Membrane array and multiplex bead analysis of tear cytokines in systemic sclerosis

Although serious ocular manifestations of systemic sclerosis (SSc) have been described, tear analysis of patients with SSc has not been performed in previous studies. Our aim was to measure a wide panel of cytokines and chemokines in tears of patients with SSc and to assess the most significant molecules with a more sensitive and specific method. Unstimulated tear samples were collected from nine patients with SSc and 12 age- and gender-matched healthy controls. The relative levels of 102 different cytokines were determined by a cytokine array, and then absolute levels of four key cytokines were determined by a magnetic bead assay. Array results revealed shifted cytokine profile characterized by predominance of inflammatory mediators. Of the 102 analyzed molecules, nine were significantly increased in tears of patients with SSc. Based on the multiplex bead results, C-reactive protein, interferon-γ-inducible protein-10, and monocyte chemoattractant protein-1 levels were significantly higher in tears of patients with SSc. Our current data depict a group of inflammatory mediators, which play a significant role in ocular pathology of SSc; furthermore, they might function as excellent candidates for future therapeutic targets in SSc patients with ocular manifestations.


Introduction
Considering its pathomechanism, SSc is an autoimmune disease with widespread small-vessel vasculopathy, immune dysregulation with production of autoantibodies, and progressive fibrosis [1].
Although the exact etiology is unknown, some predisposing factors, such as environmental and infectious agents, tissue injury, and hypoxia or oxidative stress on a susceptible genetic background, may play a role in the disease development [1,2]. One of the earliest clinical manifestations in the pathophysiology of SSc is an alteration of the peripheral microvasculature, confirmed by nailfold capillaroscopy results, which indicate that Raynaud's phenomenon and morphological changes may occur months or even years before the onset of fibrosis [3,4]. Furthermore, the severity of capillary injury correlates with the severity of internal organ involvement [5,6]. Although the crucial role of vasculopathy is obvious in the pathophysiology of SSc, the mechanism that drives disease initiation and progression in SSc is still equivocal, despite over 50 years of investigation [7].
There are only few reports available concerning ophthalmological complications in the course of systemic sclerosis. Changes in the organ of vision are thought to be the consequences of systemic complications of scleroderma or adverse effects of immunosuppressive treatment applied. Ocular symptoms may occur at any stage of the disease and may involve numerous ocular tissues. The disease course can be clinically latent or very intensive. The most common clinical manifestations of soft tissue fibrosis and inflammation in patients with SSc include increased tonus and telangiectasia of the eyelid skin. The most commonly reported lesions are periorbital edema, palpebral ectropion, and ciliary madarosis [8]. In our study, the most frequent ocular manifestation of SSc was dry eye syndrome (DES). DES in SSc is believed to be caused by fibrosis-related impairment of lacrimal gland secretion, namely the water portion of the tear film. Furthermore, the lipid layer disorder is caused by chronic blepharitis and meibomian gland dysfunction (MGD), and also the increased evaporation of tears from the ocular surface is the consequence of the restricted eyelid mobility and the consecutive reduced blinking [9]. DES was recently redefined as a multifactorial disease of the tears and ocular surface that results in symptoms of discomfort, visual disturbance, and tear film instability, leading to damage of the ocular surface [10]. Increased osmolality of the tear film [11] and inflammation of the ocular surface [12] are the two major characteristic features of this ocular surface disease. Increased levels of several inflammatory cytokines have been described previously, and accordingly, tear cytokine levels are already considered as potential markers of inflammation in DES [13].
The aim of the present study was to measure a wide panel of cytokines and chemokines in tears of patients with SSc. With a screening method, we planned to determine a group of molecules, which levels are frequently higher in tears of patients compared with healthy controls. Subsequently, we aimed to compare the levels of the mediators selected on the basis of the previous measurements between patients and controls and to assess the most significant ones with a more sensitive and specific method.

