, suggesting that TNF-α plays a role in Th17 differentiation.²⁰ 

Toll-like receptors (TLRs) are key components of the innate immune system for pathogen-associated molecular pattern recognition, which lead to the activation of multiple inflammatory pathways and initiation of systemic response against pathogens.²¹ Inappropriate activation of TLR pathways by endogenous or exogenous ligands may lead to the initiation of autoimmune responses.²¹ A recent resequencing study found that a nonsynonymous variant of TLR-4 gene was associated with BD and suggested that increased innate immune responses directed against certain bacterial components could be involved in the pathogenesis of BD.²² An increase in expression of TLRs has previously been reported from peripheral blood mononuclear cells (PBMCs) and monocytes and from oral mucosa in patients with BD as well as from PBMCs in patients with BD uveitis.^23–26 

In the present study, our aim was to evaluate the effect of interferon alfa-2a on adaptive and innate immune responses in patients who had an acute attack of BD uveitis despite conventional immunomodulatory therapies. We, therefore, analyzed the frequency of peripheral blood Treg and Th17 cells and their in vitro cytokine production as well as the expression of TLR-2, TLR-3, TLR-4, TLR-8, and TLR-9 on CD4⁺ T cells and monocytes both at the initiation of interferon alfa-2a treatment for an acute attack of BD uveitis and after achievement of clinical remission with treatment. 

Methods

Recruitment of Patients and Controls

Twenty-seven consecutive patients with BD who attended two uveitis clinics at tertiary referral centers because of an acute attack of BD uveitis despite conventional immunomodulatory therapies and who were switched to interferon alfa-2a treatment were included from August 1, 2016 to September 30, 2017. All participants met the classification criteria established by the International Study Group for Behçet's Disease.²⁷ Patients with comorbidities, such as diabetes mellitus, malignancy, coexistent autoimmune disease, and allergies were not included in the study. Twenty-seven age- and sex-matched healthy subjects were recruited from consecutive patients who were scheduled for routine ocular examination for refractive error correction at two tertiary centers. 

The protocol of this prospective cross-sectional study was approved by the institutional review board of Koc University (approval number: 2015.017.IRB1.003). All patients and control subjects were white and of Turkish descent. The study was conducted according to the tenets of the Declaration of Helsinki and a written consent was obtained from patients and control subjects. 

Ocular Evaluation and Interferon alfa-2a Treatment

All patients underwent a complete ophthalmologic examination, including best-corrected visual acuity, slit-lamp biomicroscopy, tonometry, and indirect ophthalmoscopy at all visits. Data on age, sex, age at diagnosis of BD, duration of BD uveitis, extraocular manifestations of BD observed over the disease course, and dose and duration of systemic corticosteroids, immunomodulatory agent(s), and/or other medications used at the time of study recruitment were collected. Laterality and anatomic involvement of uveitis were recorded. 

An acute attack of BD uveitis refractory to conventional immunomodulatory therapies was defined as an increase in or the development of vitreous haze, the emergence of inflammatory cells in anterior chamber, inflammatory sheathing of retinal vessels, vascular occlusion, retinal hemorrhages, retinal infiltrates, macular edema, and/or papillitis despite azathioprine and/or cyclosporine A treatment with or without oral corticosteroids. Only patients who presented within 1 week after symptom onset were included in the study. Clinical remission of uveitis was defined as the presence of less than 1 (+) vitreous haze^28,29 and absence of anterior chamber cells and all posterior segment inflammatory signs. 

Interferon alfa-2a treatment was initiated in patients with an acute attack of BD uveitis despite conventional immunomodulatory therapies and was administered using a previously described protocol.³ Briefly, patients were started on 3 million international units (MIU) of recombinant interferon alfa-2a (Roferon-A; Roche Pharmaceuticals, Nutley, NJ, USA) per day subcutaneously. After 2 to 4 weeks the dose of interferon was tapered to 3 MIU every other day or 3 MIU twice weekly depending on ocular response and/or laboratory parameters. For the purpose of the study, total dose of interferon treatment administered was calculated. Conventional immunomodulatory agents were discontinued, and oral corticosteroids were usually discontinued or tapered to 5 mg/day of prednisolone equivalent at the initiation of interferon alfa-2a. 

