Background
Age related macular degeneration (AMD) is clinically characterized by degenerative changes in the macula, the region of the retina that permits fine central vision. One of the key pathological features of AMD is the development of large drusen, extracellular deposits located between Bruch's membrane and the retinal pigment epithelium (RPE). These large drusen and the associated RPE changes are the major risk factors for the development of advanced AMD, which can be classified into two subtypes: dry (geographic atrophic) and wet (neovascular) [
1]. Inflammation has been suggested to play an important role in AMD pathogenesis [
2,
3].
Genetic studies have demonstrated strong associations between AMD and several gene variants in genes coding for complement proteins, including complement factor H (CFH), factor B/C2, and C3 [
4‐
12]. CFH is a factor that down-regulates complement activation. It is commonly thought that CFH polymorphism leads to dysregulation of alternative complement activation which may contributes to AMD pathogenesis [
13]. However, the mechanism by which CFH regulates AMD progress is still not clear. Systemic activation of the complement cascade has been implicated in AMD patients [
14‐
16]. C5a, among many alternative complement activation molecules, are elevated in peripheral blood of AMD patients [
15,
16]. Locally, C5a and C3a accumulate in drusen and are shown to promote choroidal neovascularization (CNV) [
17], which is the hallmark of wet AMD.
Recently, a subset of effector helper T cells, IL-17-producing T cell (Th17), is implicated in the pathogenesis of various autoimmune diseases including uveitis, arthritis, multiple sclerosis, psoriasis and inflammatory bowel disease [
18‐
20]. Proinflammatory cytokines, including IL-1β, IL-6, IL-23, IL-21 and TNFα, as well as transcription factor RORC, are responsible for differentiation and maintenance of Th17 cells within human body [
21‐
24]. Recent evidence from the mouse suggests that C5a provides both costimulatory and survival signals to CD4
+ T cells and induces Th17 cytokine expression [
25,
26]. However, it is still not clear if C5a can impact human T cells and if Th17 cells are associated with AMD.
Here we showed that C5a protected human CD4
+ T cells from undergoing apoptosis and C5a promoted IL-22 and IL-17 expression from CD4
+ T cells of AMD patients and normal subjects as well. Intriguingly, consistent with previous observation of elevated C5a expression in the serum of AMD patients [
15,
16], we found significantly increased levels of IL-22 and IL-17 in the sera of AMD patients, suggesting possible roles of IL-22 and IL-17 in the inflammation that contributes to AMD.
Methods
Patients
PBMCs were obtained from the peripheral blood of AMD patients and healthy subjects in compliance with institutional review board (IRB) protocols after informed consent at the National Institutes of Health (NIH). The written consents were obtained. Our study has obtained ethics approval from the neuroscience IRB of NIH. AMD subjects were diagnosed with wet AMD without accompanied systemic autoimmune diseases or other immune-related diseases, as well as polypoidal vasculopathy by experienced clinicians. We excluded patients with a history of cancer within the past 5 years or patients with active inflammatory diseases. Clinical characteristics, demographic data, and single-nucleotide polymorphism information of complement associated molecules are provided in Table
