Elevated Serum Osteopontin Levels and Genetic Polymorphisms of Osteopontin Are Associated with Vogt-Koyanagi-Harada Disease

Abstract

Purpose.: Osteopontin (OPN) is a proinflammatory cytokine involved in chronic inflammatory diseases. This study aimed to analyze the role of OPN in the pathogenesis of Vogt-Koyanagi-Harada (VKH) disease.

Methods.: Serum levels of OPN in VKH patients and healthy controls were assayed by enzyme-linked immunosorbent assay (ELISA). Peripheral blood mononuclear cells (PBMCs) or CD4⁺ T cells were cultured with anti-CD3 and anti-CD28 antibodies in the absence or presence of recombinant OPN for the determination of cell proliferation and cytokines. Cell proliferation was detected using a cell counting kit. Levels of interferon (IFN)-γ and interleukin (IL)-17 were detected by ELISA. Four single nucleotide polymorphisms (SNPs) of OPN and four SNPs of OPN receptors were genotyped in 601 VKH patients and 605 healthy controls using a polymerase chain reaction-restriction fragment length polymorphism assay.

Results.: OPN serum levels were significantly higher in patients with active VKH than in patients with inactive VKH and in healthy controls. PBMCs or CD4⁺ T cells cultured with recombinant OPN induced a marked cell proliferation and profound secretion of IFN-γ and IL-17 from patients with active VKH. A significantly increased frequency of the OPN rs4754 TT genotype (P = 0.004, pc = 0.048) was observed in VKH patients compared with healthy controls. No association could be detected among the four selected SNPs of OPN receptors and VKH.

Conclusions.: OPN may be relevant to the pathogenesis of VKH disease. The TT genotype of rs4754 may be a susceptible factor for VKH disease in a Chinese Han population.

Vogt-Koyanagi-Harada (VKH) disease is a chronic granulomatous inflammatory disorder that affects both eyes as well as the inner ear, skin, hair, and meninges of the brain. It is one of the most common uveitis entities in China and in the Far East. ^1,2 Although the exact etiology of VKH disease remains unknown, accumulating evidence suggests that both autoinflammatory and genetic factors are involved. ^{3 –6}  

Osteopontin (OPN) is a matricellular protein that, through interactions with its receptors integrins α4β1, α9β1, αv (β1, β3, β5), and CD44 variants, participates in a wide range of physiological and pathologic processes, including wound healing, bone turnover, tumorigenesis, inflammation, and immune responses. ^{7 –10} A number of studies have reported that the persistence of OPN expression may exacerbate chronic inflammatory diseases, such as Crohn's disease (CD), ¹¹ multiple sclerosis (MS), ¹² rheumatoid arthritis (RA), ¹³ and systemic lupus erythematosus (SLE). ¹⁴ Elevated OPN levels have been found in mice undergoing experimental autoimmune uveitis; the clinical symptoms could be ameliorated when experiments were performed in OPN knockout mice or when animals were treated with OPN antibodies. ^15,16 Studies have shown that OPN enhances T cell survival and proliferation ^17,18 and promotes the Th1 and Th17 responses during chronic inflammation. ^8,19,20 These two T-helper cell subpopulations have been shown to be involved in the pathogenesis of VKH disease. ²¹ Furthermore, OPN and genetic polymorphisms of its receptors were reported to be associated with MS, ^22,23 RA, ²⁴ SLE, ²⁵ and asthma. ^26,27 All these results suggest that OPN is possibly involved in the pathogenesis of these diseases. 

In this study we examined the expression and the possible function of OPN in VKH disease and investigated the association between the SNPs of OPN and its receptors with this disease. Our study showed that OPN levels were significantly increased in the sera of patients with active VKH disease compared with the sera of patients with inactive VKH disease and healthy controls. Recombinant (r) OPN could induce a markedly enhanced cell proliferation and interferon (IFN)-γ and interleukin (IL)-17 secretion from patients with active VKH. The OPN rs4754 TT genotype was associated with VKH disease in a Chinese Han population. 