Peripheral blood (35 mL) was collected into heparinized tubes on the same day of ophthalmologic examination after a decision was made to switch to interferon alfa-2a treatment for an acute attack of BD uveitis and after achievement of clinical remission of uveitis with interferon alfa-2a. Blood was drawn once from control subjects. 

Isolation of Peripheral Blood Mononuclear Cells

Peripheral blood mononuclear cells were isolated by gradient centrifugation using Histopaque 1.077 g/mL (Sigma-Aldrich, Taufkirchen, Germany). Buffy coat was collected, and cells were washed once with PBS and prepared for magnetic separation. Mononuclear cells were seeded into a six-well plate and incubated for 2 hours at 37°C in 5% CO₂ incubator to induce monocyte adherence to the plate. Adhered monocytes were then trypsinized and collected for RNA isolation and TLR expression studies. 

Isolation of CD4⁺, CD4⁺CD25⁺, and CD4⁺CD25⁻ T Cells

MACS Human CD4⁺CD25⁺ regulatory T cell isolation kit (Miltenyi Biotech, Gladbach, Germany) was used to isolate Treg cells from peripheral blood. Biotinylated mouse anti-human CD4 antibody cocktail containing all leukocyte markers except for CD4 was used for negative selection of CD4⁺ T cells, and anti-CD25 microbeads were used for positive selection of CD4⁺CD25⁺ and negative selection of CD4⁺CD25⁻ T cells. CD4⁺ T cells were collected for RNA isolation and TLR expression studies. 

Cell Culture and In Vitro Stimulation of CD4⁺CD25⁺ Treg and CD4⁺CD25⁻ T Cells

RPMI medium containing 10% FBS and 1% penicillin and streptomycin (Thermo Fischer Scientific, Waltham, MA, USA) was used for cell culture. For the stimulation of Treg cells, 10 μg/mL phytohemagglutinin-L (Roche Diagnostics, Mannheim, Germany) was prepared in 3 mL RPMI. Treg pellets were resuspended in the medium and seeded into six-well plates for 72 hours at 37°C. After 72 hours, supernatants were collected for analysis by ELISA cytokine assays. For the stimulation of CD4⁺CD25⁻ T cells, 3 mL RPMI medium containing 2 μg/mL anti-CD3, 2 μg/mL anti-CD28, 5 μg/mL anti-IFN-γ, 1 μg/mL anti-IL-4, 5 μg/mL anti-IL-1β, 1 μg/mL anti-IL6 (BD Biosciences, San Jose, CA, USA), and 10 μg/mL anti-IL23 (Sigma-Aldrich) was added onto CD4⁺CD25⁻ T cells, and cells were seeded into six-well plates for 6 days at 37°C. Fresh medium containing 10 μg/mL IL-23 was added onto the cells on the second and fifth day. On the sixth day, supernatants were collected for ELISA analyses. Cells were suspended with 3 mL RPMI medium containing 20 ng/mL PMA (Sigma-Aldrich), 5 μg/mL PHA-L, and 3 μL protein transport inhibitor Brefeldin-A (BD Biosciences) to stimulate IL-17A synthesis. After 12 hours of incubation at 37°C, cells were collected for IL-17A staining and flow cytometry and RNA isolation for RORγt gene expression analysis. 

Flow Cytometry

Treg and Th17 cells were determined by flow cytometry using human Foxp3 buffer set and Cytofix/Cytoperm Kit (BD Biosciences), respectively, following the manufacturer's instructions. Monoclonal mouse anti-human CD3 PECy7, CD4 FITC, CD25 APC, Foxp3 PE, and IL-17A PE antibodies (BD Biosciences) were used to identify Treg and Th17 cells. Th17 cells were determined by flow cytometry after in vitro stimulation of peripheral blood cells in culture. Samples were run under BD Accuri C6 flow cytometer (BD Biosciences) and analyzed by FlowJo software v.10 (Tree Star, Inc., Ashland, OR, USA). Fifty thousand cells were counted in lymphocyte gate. Treg and Th17 cells were determined as CD3⁺CD4⁺Foxp3⁺ and CD3⁺CD4⁺IL17A⁺ cells, respectively. 