1 and
2.
Table 1
Clinical information of AMD patients
1 | W, 94, F | Wet OU | TC | GG | CC | anti-VEGF, PDT | 19 | H |
2 | W, 80, F | Wet OU | TT | GG | CC | anti-VEGF, PDT | 8 | W, L, X |
3 | W, 97, F | Wet OU | TC | GG | CC | anti-VEGF, Isteroids | none | W |
4 | W, 92, F | Wet OU | TC | GG | CC | anti-VEGF | 14, 16, 21 | Z, G, O, BB |
5 | W, 91, M | Wet OU | TC | GG | CC | anti-VEGF | 2, 7, 18, 14 | W, H, T, G, D, C, B |
6 | W, 91, M | Wet OU | TC | GG | CG | anti-VEGF | 10, 11, 14, 15 | G, C, H, B, E |
7 | W, 83, F | Wet OU | CC | GG | CG | anti-VEGF, PDT | 8, 14, 16 | W, Z, G |
8 | W, 79, F | Wet OD, Dry OS | CC | GG | CG | anti-VEGF, Laser Rx | 14, 15, 21 | W, C, G, M, BB |
9 | W, 74, M | Wet OU | CC | GG | GG | anti-VEGF, Laser Rx | 7, 11, 14, 15 | W, BB, E, Y, D, H, L, C, S |
10 | W, 77, M | Wet OD, Dry OS | TC | GG | CG | anti-VEGF, Isteroids | 14, 15 | W, L, C, H, R |
11 | W, 74, M | Dry OD, Wet OS | TT | GG | CC | anti-VEGF | 3, 15, 16 | W, Z, A |
12 | W, 82, F | Wet OU | TT | GG | GG | anti-VEGF, PDT | 10, 11 | W, E, BB, L, F |
13 | W, 81, F | Wet OU | CC | GG | CG | anti-VEGF, laser Rx | 7, 14, 15 | C, G, V |
14 | W, 75, M | Wet OD, Dry OS | TC | GG | CC | None | 14, 15, 23 | C, G, H |
15 | W, 67, M | Wet OD, Dry OS | TC | GG | CC | anti-VEGF, PDT | 8, 15, 20 | W, L, X, C, H, F |
16 | W, 72, F | Wet OU | TC | GG | CC | anti-VEGF, Isteroids | 12, 16 | Z, I, G, W |
17 | W, 74, M | Wet OU | CC | CG | GG | anti-VEGF, laser Rx, PDT | 20 | F, V |
18 | W, 77, F | Wet OU | CC | - | CC | anti-VEGF, PDT, Isteroid | 8 | X |
19 | W, 75, M | Dry OD, Wet OS | CC | GG | CC | anti-VEGF | 11, 15 | W, E, C, L |
20 | W, 82, M | Wet OU | - | - | - | none | 3, 5, 6, 7, 11, 19, | P, B, G, A, E, D |
21 | W, 72, M | Wet OU | CC | GG | CG | anti-VEGF, Isteroid | 7, 14 | H, G, D |
22 | W, 83, M | Wet OD, Dry OS | TC | GG | CC | anti-VEGF, Isteroid | 14, 15 | W, L, H, G, C |
23 | W, 83, M | Wet OD, Dry OS | TT | GG | GG | anti-VEGF, PDT | 14, 15, 22 | B, W, G, C, |
24 | W, 88, F | Dry OD, Wet OS | TC | GG | CC | anti-VEGF | 19 | W, H, |
25 | W, 70, M | Wet OD, Dry OS | TC | GG | CC | anti-VEGF | 10 | W |
26 | W, 83, F | Wet OD, Dry OS | TT | GG | CC | anti-VEGF, Isteroids | 8, 13, 14, 17, | Q, G, Z, W |
27 | W, 80, M | Dry OD, Wet OS | CC | GG | CG | anti-VEGF | 7, 8, 14, 15, 21, 25 | W, L, BB, Q, G, C, M, AA, D, BB |
28 | W, 95, M | Dry OD, Wet OS | TT | CG | CC | anti-VEGF | 1, 14, 15, 22 | W, E, B, K, BB, C, G, T, |
29 | W, 79, F | Wet OU | CC | GG | CG | anti-VEGF, PDT | 7, 21, 24 | W, L, D, V, CC, N |
30 | W, 84, M | Wet OU | TC | GG | CC | PDT, Laser Rx | 5, 8 | G, H, C, X, A, |
31 | W, 80, M | Wet OD, Dry OS | CC | CG | CG | anti-VEGF | 4, 8, 15 | W, BB, L, B, E, N, D, X |
32 | W, 97, F | Wet OU | - | - | - | None | 2 | J, B |
33 | W, 77, M | Wet OU | TC | GG | CC | anti-VEGF, Laser Rx | 11 | W, L, E |
34 | W, 57, M | Wet OU, Wet OS | TT | GG | CG | anti-VEGF | 3, 7, 14, 15, 26 | A, C, D, G, BB, DD |
35 | W, 67, F | Wet OU | CT | GG | CG | anti-VEGF | 14, 15, 16, 21 | G, M, Z, BB, EE |
36 | W, 82, F | Wet OU | CT | CG | CG | anti-VEGF | 7, 8, 14, 15, 16 | C, G, K, Q, Z, BB |
37 | W, 84, F | Wet OS | CC | GG | CC | anti-VEGF | 15, 16 | CC, L, G, Z, |
38 | W, 83, F | Wet OU | CT | GG | CG | anti-VEGF | 9, 14, 15, 28 | B, C, BB |
39 | W, 84, F | Wet OS | TT | GG | CC | anti-VEGF | 16, 27 | Z, CC |
40 | W, 90, F | Wet OU | CC | GG | CC | anti-VEGF | 27 | B, G, H, O, BB, CC |
Table 2
Healthy Donor Information