Patients, Materials, and Methods

Patients and Controls

Forty-seven patients with VKH disease (25 men, 22 women; mean age, 34.2 years) and 41 healthy controls (22 men, 19 women; mean age, 32.9 years) were included in this study on the measurement of OPN expression and the role of OPN in T cells. The 27 patients with active VKH typically showed mutton fat keratic precipitates, cells, and flare in the anterior chamber, iris nodules, and sunset glow fundus, whereas the 20 patients with inactive VKH showed only sunset glow fundus and, in some patients, choroidoretinal atrophy. ²⁸ No immunosuppressive agents or prednisone were used in the patients with active VKH disease before referral to our hospital and blood sampling. The patients showed no active intraocular inflammation (inactive uveitis stage) for at least 3 months after treatment with prednisone alone or combined with either chlorambucil or cyclosporin A for >1 year. Blood samples were obtained from patients with inactive VKH at least 3 months after termination of all medications. Six hundred one VKH disease patients and 605 age-, sex-, and ethnically matched healthy controls (all Chinese Han population) were included for SNP analysis of OPN and its receptors. All patients and controls were enrolled at the First Affiliated Hospital of Chongqing Medical University (Chongqing, China) or the Uveitis Study Center of the Sun Yat-sen University (Guangzhou, China). The diagnosis of VKH disease was made according to the diagnostic criteria revised for VKH disease in an international committee on nomenclature. ²⁸ VKH disease patients were assessed at the time of diagnosis and are summarized in Table 1. All procedures followed the tenets of the Declaration of Helsinki. The local institutional ethics committee approved the study, and written informed consent was obtained from all the subjects. 

Table 1.

 

Clinical Characteristics of VKH Disease Patients

Clinical Features	Total n (%)
Men	314 (52.2)
Women	287 (47.8)
Uveitis	601 (100)
Headache	316 (52.6)
Tinnitus	271 (45.1)
Alopecia and poliosis	285 (47.4)
Vitiligo	133 (22.1)

 Patient age at disease onset was 34.7 ± 9.9 years.

OPN Measurement

Blood samples were centrifuged at 3000g for 10 minutes after clotting for 30 minutes at room temperature and were stored at −70°C until analysis. OPN serum levels were measured using human OPN kits (DuoSet ELISA Development; R&D Systems, Minneapolis, MN) according to the manufacturer's instructions. The limit of detection was 62.5 pg/mL. 

Proliferation Assay

Cell proliferation was measured using a cell counting (CCK-8; Sigma-Aldrich, St. Louis, MO) assay. ²⁹ PBMCs were transferred to a 96-well (200 μL/well) microplate at a density of 5 × 10⁵ cells/mL, stimulated with anti-CD3 (1 μg/mL; eBioscience, San Diego, CA) and anti-CD28 antibodies (1 μg/mL; eBioscience) in the presence or absence of rOPN (1 μg/mL; R&D Systems) at 37°C in humidified 5% CO₂ for 5 days. CCK-8 (20 μL) was added to each well and incubated for an additional 4 hours. Absorbance was measured at 450 nm on a microplate reader (SpectraMax M2e; Molecular Devices, Inc., Eugene, OR). 

Cytokine Measurements

Anticoagulated blood samples were obtained using vacuum tubes containing EDTA. PBMCs were isolated using Ficoll-Hypaque density gradient centrifugation. CD4⁺ T cells were purified by human (h) CD4 microbeads according to the manufacturer's instructions (Miltenyi Biotec, Palo Alto, CA). PBMCs and CD4⁺ T cells were stimulated with anti-CD3 (1 μg/mL; eBioscience) and anti-CD28 (1 μg/mL; eBioscience) antibodies in the presence or absence of rOPN (1 μg/mL; R&D Systems) at 37°C in humidified 5% CO₂ for 72 hours at a density of 1 × 10⁶ cells/mL. The concentration of IFN-γ and IL-17 in cell culture supernatants was measured with human OPN kits (DuoSet ELISA Development; R&D Systems) with a detection limit of 15.6 pg/mL. 