Molecular Studies

RNA isolation was done by using NucleoSpin RNA isolation kit (Macherey-Nagel, Düren, Germany) and by following the manufacturer's instructions. For cDNA synthesis, M-MLV reverse transcriptase kit (Thermo Fischer Scientific) was used. Expression of TLRs and RORγt was determined by using TaqMan real-time ready primer and probe sets (Roche Diagnostics) specific to TLR-2, TLR-3, TLR-4, TLR-8, TLR-9, and RORγt genes. Beta-2 microglobulin expression was used as the housekeeping gene. Gene expression was calculated by using the 2^ddCP comparative CP method. 

Immunoassays

Platinum ELISA assay kits for IL-10, TGF-β₁, IL-2, IL-17, IL-6, IL-21, IL-22, IL-23, IFN-γ, TNF-α (Thermo Fischer Scientific), and IL-35 (Sunred Biolabs, Shanghai, China) were used to detect cytokine concentrations in the supernatants of Treg and Th17 cell cultures. All tests were done in duplicate. 

Fluorescein Angiography

Patients underwent fluorescein angiography on the same day of blood sampling at enrollment and after achieving clinical remission of uveitis with interferon alfa-2a treatment. Fluorescein angiography images were obtained from the posterior pole and retinal periphery at early and late phases (5–10 minutes) using the Heidelberg retinal angiograph 2 platform (Heidelberg Engineering, Heidelberg, Germany) with the 102° noncontact ultra-wide-field lens after an intravenous injection of 2.5 mL 10% sodium fluorescein (Fluorescite; Alcon, Inc., Fort Worth, TX, USA). 

An angiographic scoring system developed by the Angiography Scoring for Uveitis Working Group and with proven interobserver reliability was used for scoring fluorescein angiography.^30–33 The angiographic scoring system used herein is a semiquantitative system in which fluorescein angiography signs include nine categories.³⁰ The total maximum score for a given eye might be 40 in this scoring system. All angiograms were graded by two experienced ophthalmologists (S.O. and M.O.). 

Statistical Evaluation

SPSS 24.0 software was used for statistical analysis (SPSS, Inc., Chicago, IL, USA). For group comparisons, Mann-Whitney U and Wilcoxon tests were used. Association of parameters was assessed using Spearman's coefficient of correlation. A correlation coefficient of 0.1 to 0.3 was considered as weak correlation, 0.4 to 0.6 as moderate correlation, 0.7 to 0.9 as strong correlation, and 1.0 as perfect correlation.³⁴ Results are given as mean ± SD unless specified otherwise. A P value less than 0.05 was considered statistically significant. 

Results

Nineteen (70.4%) male and 8 (29.6%) female patients with an acute attack of BD uveitis despite conventional immunomodulatory therapies and 27 healthy control subjects were recruited. There was no statistically significant difference between the age of patients and control subjects (31.37 ± 7.28 vs. 31.4 ± 7.41 years, P = 0.869). Age at the onset of BD was 28.74 ± 5.89 years (range, 17–40 years). Median duration of BD uveitis was 12 months (range, 2–107 months). See Supplementary Table S1 for frequency of extraocular manifestations of BD. All patients had bilateral involvement and panuveitis except one patient with bilateral posterior uveitis. Data on use of systemic corticosteroids, immunomodulatory agent(s), and/or other medications at enrollment are presented in the Table. 

Table

 

Dose and Duration of Systemic Corticosteroids, Immunomodulatory Agent(s), and/or Other Medications Used at the Time of Study Recruitment in 27 Patients With Active Behçet Disease Uveitis

 

Interferon alfa-2a treatment led to clinical remission of uveitis in 19 (70.4%) patients (16 male and 3 female; age, 31.15 ± 6.08 years). Median total interferon dose used was 339 MIU (range, 135–810 MIU), and median duration of treatment was 5 months (range, 3.5–10.5 months). Oral methylprednisolone (4 mg/day) was continued in 10 and colchicine (1 mg/day) was continued in 2 patients during interferon treatment. Interferon treatment was discontinued due to inefficacy in four patients (14.8%), adverse effects in three (11.1%; depression in two, elevated liver enzymes in one), and intolerance to treatment in one (3.7%) patient. 