1 | W, 61, F | CT | CG | CC |
2 | W, 72, M | CT | GG | CC |
3 | W, 69, M | - | GG | GG |
4 | W, 71, F | CT | GG | CG |
5 | W, 75, M | TT | GG | CC |
6 | W, 65, F | TT | GG | CC |
7 | W, 66, F | TT | GG | CC |
8 | W, 66, M | CT | GG | CC |
9 | W, 73, M | TT | GG | CC |
10 | W, 61, M | CC | GG | CC |
11 | W, 69, M | TT | GG | CG |
12 | W, 73, F | CT | GG | CC |
13 | W, 65, F | CT | GG | CC |
14 | W, 75, F | CT | GG | CC |
15 | W, 69, M | CT | GG | CC |
16 | W, 65, F | CC | GG | GG |
17 | W, 65, F | CT | GG | CC |
18 | W, 65, F | - | CG | CG |
19 | W, 62, F | TT | CG | CC |
20 | W, 71, M | TT | GG | CG |
21 | W, 72, M | CT | GG | CC |
22 | W, 62, M | TT | GG | CC |
23 | W, 71, F | CC | GG | CG |
24 | W, 66, F | TT | GG | CG |
25 | W, 65, F | CT | GG | CC |
26 | W, 61, F | CT | GG | CC |
27 | W, 63, F | CT | CG | CC |
28 | W, 64, F | CT | GG | CC |
29 | W, 68, M | CT | CG | CG |
30 | W, 70, M | CT | GG | - |
31 | W, 87, F | - | - | - |
32 | W, 59, M | CC | CC | CC |
33 | W, 64, F | - | - | CC |
34 | W, 61, M | TT | GG | CG |
35 | W, 66, M | CC | CG | CG |
36 | W, 65, M | CC | GG | CG |
37 | W, 65, F | CT | GG | GG |
38 | W, 60, F | CT | GG | CG |
39 | W, 63, F | TT | GG | CG |
40 | W, 66, F | CT | CC | CC |
41 | W, 77, F | - | - | - |
42 | W, 65, M | CT | GG | CC |
43 | W, 66, M | CT | GG | CG |
44 | W, 62, M | CT | GG | CC |
45 | W, 66, M | - | - | - |
Cell sorting
To sort CD4+ T cells and monocytes, 1 × 107 PBMCs were stained with allophycocyanin-labeled CD3 (clone UCHT1, BD Biosciences), PE-labeled CD4 (clone RPA-T4, BD Biosciences), or FITC labeled CD14 (clone M5E2, BD Biosciences) for 20 minutes in 1% BSA PBS staining buffer. Cells were then washed and subsequently sorted on a FACS Aria (BD Biosciences). BD FACSDiva software was used to sort the cells.
Cell culture and flow cytometry
PBMC cells were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum (Gemini Bioproducts, West Sacramento, CA) supplemented with 2 mM glutamine and 1× antibiotics. For T cell and monocytes separation, PBMCs were cultured in the same RPMI medium described above and then stained with anti-CD3 and anti-CD4 antibodies for T cell and anti-CD14 for monocyte separation. Cells were treated with or without C5a (50 ng/ml from R&D Systems, endotoxin level <1.1 EU per 1 μg of protein) and a C5aR antagonist (2.5 ug/ml from Jerini Ophthalmic Inc, also called JPE-1375, is a hexameric linear peptidomimetic molecule that inhibits C5a binding to the human C5aR). Anti-B7.1 and B7.2 antibodies (10 μg/ml of each) or anti-IL-1β (10 μg/ml) and anti-IL-6 (10 μg/ml) neutralization antibodies were added into the cell culture in indicated experiments. Intracellular staining was performed after 5 days of C5a culture. Cells were stimulated with PMA (10 ng/ml), ionomycin (0.5 μg/ml) and Golgistop for 4 hours at 37°C before intracellular staining. 5 × 105 cells were stained with FITC labeled CD45RA (clone HI100, BD Biosciences), PE-IL-22 (clone 22URTI, eBioscience), or PE-IL-17A (clone eBio64DEC17, eBioscience), perCP-CD4 (clone SK3, BD Biosciences) and allophycocyanin-labeled CD3. The intracellular staining procedure was based on the BD Bioscience protocol. Briefly, cells were firstly stained with cell surface markers (anti-CD4, anti-CD45RA), and then permeabilized and proceeded to intracellular staining (anti-LI-17A or IL-22). Cells were acquired by a FACSCalibur flow cytometer (BD Biosciences) and analyzed by FlowJo software (TreeStar, San Jose, CA).