Genotyping

Genomic DNA was extracted by the QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA). The SNPs of OPN and its receptors were examined by polymerase chain reaction-restriction fragment length polymorphism assay with restriction endonucleases according to the method described previously. ³⁰ Direct sequencing was also performed by the Invitrogen Biotechnology Company (Shanghai, China) using randomly selected subjects (10% of all samples) to validate the method used in this study. 

Statistical Analysis

Statistical analysis was performed using one-way ANOVA and paired-sample t-test. Data are expressed as mean ± SD. The Hardy-Weinberg equilibrium and the frequency of genotypes and alleles were evaluated using the χ² test. Data were analyzed using statistical software (SPSS, version 13.0; SPSS Inc., Chicago, IL). P < 0.05 was considered to be statistically significant. 

Results

Expression of OPN in the Sera of VKH Patients

Sera of 21 patients with active VKH, 17 patients with inactive VKH, and 22 healthy controls were collected for cytokine assay. Concentrations of OPN were significantly elevated in patients with active VKH (5.39 ± 1.66 ng/mL) compared with those of patients with inactive VKH (4.07 ± 1.62 ng/mL; P = 0.027) and healthy controls (3.63 ± 2.01 ng/mL; P = 0.002). There was no significant difference between patients with inactive VKH and healthy controls (P = 0.445; Fig. 1) 

Figure 1.

 

OPN levels in the sera of healthy controls (n = 22), patients with inactive VKH (n = 17), and patients with active VKH (n = 21). Data are shown as mean ± SD.

Effect of OPN on the Proliferation of PBMCs

PBMCs separated from VKH patients and healthy controls were stimulated with anti-CD3 and anti-CD28 antibodies in the presence or absence of rOPN for 5 days, and cell proliferation was assayed. PBMCs from patients with active VKH showed a significantly higher cell proliferation compared with those of patients with inactive VKH and controls. OPN significantly promoted the proliferation of PBMCs from patients with active VKH (P < 0.001) but not from those of patients with inactive VKH or healthy controls (Fig. 2). 

Figure 2.

 

Effect of OPN on the cell proliferation of PBMCs of healthy controls (n = 8), patients with inactive VKH (n = 7), and patients with active VKH (n = 11). Data are shown as mean ± SD.

Effect of OPN on the Production of IFN-γ

PBMCs and CD4⁺ T cells from VKH patients and controls, cultured with OPN in the presence of anti-CD3 and anti-CD28 antibodies, were used to examine the influence of this molecule on the production of IFN-γ. IFN-γ production by both PBMCs and CD4⁺ T cells was significantly higher in patients with active VKH than in patients with inactive VKH (PBMCs, P < 0.001; CD4⁺ T cells, P < 0.001) or healthy controls (PBMCs, P < 0.001; CD4⁺ T cells, P < 0.001). OPN significantly promoted the production of IFN-γ by PBMCs (P = 0.023) and CD4⁺ T cells (P = 0.017) from patients with active VKH but not from patients with inactive VKH or healthy controls (Figs. 3A, 3B). 

Figure 3.

 

IFN-γ production by PBMCs and CD4⁺ T cells. Cells were cultured with anti-CD3 and anti-CD28 antibodies in the presence or absence of OPN for 72 hours. (A) IFN-γ production by PBMCs from healthy controls (n = 15), patients with inactive VKH (n = 8), and patients with active VKH (n = 8). (B) IFN-γ production by CD4⁺ T cells from healthy controls (n = 8), patients with inactive VKH (n = 7), and patients with active VKH (n = 8).