Median total fluorescein angiography score of 19 patients (38 eyes) significantly decreased from 17.5 (range, 0–26) at the initiation of interferon alfa-2a to 2.5 (range, 0–15) after clinical remission of uveitis was achieved with interferon treatment (P < 0.001). Figure 1 demonstrates fluorescein angiography in one representative patient. 

Figure 1

 

Fluorescein angiography (FA) and spectral-domain optical coherence tomography (SD-OCT) of the right eye in a representative patient (Heidelberg Engineering). A 25-year-old female patient who presented with bilateral panuveitis despite 2 mg/kg/day azathioprine and 6 mg/day methylprednisolone was switched to 3 MIU/day interferon alfa-2a treatment. Her second blood sample was collected at the 10th month of interferon treatment. The total dose of interferon alfa-2a administered was 780 MIU. Late-frame FA at the initiation of interferon alfa-2a shows retinal vascular staining and leakage and diffuse retinal capillary leakage at the posterior pole and retinal periphery, optic disc staining and leakage, and macular edema with pooling of the dye in cystic spaces. There is blockage of fluorescence due to inflammatory vitreous condensation in the inferior retinal periphery. Total FA score was calculated as 24 (A). FA obtained during clinical remission of uveitis shows complete recovery of angiographic signs with partial optic disc staining at the late phase of the angiogram; total FA score was graded as 1. There is blockage of fluorescence due to vitreous opacity beneath the optic disc and at the inferior retinal periphery (B). SD-OCT B-scan shows cystoid macular edema and presence of subretinal fluid at the initiation of interferon treatment (C). SD-OCT confirms resolution of macular edema and reinstitution of foveal microstructure during remission of uveitis (D).

Comparison of Study Outcomes Between 27 BD Patients Who Had an Acute Attack of Uveitis Despite Conventional Immunomodulatory Therapies and Healthy Control Subjects

Figures 2A and 3A illustrate flow cytometric detection of CD3⁺CD4⁺Foxp3⁺ Treg and CD3⁺CD4⁺IL17A⁺ Th17 cells in peripheral blood. Figures 2B and 2D and 3B and 3D show comparison of CD3⁺CD4⁺Foxp3⁺ Treg and CD3⁺CD4⁺IL17A⁺ Th17 cell percentages and mean fluorescence intensities (MFI) among patients with an acute attack of BD uveitis despite conventional immunomodulatory therapies and healthy control subjects. Treg and Th 17 cell percentages were significantly higher in patients when compared to controls (6.76 ± 1.13 vs. 2.48 ± 0.35, and 35.58 ± 7.67 vs. 13.42 ± 3.06, respectively; P < 0.001). Median RORγt mRNA expression by Th17 cells was significantly elevated in patients (4.84 [range, 1.15–539.34] vs. 1, P < 0.001). 

Figure 2

 

Detection of CD3⁺CD4⁺Foxp3⁺ Treg cells in peripheral blood in one representative patient with BD uveitis and corresponding control subject. Monoclonal mouse anti-human CD3 PECy7, CD4 FITC, CD25 APC, and Foxp3 PE antibodies were used. The assays were reproducible. Each sample's cutoff for double-positive gate was determined based on the sample's negative control. Numbers in the histograms indicate the percentages of cells positive for Foxp3 and CD4 (A). Comparison of peripheral blood CD3⁺CD4⁺Foxp3⁺ Treg cell percentages and mean fluorescence intensities (MFI) between 27 BD patients who had an acute attack of uveitis despite conventional immunomodulatory therapies and healthy control subjects (B, D); and in 19 BD patients at the initiation of interferon alfa-2a therapy and during uveitis remission with ongoing therapy (C, E). Cell percentages and MFI values are presented as mean ± SD (B–E).