Patients and healthy controls
Nine patients with SSc and 12 healthy controls were included in our study. Patients were enrolled from the outpatient clinic at the Department of Rheumatology, University of Debrecen [nine females, mean (SD) age 61.85 (48-74) years]. Diagnosis of SSc was based on the corresponding international criteria [14,15]. Subjects went through ophthalmological examination and basal tear sample collection at the Department of Ophthalmology during a 2-month period. Tear collection was performed between 9 and 11 am. The diagnosis of secondary Sjögren's syndrome was excluded in all patients. We enrolled 12 age-matched female healthy controls, without history of any autoimmune or ocular disorders. Patients did not take immunosuppressive medications at the time of tear sampling. Written informed consent was obtained from all patients and controls. Study protocol was approved by the local bioethics committee and followed the tenets of the Declaration of Helsinki.

Tear sample collection
Unstimulated tear samples were collected from nine patients with SSc and 12 age-and gender-matched healthy controls. Glass capillary tubes (Haematokrit-Kapillaren, 75 ll, L 75 mm, Hirschmann Laborgerate, Germany) were used to collect tears from the lateral canthus of open eyes, avoiding additional reflex tearing as much as possible. Tear collection was carried out from 9 to 11 am. The samples were obtained from both eyes of each individual and were pooled afterward. Each sample was transferred into a sterile collection tube, carried on dry ice to the laboratory, and stored at -80°C until further assessment.

Determination of tear cytokine levels
To remove cells, cellular debris and contaminant particles samples were centrifuged (10 min, 15,000 rpm, 4°C) prior to use. Tear samples of controls and patients were used for cytokine profiling. The relative levels of 102 different cytokines were determined by Proteome Profiler Human XL Cytokine Array Kit (R&D Systems) using 50 ll samples according to the manufacturer's instructions. The pixel density in each spot of the array was determined by ImageJ software.
Total protein concentrations were determined by Bradford's method in both series of experiments.

Statistical analysis
Cytokine array data were representative of four control and four SSc samples. Integrated density values obtained from densitometry were corrected with background levels, than they were normalized to positive control spots and total protein content of the samples.
In case of Luminex measurements, values represented the mean of nine control and 12 SSc samples. Concentration values calculated by Bio-Plex Manager software were normalized to total protein content of the samples. The data were expressed as mean ± SD. Cytokine levels of patients and controls were tested for statistical significance using two-tailed Mann-Whitney U test, where a p \ 0.05 was considered to be statistically significant.
Integrated density values were normalized to positive control spots and total protein content of the samples. Cytokine array data are representative of four control and four SSc samples (Fig. 1c).
Normalized density of CFD was 50. 35  Mean total protein values did not differ significantly in tears of patients and controls (Fig. 1d). Mean total protein value was 40.9239 lg/ml in tears of patients with SSc and 42.536 lg/ml in tears of healthy controls (p = 0.86).

Multiplex cytokine bead assay results
With the more sensitive and more specific Luminex bead assay, four key molecules were determined in tears of nine patients with SSc and 12 healthy controls.
Based Except for CFD, differences between patient and control values were statistically significant: p = 0.01 for CRP, p = 0.001 for IP-10, and p = 0.1187 for MCP-1. Values represent the mean (±SD) of the nine patient and 12 control samples, which are the fold change in normalized cytokine levels (Fig. 2a). The difference between total protein values of control and SSc tear samples was not significant (p = 0.37263). Mean total protein was 818.46 (779.94-1162.4) lg/ml in tears of patients and 872.46 (771.78-1359.5) lg/ml in tears of controls (Fig. 2b).