Effect of OPN on the Production of IL-17

PBMCs and CD4⁺ T cells from VKH patients and healthy controls cultured with OPN in the presence of anti-CD3 and anti-CD28 antibodies were used to evaluate the influence of this molecule on IL-17 production. A significantly increased production of IL-17 was observed in patients with active VKH compared with patients with inactive VKH (PBMCs, P < 0.001; CD4⁺ T cells, P = 0.002) or controls (PBMCs, P < 0.001; CD4⁺ T cells, P < 0.001). OPN significantly promoted the production of IL-17 by PBMCs (P = 0.026) and CD4⁺ T cells (P = 0.022) from patients with active VKH but not from patients with inactive VKH or healthy controls (Figs. 4A, 4B). 

Figure 4.

 

IL-17 production by PBMCs and CD4⁺ T cells. Cells were cultured with anti-CD3 and anti-CD28 antibodies in the presence or absence of OPN for 72 hours. (A) IL-17 production by PBMCs from healthy controls (n = 14), patients with inactive VKH (n = 8), and patients with active VKH (n = 8). (B) IL-17 production by CD4⁺ T cells from healthy controls (n = 10), patients with inactive VKH (n = 7), and patients with active VKH (n = 8).

Genetic Polymorphisms of OPN and Its Receptors in VKH Patients

Blood samples from 601 VKH patients and 605 healthy controls were genotyped for OPN rs4754, rs9138, rs1126616, and rs1126772 polymorphisms and for its receptors integrin α4 (ITGA4) rs1449263, integrin αv (ITGAV) rs3738919, integrin β3 (ITGB3) rs5918, and CD44 rs8193 polymorphisms. The genotype and allele distribution of the SNPs of OPN and its receptors in all patients and healthy controls did not deviate from the Hardy-Weinberg equilibrium. No differences were observed in the distribution of age and sex ratios between VKH disease patients and healthy controls. Table 2 shows the rs4754 allele and genotype frequencies. The frequency of the rs4754 TT genotype was significantly increased in VKH disease patients compared with controls (P = 0.004; pc = 0.048; odds ratio, 1.830; 95% CI; 1.200–2.789). An increased frequency of the rs4754 T allele was also observed in VKH disease patients compared with healthy controls (P = 0.023), but this significance was lost after the Bonferroni correction (pc = 0.092). No significant difference was found regarding the distribution of genotype and allele of the other seven selected SNPs between VKH disease patients and healthy controls (Supplementary Table S1). Because the VKH disease patients exhibited a variety of extraocular manifestations, such as tinnitus, vitiligo, poliosis, and central nervous system signs, a stratification analysis was also performed to investigate the association of the tested SNPs with these clinical parameters. No association was observed, however, between the selected SNPs with any of the clinical manifestations of VKH disease. Similar results were observed after stratification analysis based on the ages of the patients (data not shown). 

Table 2.

 

Genotype and Allele Distributions of OPN rs4754 in VKH Disease Patients and Healthy Controls

Genotype Allele	VKH (%) n = 601	Control (%) n = 605	χ2	P	pc	OR (95% CI)
CC	304 (50.58)	329 (54.38)	1.744	0.187	2.244	0.859 (0.685–1.077)
CT	233 (38.77)	239 (39.50)	0.068	0.794	9.528	0.970 (0.769–1.222)
TT	64 (10.65)	37 (6.12)	8.074	0.004	0.048	1.830 (1.200–2.789)
C	841 (69.97)	897 (74.13)	5.196	0.023	0.092	0.813 (0.680–0.972)
T	361 (30.03)	313 (25.87)	5.196	0.023	0.092	1.230 (1.029–1.470)

Discussion

In this study, we showed that the serum OPN level was significantly increased in the blood of patients with active VKH disease. Recombinant OPN enhanced cell proliferation and upregulated IFN-γ and IL-17 production in patients with active VKH disease. The frequency of the TT genotype of OPN rs4754 was positively associated with VKH disease. These results suggest that OPN may be associated with an increased risk for VKH disease. 