Figure 3

 

Detection of CD3⁺CD4⁺IL17A⁺ Th17 cells in peripheral blood in one representative patient with BD uveitis and corresponding control subject. After Th17 cell culture and in vitro stimulation, monoclonal mouse anti-human CD3 PECy7, CD4 FITC, CD25 APC, and IL-17A PE antibodies were used. The assays were reproducible. Each sample's cutoff for double-positive gate was determined based on the sample's negative control. Numbers in the histograms indicate the percentages of cells positive for IL-17A and CD4 (A). Comparison of peripheral blood and CD3⁺CD4⁺IL17A⁺ Th17 cell percentages and mean fluorescence intensities (MFI) between 27 BD patients who had an acute attack of uveitis despite conventional immunomodulatory therapies and healthy control subjects (B, D); and in 19 BD patients at the initiation of interferon alfa-2a therapy and during uveitis remission with ongoing therapy (C, E). Cell percentages and MFI values are presented as mean ± SD (B–E).

Patients had significantly lower IL-10 concentration in supernatants of CD4⁺CD25⁺ Treg cell cultures than control subjects (P = 0.022, Fig. 4A). IL-17, IL-6, IL-21, IL-22, IL-23, IFN-γ, and TNF-α cytokine levels in CD4⁺CD25⁻ Th17 cell culture supernatants were significantly higher in patients when compared to control subjects (P ≤ 0.01, Figs. 5A–G). There were no significant differences in Treg TGF-β (2503.87 ± 271.31 vs. 2504.18 ± 384.50, P = 0.634), Treg IL-35 (Fig. 4B), Treg IL-2 (Fig. 4C), and Th17 IL-2 (139.41 ± 457.28 vs. 63.04 ± 134.17, P = 0.065) concentrations between patients and controls. IL-17, IL-6, IL-21, IL-22, IL-23, IFN-γ, and TNF-α levels in CD4⁺CD25⁺ Treg cell supernatants and IL-10, TGF-β, and IL-35 levels in CD4⁺CD25⁻ Th17 cell supernatants were under the detection limits of ELISA kits. 

Figure 4

 

Comparison of cytokine levels in supernatants of CD4⁺CD25⁺ Treg cell cultures between 27 BD patients who had an acute attack of uveitis despite conventional immunomodulatory therapies and healthy control subjects (A–C); and in 19 BD patients at the initiation of interferon alfa-2a therapy and during uveitis remission with ongoing therapy (D–F). Cytokine concentrations are presented as mean ± SD.

Figure 5

 

Comparison of cytokine levels in supernatants of CD4⁺CD25⁻ Th17 cell cultures between 27 BD patients who had an acute attack of uveitis despite conventional immunomodulatory therapies and healthy control subjects (A–G); and in 19 BD patients at the initiation of interferon alfa-2a therapy and during uveitis remission with ongoing therapy (H–N). Cytokine concentrations are presented as mean ± SD.

TLR-2, TLR-3, TLR-4, TLR-8, and TLR-9 mRNA expression by peripheral blood CD4⁺ T cells and monocytes was significantly increased in patients compared to controls (P < 0.001, Figs. 6, 7). 

Figure 6

 

Comparison of TLR mRNA expression by CD4⁺ T cells between 27 BD patients who had an acute attack of uveitis despite conventional immunomodulatory therapies and healthy control subjects (A–E); and in 19 BD patients at the initiation of interferon alfa-2a therapy and during uveitis remission with ongoing therapy (F–J). The mRNA expressions are presented as median. TLR expression of patients was normalized to healthy control subjects (asterisks).

Figure 7

 

Comparison of TLR mRNA expression by monocytes between 27 BD patients who had an acute attack of uveitis despite conventional immunomodulatory therapies and healthy control subjects (A–E); and in 19 BD patients at the initiation of interferon alfa-2a therapy and during uveitis remission with ongoing therapy (F–J). The mRNA expressions are presented as median. TLR expression of patients was normalized to healthy control subjects (asterisks).

Comparison of Study Outcomes in 19 BD Patients at the Initiation of Interferon alfa-2a Therapy and During Uveitis Remission With Ongoing Therapy

Figures 2C and 2E and 3C and 3E show comparison of CD3⁺CD4⁺Foxp3⁺ Treg and CD3⁺CD4⁺IL17A⁺ Th17 cell percentages and MFIs at the initiation of interferon alfa-2a for an acute attack of BD uveitis and after achievement of uveitis remission. A significant decrease in Treg and Th17 cell percentages was observed during uveitis remission with ongoing interferon alfa-2a treatment (6.86 ± 1.11 vs. 2.93 ± 0.84, and 33.52 ± 4.38 vs. 13.27 ± 6.04, respectively; P < 0.001). Median Th17 RORγt mRNA expression decreased significantly with interferon alfa-2a treatment (4.93 [range, 1.15–85.13] vs. 0.06 [range, 0.01–2.35], P < 0.001). 