Cytokine Analysis
Sera from patients or supernatants collected from cell culture were tested by ELISA for IL-22 and IL-17, or sent for multiplex cytokine analysis (Aushon Biosystems). IL-22 and IL-17A ELISA kits were purchased from R&D Systems, Inc. (Minneapolis, MN) and were performed based on kit protocols.
Apoptosis Assay
Apoptotic cells were detected by staining cells with both the annexin-V-FITC (BD Biosciences) and TUNEL technology (Roche, Indianapolis, IN) according to the manufacturer's instructions. Phospho-Bad expression was detected by western blot using anti-Phospho-Bad antibody (Cell Signaling Technology).
SDS-PAGE and Western blotting
A total of 5 million T cells were lysed in 100 μl lysis buffer [50 mM Tris-Cl, 1% Triton X-100, 100 mM NaCl, 2 mM EDTA, 50 mM NaF, 50 mM glycerol-phosphate, 1 mM NaVO4 and 1× protease inhibitor cocktail (Roche)]. Complete cell lysis was achieved by immediately vortexing the cells and then boiling in an equal amount of 2 × SDS protein loading buffer at 95°C for 5 minutes. Cell debris was removed by centrifugation at 12, 000 rpm for 3 min. Twenty microliter of each sample was loaded into a 12% SDS-polyacrylamide gel containing a 4% stacking gel. Immunoblotting was carried out. Primary antibodies of anti-Phospho-Bad, anti-Bad were purchased from Cell Signaling Technology (Beverly, MA). Anti-β-actin antibody was from Santa Cruz Biotechnology, Inc.(Santa Cruz, CA).
SNP Genotyping
Genomic DNA was extracted from the peripheral blood of each individual using Promega Wizard Genomic DNA Purification kit. The samples were analyzed by TaqMan genotyping assay using the Real-time PCR system 7500 (Applied Biosystems, Foster City, CA, USA). The primers and probes for C2/CFB rs9332739 and C3 rs2230199 were from the inventory SNP assay while CFH rs1061170 were custom-designed from Applied Biosystems. Genotypes were determined based on the fluorescence intensities of FAM and VIC. The call rates of 3 assays were >98.5% and the call accuracies (consistency of duplicate wells of selected samples) were 100%.
Statistical Analysis
Non-parametric methods (Wilcoxon two-sample test) were used since the expression of IL17 and IL22 does not follow a parametric distribution. To evaluate if the expression of these 2 cytokines follows a normal distribution, we visually checked the histograms as well as used the Kolmogorov-Smirnov method. For the association study between IL-22/IL-17 and some characteristics of patients (CFH, C2/CFB, C3 genotypes, gender, co-morbidities of diabetes, hypertension and hypercholesterolemia), Wilcoxon's nonparametric two-sample rank sum test was used. Age was analyzed using Pearson correlation. The software used for all the analyses was "The SAS System", version 9.2.
Discussion
In this study, we have provided evidence that C5a induced IL-22 and IL-17A expression from human CD4
+ T cells. Importantly, consistent with previous observations of elevated C5a expression in the serum of AMD patients from different cohorts [
15,
16], we observed significantly increased levels of IL-22 and IL-17A in the sera of AMD patients. However, so far, we do not have direct evidence showing that the elevated Th17 cytokine levels in AMD patients' sera are due to higher C5a expression in AMD patients. C5a may be one of the many factors related to this observed effect. Other unknown factors may also contribute to this T cell activation seen in AMD patients. Interestingly, the findings that C5a specifically promoted the Th17 family cytokine production, but not IFNκ nor IL-4, also correlated with the fact that there were similar IFNκ and IL-4 levels in the sera of AMD patients as compared to controls. The dysregulation of the complement system has been linked to multiple neurodegenerative diseases including Alzheimer's disease, Parkinson's disease, as well as AMD [
29]. The induction of inflammatory Th17 cytokines, including IL-22 and IL-17, by complement component C5a could potentially elucidate the general mechanism by which inflammation contributes to the pathogenesis of these diseases previously referred to as degenerative. Our results support a role for C5a in protecting CD4
+ T cells from undergoing apoptosis (Figure
3). These findings suggest that the enhanced effector T cell function by C5a is at least partially mediated by limiting pro-inflammatory cell death.