OPN is a matricellular protein that mediates diverse biological functions. Recently, OPN was reported to promote cell-mediated immune responses and to play a role in chronic inflammatory diseases. ^20,31 Our findings showing elevated serum OPN levels in VKH are in line with those observed by us in Behçet's disease. ³⁰ Interestingly, these results are also consistent with those reported previously in chronic autoimmune diseases, such as CD, ^11,20 MS, ¹² RA, ¹³ and SLE. ¹⁴ These results seem to show that the upregulation of OPN may be a common event for these diseases. Unexpectedly, we were unable to find an association between elevated OPN levels and disease activity in our VKH patient group. This might have been due to the small sample size or to the classification of disease activity in the eye. 

Given that T-cell proliferation is one of the important factors in immune response and inflammation, we further investigated the effect of OPN on cell proliferation. Our results showed that OPN promoted a small but significantly higher proliferation of PBMCs from patients with active VKH disease compared with patients with inactive VKH or healthy controls. These findings suggest that the sensitivity of PBMCs to OPN in VKH is higher. This higher response was identified on stimulation with anti-CD3/CD28 antibodies. Costimulation by a combination of anti-CD3 and anti-CD28 antibodies is a generally accepted immunologic method, and, because it is not physiological, testing costimulation with an immunopathogenic ocular antigen could provide additional information to support our hypothesis. Antigens that could possibly be used are tyrosinase and retinal S-antigen. Showing an enhancement of lymphocyte proliferation, cytokine production, or both by OPN is necessary for a definitive conclusion regarding the role of OPN in the pathogenesis of VKH. Another limitation of our study was that we did not show whether the identified response was unique to VKH disease or whether it could also be observed using cells from patients with other forms of uveitis, such as Behçet's disease. 

The OPN concentrations needed to generate a T-cell response for our in vitro studies were in the microgram range, which is in accordance with the doses used by other authors. ^8,32 Although OPN levels observed in serum are often in the nanogram range, earlier studies have shown that the concentration at other sites in the body may be in the microgram range. Recent studies have, for instance, shown that the synovial fluid of rheumatoid arthritis patients contains a mean OPN level of 15 μg/mL. ¹⁹ The mean OPN level in the vitreous fluid of patients with diabetic retinopathy was 2.1 μg/mL. ³³  

It has been reported that OPN is expressed by a wide variety of cell types, such as T and B lymphocytes, dendritic cells, macrophages, neutrophils, NK cells, osteoclasts, endothelial cells, epithelial cells, hepatocytes, and vascular smooth muscle cells. ^7,8,34 It is not known whether OPN production by these types of cells is altered in active VKH disease. 

Given that IFN-γ plays an important role in the pathogenesis of chronic inflammatory diseases including VKH disease, ²¹ we further examined the effect of OPN on the secretion of IFN-γ. Consistent with our previous study, ³⁵ a significantly increased production of IFN-γ was observed in patients with active VKH. Sato et al. ³⁶ and Agnholt et al. ¹¹ reported that OPN stimulation increased IL-12 and IFN-γ production in patients with CD but not in healthy controls. Heilmann et al. ²⁰ postulated a dual function for OPN in inflammation. During acute inflammation OPN is thought to reduce tissue damage, whereas during chronic inflammation it promotes the Th1 response and strengthens inflammation. Our study also showed that OPN was able to increase the production of IFN-γ by PBMCs and CD4⁺ T cells from patients with active VKH, but not in patients with inactive VKH or healthy controls. Differences in response to OPN between the groups tested might have been due to different receptor levels or downstream signaling in cells of patients with inactive VKH. It is also possible that the observed effects can be attributed to differences in the populations of cells in the peripheral blood of patients with inactive VKH versus active VKH, such as the amount of Tregs. ³⁷ Additional studies are needed to clarify this point. 