There was a significant increase in IL-10 and IL-35 and a significant decrease in IL-2 expression by Treg cells during interferon treatment (P ≤ 0.005, Figs. 4D–F). Interferon alfa-2a treatment led to a significant decrease in IL-17, IL-6, IL-21, IL-22, IL-23, IFN-γ, and TNF-α cytokine levels in CD4⁺CD25⁻ Th17 cell supernatants (P ≤ 0.003, Figs. 5H–N). There were no significant differences in Treg TGF-β (2541.22 ± 271.89 vs. 2463.82 ± 234.14, P = 0.212) and Th17 IL-2 (138.91 ± 377.07 vs. 98.44 ± 194.42, P = 0.748) concentrations when compared before and after interferon alfa-2a treatment. 

The mRNA expression of all TLRs by peripheral blood CD4⁺ T cells and monocytes decreased significantly during interferon alfa-2a–induced remission of BD uveitis (P < 0.001, Figs. 6, 7). 

When intercorrelations between the total interferon alfa-2a dose administered and study parameters were calculated, an increase in the total interferon alfa-2a dose administered was found to be associated with a moderately significant decline in IL-2 (−0.496, P = 0.031), IL-21 (−0.446, P = 0.043), IL-23 (−0.551, P = 0.01), and TNF-α (−0.663, P = 0.002) levels in Th17 cell supernatants. 

Comparison of Study Outcomes Between BD Patients Who Achieved Uveitis Remission With Interferon Therapy and Those Who Failed

Three out of four patients in whom interferon treatment was ineffective agreed to provide follow-up blood samples at the termination of treatment. Comparison of study parameters at the initiation of interferon treatment revealed a significantly lower CD3⁺CD4⁺IL-17A⁺ Th17 cell percentage and IL-23 expression by CD4⁺CD25⁻ Th17 cells in 19 patients in whom interferon led to remission of uveitis when compared to 3 patients in whom interferon treatment failed (33.52 ± 4.38 vs. 52.33 ± 4.56, P = 0.001, and 32.79 ± 22.55 vs. 77.36 ± 34.77, P = 0.021; respectively). We did not observe any significant difference in study outcomes between the follow-up blood samples in patients who achieved uveitis remission with interferon and in those who failed. 

Discussion

In this study, we found an increase in peripheral blood Foxp3⁺ Treg and IL-17A⁺ Th17 cells, and an increased expression of Th17 RORγT mRNA in patients with an acute attack of BD uveitis while on conventional immunomodulatory therapies. Despite an increase in Treg cells, in vitro expression of IL-10, a regulatory cytokine with anti-inflammatory effects, by Treg cells was decreased, suggesting a dysfunction in Treg cells leading to defective control of inflammatory responses. Our finding of an increased frequency of Th17 cells and their increased expression of proinflammatory cytokines in patients with active BD uveitis confirms previous studies that suggest a significant role of Th17 immune responses in BD.^12–18 We also observed an increase in the expression of TLR-2, TLR-3, TLR-4, TLR-8, and TLR-9 by peripheral blood CD4⁺ T cells and monocytes during an acute attack of BD uveitis. Follow-up investigations after interferon alfa-2a treatment, which resulted in clinical remission of uveitis in 70.4% of the patients, revealed resolution of pathologic adaptive and innate immune system responses that included induction of IL-10 expression by Treg cells, suppression of Th17 cells, and reduced expression of TLRs on CD4⁺ T cells and monocytes. 