We found that monoctytes are necessary for C5a induced Th17 cytokine production through two mechanisms: 1) promoting the direct interaction between monocytes and T cells; 2) indirectly stimulating the production of IL-1β and IL-6 from monocytes. C5a can bind to the trans-membrane receptors C5aR/CD88 and C5L2 (GPR77) which are expressed on monocytes. C5L2 is expressed at much lower levels as compared to CD88. C5a binding to CD88 leads to a number of functional changes including activation of inflammation. However, the pathophysiological role of C5L2 is currently controversial with both pro-inflammatory and anti-inflammatory roles reported [
30]. Previous reports from rodent models have shown that C5a has a direct effect on T cells by interacting with the C5a receptor expressed on T cells, a finding which is different from what we have observed in humans [
26]. Fang
et al. recently demonstrated that C5a itself has no effect on Th17 cytokine production in mouse [
31]. However, C5a synergizes with TLR4 to produce serum factors that drive Th17 cell differentiation [
31]. Liu
et al. reported that local interactions among C3a/C5a, C3aR/C5aR, antigen presenting cells (APC) and T cells are important for IFNκ and IL-17 production of T cells in a murine EAE (Experimental autoimmune encephalomyelitis) model [
32]. In another murine sepsis model, Xu
et al. shows that C5a affects the crosstalk between DC and gamma/delta T cells and results in a large production of IL-17[
33]. In a human study, Hueber and colleagues showed that C5a induces IL-17 from human mast cells [
34]. Our work is the first human study showing that monocytes play an essential role in C5a promoted expression of Th17 cytokines from CD4
+ T cells.
Several research teams have reported that a common SNP of CFH, Tyr
402His, has a particularly strong association with AMD [
4,
6,
8]. We sub-grouped IL-22/IL-17 expression based on the subjects' CFH SNP genotypes and found that AMD patients with higher IL-22/IL-17 cytokine expression were likely to have the risk CFH allele (TC/CC) (Figure
4). However, serum IL-22/IL-17 cytokine levels showed no difference between the two CFH genotype groups (TT versus TC/CC) in controls. These results suggest that this CFH SNP does not explain the elevated Th17 cytokine expression. However, this genetic variant may be one of the many factors influencing Th17 cytokine expression.
Dysregulation of alternative complement activation has been reported to be involved in AMD pathogenesis. The drusen of AMD donor eyes contain almost all molecules of the alternative complement pathway, including CFH, C3, C5, C3a, C5a, and the membrane attack complex (MAC) [
35‐
37]. These results suggest the role of the complement system in the eye. The products of complement activation can also be detected in the blood of AMD patients. Scholl
et al.[
16] found higher levels of alternative complement activation molecules in the blood from an AMD cohort, including Ba, C3d, MAC, C3a, and C5a. A subsequent study in a larger independent cohort of patients and controls confirmed these results, showing the activation of the alternative pathway of complement in blood is associated with genetic polymorphisms in complement factor B and increases with age [
14]. Reynolds and colleagues also found an increased plasma concentration of C5a and Bb in advanced AMD [
15]. In addition, a recent report has shown that immunization with carboxyethylpyrrole generated by oxidative damage to DHA (Docosahexaenoic acid) present in the drusen and plasma from AMD-affected individuals is sufficient to produce AMD like lesions in mice and antibody titers of carboxyethylpyrrole correlates with disease pathology, suggesting the involvement of the acquired immune pathway in disease pathology [
38]. In this study, we found C5a induced Th17 cytokine expression from human T cells in vitro, which correlates with the increased levels of Th17 cytokines in AMD blood. IL-22 has been shown to induce apoptosis of fetal retinal pigment epithelium (RPE) cells and reduce RPE cell electrical resistance in culture [
39]. However, whether systemic observations reflects pathological events in the eye and how systemic activation may ultimately be manifest in the eye remain to be defined.
To date there is no effective treatments other than attempts to slow the progression of geographic atrophy form of AMD, while neovascular AMD is treated with anti-VEGF medications injected directly into the eye [
40,
41]. Previous attempts at controlling the wet form of AMD with corticosteroid therapy have shown that the beneficial effect is transient with a significant side-effect risk profile [
42]. Health improving behavior (no-smoking), diet, and exercise may be preventive measures for AMD [
43]. Several compounds targeting complement pathway are currently in clinical trials [
13]. We recently reported that immunotherapy directed against T-cell activation resulted in patients with recurrent CNV requiring fewer injections of anti-VEGF [
44]. The dysregulation of the acquired immune pathway we describe here may provide us with new therapeutic strategies.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
BL, RBN have conceived and designed the research and drafted the manuscript; BL, LW, JT, ZL, SC have performed the experiments. CM, HNS, CCC, MLK, EC, FF have provided materials and clinical samples and help analyzed the clinical data. EA, BL performed statistical analysis. All authors read and approved the final manuscript.