An upregulated Th17 response has been found to play an important role in chronic inflammatory diseases, including VKH disease. ³⁵ Our study further examined whether OPN could influence the expression of IL-17, a typical cytokine of Th17 cells, in VKH patients. The result showed significantly increased IL-17 production by PBMCs and CD4⁺ T cells in patients with active VKH. OPN significantly induced the production of IL-17 by PBMCs and CD4⁺ T cells from patients with active VKH. These results confirm other studies showing a stimulatory effect of OPN on IL-17 production. ^8,19  

VKH disease is a chronic granulomatous inflammatory disorder whose exact etiology remains unknown, whereby a genetic predisposition has been shown to play an important role. ⁴ A variety of studies have shown that genes in the human leukocyte antigen (HLA) region are the most powerful risk factor for VKH disease in China, Japan, and other countries. ^{4,38 –41} Recently, several non-HLA genes, primarily those relevant to inflammation, have also been shown to be associated with VKH disease. ^{42 –44} OPN, as a proinflammatory cytokine, plays a pivotal role during the inflammatory response, ³¹ and a number studies have reported that the polymorphisms of OPN contributed to susceptibility to chronic inflammatory diseases. ^{23,25,45 –47} We extended these studies to VKH disease and selected SNPs of OPN based on earlier reports. Rs4754 (8090), rs9138(+1239A/C), rs1126616 (9250), and rs1126772 (9583) polymorphisms were associated with some chronic inflammatory diseases ^{23,25,45 –47} and were, therefore, chosen as candidate polymorphisms in our study. Our results showed that the frequency of the OPN rs4754 TT genotype was associated with risk for VKH. The association we observed was not consistent with that reported in RA in Chinese patients or MS in Japanese patients. ^23,48 The discrepancy may be explained by the different genetic backgrounds in the populations analyzed or by the fact that the etiology and pathogenesis of VKH disease are different from those of the other chronic inflammatory diseases investigated thus far. ¹ We did not find an association between VKH and OPN rs9138, rs1126616, or rs1126772. In addition, further analysis of OPN SNPs with VKH based on extraocular features and age at onset did not reveal detectable associations. 

Because OPN exerts its role through interaction with its receptors—integrins and CD44 ⁸ —the polymorphisms of these receptors were also investigated. To date, several SNPs, such as integrin α4 rs1449263, integrin αv rs3738919, and CD44 rs8193, have been associated with chronic inflammatory diseases ^22,24,26 and were, therefore, chosen as candidates in our study. We also included the integrin β3 Leu33Pro polymorphism (rs5918) because it is associated with an enhanced haptotactic migratory response to OPN. ⁴⁹ None of the tested SNPs for the OPN receptors showed a significant association with VKH, and no association was found when patients were subdivided according to the clinical features or age. 

In conclusion, this study indicated that an increased expression of OPN in serum was associated with the clinical severity of VKH disease. OPN could significantly enhance cell proliferation and upregulated the expression of IFN-γ and IL-17 on stimulation with anti-CD3/CD28 antibodies in VKH patients with active disease. These results suggest that an increased expression of OPN may be one of mechanisms involved in the development of VKH disease. Other important findings in our study are that OPN rs4754 is associated with VKH disease and that the TT genotype may confer a risk for VKH disease. However, it is unclear whether and how this SNP exerts its role in VKH disease. Further studies are needed to clarify this. 

Supplementary Materials

Table st1, DOC - Table st1, DOC 

Footnotes

 Supported by Natural Science Foundation Major International (Regional) Joint Research Project (30910103912), National Natural Science Foundation Project (30973242), Program for the Training of a Hundred Outstanding S&T Leaders of Chongqing Municipality, Key Project of Health Bureau of Chongqing, Project of Medical Science and Technology of Chongqing, Key Project of Natural Science Foundation of Chongqing (CSTC, 2009BA5037), Chongqing Key Laboratory of Ophthalmology (CSTC, 2008CA5003), and the Fund for PAR-EU Scholars.

Footnotes

 Disclosure: M. Chu, None; P. Yang, None; R. Hu, None; S. Hou, None; F. Li, None; Y. Chen, None; A. Kijlstra, None

The authors thank all the donors enrolled in the study.