Our study revealed an increased frequency of peripheral blood CD3⁺CD4⁺Foxp3⁺ Treg cells in patients with active BD uveitis. Nanke and associates⁸ demonstrated that the levels of CD4⁺CD25⁺ bright Treg cells were significantly decreased in the peripheral blood before an attack of uveitis in patients with BD; however, the authors did not evaluate Treg cells during the acute uveitis episode. Sugita and associates⁹ showed that patients with active BD uveitis had a lower proportion of peripheral blood CD4⁺ Foxp3⁺ Treg cells in comparison to healthy and diseased controls. They also reported a significantly lower proportion of Treg cells during the active stage than in the remission of BD uveitis.⁹ The authors' definition of active uveitis was based on the occurrence of at least one attack in the last 3 months of follow-up, which may explain the discrepancy between their findings and the results of the present study, in which blood sampling was done within 1 week of symptom onset. While we found a significant decline in the number of Treg cells during clinical remission of uveitis induced by interferon alfa-2a treatment, possibly reflecting a decreased demand to control the activity of effector T cells, Sugita and associates⁹ reported that a higher population of Treg cells during infliximab treatment was associated with better suppression of recurrent uveitis attacks. In a study showing results similar to ours, Hamzaoui and associates¹⁰ revealed an elevated CD4⁺CD25⁺ Treg cell percentage and Foxp3 mRNA expression in BD patients who had active systemic and intraocular inflammation when compared to healthy control subjects and to patients with BD who were in remission. Our patients did not have severe active systemic inflammation as they were either on azathioprine or cyclosporine A or a combination treatment thereof. Demir and associates¹¹ also showed significantly higher CD4⁺CD25⁺ Foxp3⁺ Treg cell counts in 25 patients with systemically active BD, 13 of them having active uveitis, and in 25 patients with BD in remission stage when compared to controls. A relative increase in the number of Treg cells would be expected in inflammatory diseases due to an increase in polarized effector T cells, and we believe that it is not their scarcity but their dysfunctionality that leads to a pathologic state in patients with BD uveitis. There may also be a disruption in the recruitment of Treg cells to the inflamed tissues and the eye in patients with BD. Our finding of a significantly lower expression of Foxp3 MFI by Treg cells, despite a relative increase in Treg cell counts, implicates a functional impairment in Treg cells in patients with BD uveitis. 

We also demonstrated that in vitro IL-10 concentration in the supernatants of CD4⁺CD25⁺ Treg cell cultures was significantly lower in patients with BD uveitis when compared to controls. A genome-wide association study (GWAS) revealed a single nucleotide polymorphism in the gene encoding IL-10 in patients with BD, suggesting that a variant associated with lower production of IL-10 can predispose to BD.³⁵ The findings of the current study support this observation in the BD GWAS study and associate lower concentrations of IL-10 with active BD. We assume that despite an increase in the frequency of Foxp3⁺ Treg cells in the peripheral blood, they have a functional defect reflected by decreased IL-10 expression. We speculate that interferon alfa-2a induces remission of uveitis by leading to a significant increase in the expression of the anti-inflammatory cytokines IL-10 and IL-35 by Treg cells after regaining their function during treatment. 

In our study, interferon alfa-2a treatment resulted in the reversal of all pathologic Th17 cell immune responses observed in patients with active BD uveitis despite conventional immunomodulatory therapies. We found a significant reversal of the increased frequency of peripheral blood CD3⁺CD4⁺IL17A⁺ Th17 cells, increased expression of Th17 RORγt mRNA, and their proinflammatory cytokine production potential. Lightman and associates³⁶ randomized patients with BD to receive pegylated interferon alfa-2b or not in addition to their standard care therapy, which included conventional immunomodulatory agents and systemic corticosteroids. Pegylated interferon significantly lowered systemic corticosteroid requirement and was associated with a significant decrease in Th17 cells and an increase in Treg cells. However, their results are not directly comparable to ours as only 42% of their patients had eye disease and as patients were not switched to interferon; rather, interferon was added to their immunosuppressive regimen. Sugita and associates²⁰ showed that Th17 cells from peripheral blood of patients with active BD uveitis exposed to infliximab in vitro failed to produce IL-17 and showed diminished expression of Th17 RORγt, suggesting that TNF-α plays a role in Th17 polarization. These authors also showed favorable effect of infliximab on IL-17 and CD4⁺IL17⁺ Th17 cell immune responses in ocular fluids of patients with BD uveitis and in an animal model of experimental autoimmune uveitis. Both infliximab and interferon alfa-2a, which appear to inhibit pathologic Th17 responses in patients with BD uveitis, have been included in European League Against Rheumatism recommendations as equally preferable biologic agents for severe ocular involvement in BD.³⁷ However, we previously emphasized the advantages of interferon alfa-2a over infliximab therapy in patients with BD uveitis.³⁸ 