Discussion
Many ocular manifestations of SSc have been described including conjunctival telangiectasia, keratoconjunctivitis sicca, and filamentous keratitis [16]. Although in systemic autoimmune diseases ocular manifestations have significant debilitating roles, tear analysis has been lacking from the investigations repertoire. Since tear represents the local homeostasis of the ocular surface better than serum, this makes tear an ideal sample for assessing ocular pathology in the disease. Tear investigation is a challenging research field due to the fact that the sample collection is noninvasive, but has an insurmountable limitation regarding the quantity of sample obtainable [17]. Although diurnal cycle has a determining effect on the amounts of tear cytokines [18], tear collection was performed every day in the same time interval for the elimination of this important influencing factor. In this project, we did not perform a longitudinal assessment on tear cytokines, which could be a limitation factor of the paper; therefore, further studies are needed for better understanding of the diurnal cycle of tear cytokines in SSc. Based on our cytokine array results, nine cytokines and chemokines had significantly higher levels in tears of patients with SSc. This screening method was performed for the assortment of 102 cytokines, selecting the most relevant ones in the pathogenesis of SSc for Fig. 1 Tear cytokine profile of the control group and patients with SSc. Non-stimulated tear samples of healthy controls and patients with SSc were analyzed by cytokine array, detecting 102 different cytokines (a, b), which revealed a shifted cytokine profile characterized by predominance of inflammatory mediators (c). Values represent the mean (±SD) of four control and four SSc samples (two-tailed Mann-Whitney U test, where *p \ 0.05; **p \ 0.01; ***p \ 0.000). The results represent the fold change in tear cytokine levels normalized to total protein contents, which did not differ significantly in tears of patients and controls (d). CHI3L1 chitinase-3-like protein 1, CFD complement factor D, CRP Creactive protein, EGF epidermal growth factor, IP-10 interferon gamma-induced protein-10 (CXCL-10), MCP-1 monocyte chemotactic protein 1 (CCL2), MIG monokine induced by gamma interferon (CXCL-9), MMP-9 matrix metallopeptidase-9, VDBP vitamin D-binding protein further experiments. All of the molecules, which appeared to be significantly higher in tears of patients, are molecular players of the immune responses and inflammatory processes, which confirms the presence of ocular surface inflammation possibly as a consequence of DES in patients with SSc.
Previously, CHI3L1 has not been described in relation to the pathomechanism of SSc. It is a protein, which takes part in the processes of inflammation and tissue remodeling. We have found elevated levels of CHI3L1 in patients with SSc, which result correlates well with the fact that inflammation and tissue injury caused by hypoxia and oxidative stress are always present in the course of SSc.
In fact, different pathways may lead to vascular dysfunction processes in SSc, such as direct vascular damage or pro-inflammatory responses. Studies in different diseases have shown functional links between activated complement molecules and these pathways. A serine protease, CFD, which is also known as adipsin, plays a key role in these processes [19,20]. CFD is the rate-limiting enzyme in the activation cascade of the alternative pathway, and its level in blood is quite low. Our cytokine array results have shown increased CFD levels, which confirms the role of the complement system in the ocular pathology of SSc.
Levels of EGF were also elevated in tear samples of patients with SSc. EGF is a growth factor that stimulates cell growth, proliferation, and differentiation [21]. Elevation of EGF may be explained by the above-mentioned processes of vasculopathy. The next molecule, which appeared to be higher in patients' tears, is matrix metallopeptidase-9 (MMP-9). As a protease of the matrix metalloproteinase (MMP) family, it is involved in the breakdown of extracellular matrix in normal physiological processes, such as embryonic development, reproduction, angiogenesis, bone development, wound healing, and cell migration, as well as in pathological processes, such as intracerebral hemorrhage, arthritis, and metastasis [22][23][24]. In a study of Kim et al. [25], serum MMP-9 concentrations were elevated in patients with SSc and correlated well with skin scores. Their results suggest that the increased MMP-9 concentrations may be due to the overproduction by dermal fibroblasts and also that the enhanced production of MMP-9 may contribute to fibrogenic remodeling during the progression of skin sclerosis in SSc. Our results of tear cytokine array are parallel with the finding that MMP-9 is increased in the course of SSc. In a previous study, expression of antiangiogenic