An important finding of this study was that CD3⁺CD4⁺IL17A⁺ Th17 cell percentage and IL-23 expression by CD4⁺CD25⁻ Th17 cells were significantly lower at the initiation of interferon treatment in patients in whom interferon alfa-2a induced clinical remission compared to those in whom treatment failed. This finding suggests that peripheral blood Th17 cell count might be used as a predictor of disease severity as well as the response to a standard dose of interferon alfa-2a treatment in patients with BD uveitis. We speculate that patients with higher numbers of Th17 cells may require higher doses of interferon alfa-2a or alternative treatments, and that the potential of this finding to serve as a biomarker to guide treatment in patients with BD uveitis needs to be explored in a larger number of patients. 

Our study is the first to evaluate the effect of interferon alfa-2a treatment on the expression of a large set of TLRs by peripheral blood CD4⁺ T cells and monocytes in patients with BD uveitis. We demonstrated significantly higher expression of TLR-2, TLR-3, TLR-4, TLR-8, and TLR-9 mRNA by CD4⁺ T cells and monocytes in patients with active BD uveitis as compared to healthy control subjects. Interferon alfa-2a led to a significant decrease in expression of studied TLRs on CD4⁺ T cells and monocytes after clinical remission of uveitis was achieved. Liu and associates²⁶ documented a markedly higher expression of TLR-2, TLR-3, TLR-4, and TLR-8 mRNA from PBMCs in patients with active BD uveitis. A significant cell surface positivity of these TLRs on PBMCs, and on isolated CD4⁺ T cells and monocytes, was documented as well. TLRs exhibit a critical role in innate immunity against pathogens; and infectious agents, particularly Streptococcus spp. and herpes simplex virus type 1, have long been seen as a triggering factor in the pathogenesis of BD.^21,39 Monocytes play important roles in host defense as critical cellular components of the innate immune system, and TLR signaling from CD4⁺ T cells has been shown to promote Th17 responses in autoimmune disease.^40,41 However, it is still not completely clear whether an intrinsic enhanced TLR expression in BD patients makes them more susceptible to the recognition of pathogens or not, and how exactly infection serves as a trigger for a dysregulated immune response in BD. 

Limitations of this study are lack of a diseased control group and variable total dose and duration of interferon alfa-2a treatment administered at follow-up blood sampling. Although fluorescein angiographic remission is the ultimate goal in the treatment of BD uveitis, the timing of follow-up blood sampling was not based on achievement of angiographic remission. However, we observed a statistically significant decrease in the total fluorescein angiography scores in the angiograms obtained after interferon-induced clinical remission of uveitis was achieved. Another limitation of this study was the small number of patients who failed interferon therapy, which did not allow us to draw strong conclusions from the outcomes in this subgroup. 

This novel study that assesses the effect of interferon alfa-2a in patients with active BD uveitis despite immunomodulatory therapies shows that interferon alfa-2a treatment leads to a decline in the number of dysfunctional Treg cells in patients with refractory BD uveitis, along with control of disease activity, and restores their function as reflected by an increase in IL-10 production. Interferon alfa-2a also suppresses aggravated Th17 immune responses and the expression of TLRs on CD4⁺ T cells and monocytes. The findings of the present study suggest that interferon alfa-2a might orchestrate immune system interactions to restore homeostasis such as those established for Treg/Th17 cell plasticity, Foxp3/RORγt interaction, and TLR-mediated regulation of Treg cell function.^21,42 Further studies are required to assess the effect of type 1 interferon therapy on immunologic features of ocular fluids and tissues in patients with BD uveitis and to investigate their correlation with the findings observed in the peripheral blood. 

Acknowledgments

The authors thank medical students Burak Zabun and Kahraman Aksoy for their participation in laboratory studies. 

Supported by The Scientific and Technological Research Council of Turkey (TUBITAK Research Grant Project 1001, Project No: 115S541) 

Disclosure: O. Albayrak, None; M. Oray, None; F. Can, None; G. Uludag Kirimli, None; A. Gul, None; I. Tugal-Tutkun, None; S. Onal, None