chemokines and their receptors was determined in sera and skin of patients with SSc [26]. Based on their results, MIG and its receptor are elevated in serum and highly expressed in skin of patients with SSc. We have found also increased levels of MIG in tear samples of patients, which strengthens the fact that dysregulated angiogenesis is an important feature in the pathomechanism of SSc. The next protein that appeared to be higher is VDBP, which belongs to the albumin gene family. VDBP is a multifunctional protein found in plasma, ascitic fluid, and cerebrospinal fluid and on the surface of many cell types. It binds to vitamin D and its plasma metabolites and transports them to target tissues [27]. Others have measured significant quantities of VDBP-actin complexes in the plasma following injury [28]. The presence of tissue injury is likely to be the explanation of our results, namely the elevated levels of VDBP in tears of patients with SSc.
Based on our results of multiplex bead assay, the three molecules that showed significant differences in tears of patients and controls were IP-10, MCP-1, and CRP. Previous studies have already demonstrated elevated levels of these markers in sera of patients with SSc.
General markers of inflammation, such as CRP, are expected to be higher in a disease such as SSc. In earlier trials, CRP appeared to be elevated in sera of patients with SSc and was associated with poor survival. Therefore, it may be a useful indicator of disease activity and severity in SSc [29,30].
Recent reports have shown that the serum and/or the tissue expressions of IP-10 are increased in various bacterial, viral, fungal, and protozoal infections [34] and also in autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus, multiple sclerosis, autoimmune thyroid diseases, type 1 diabetes mellitus, Addison's disease, and at last but not least SSc [32,[35][36][37]. CXCL-10 is secreted by cluster of differentiation (CD)4?, CD8?, natural killer, and natural killer T cells, and it is dependent on interferon-c. CXCL-10 can also be secreted by several other cell types including endothelial cells, fibroblasts, keratinocytes, thyrocytes, and preadipocytes. Detecting high level of CXCL-10 in peripheral fluids is therefore a marker of host immune response [30], which correlates well with our results of cytokine bead assay measurements.
Finally, MCP-1, which is a key participant of the fibrotic processes in SSc, also appeared to be higher in patients' tears. MCP-1, which recruits monocytes, memory T cells, and dendritic cells to the sites of inflammation, is produced by either tissue injury or infection [38]. It is known as one of the most pathogenic chemokines during the development of inflammation and fibrosis in SSc [39]. Not only MCP-1 is a chemoattractant molecule for monocytes and T cells, but also it induces Th2 cell polarization and stimulates collagen production by fibroblasts [40]. Hasegawa et al. [41] have previously shown that serum MCP-1 levels are elevated when skin and lung are affected in patients with SSc. It has also been reported that cultured dermal fibroblasts from patients with SSc show augmented expressions of MCP-1 mRNA and protein [42].
Of the last three molecules, IP-10 and MCP-1 are the two, which molecular characteristics make them potential candidates for therapies against the pathological consequences of diseases such as SSc. Monitoring of these factors in tears of patients with SSc can be a noninvasive alternative instead of serum investigation. Furthermore, in patients with ocular manifestations, such as DES, tear analysis is far more informative and serves information of the ocular surface; hereby, it could help us in choosing the appropriate treatment, particularly artificial tears or antiinflammatory eyedrops. Further studies are needed to understand the signaling pathways regulating IP-10 and MCP-1, with the aim of developing new interventions against autoimmune diseases mediated by these chemokines [32,34] as well as inventing novel therapeutic possibilities for the ocular manifestations of SSc.
A methodological limitation of our study is that we could not perform further analyses to confirm array results because of the small tear volume available.
In a previous study of Li et al., tear film cytokines were analyzed by antibody protein array. They found that the presence of factors that exhibit an affinity for plastic, capture antibodies, and IgG and create a complex series of matrix effects profoundly impacts the reliability of dot-ELISA. These effects include elevated levels of background reactivity and reduction in capacity to bind targeted protein, which limit the ability to obtain meaningful results [43].
PCR would be a feasible method to confirm array results.
However, a longer-term prospective study in a larger population with extension of the ophthalmological examinations is needed to confirm their clinical utility.