Introduction

Diabetes mellitus is associated with numerous complications, which can be life-threatening, if not controlled. Of these complications, diabetic retinopathy has gained much interest as a severe progressive ocular pathology and the leading cause of blindness among aged patients [1]. Recent advances have shown that inflammation arising during diabetes has a key role in diabetic retinopathy development and progression [2]. Diabetes biochemical alterations, including increased advanced glycation end-products formation, protein kinase C activity and flux of polyols and hexosamines, induce an increase in oxidative stress and inflammatory responses [3]. The latter response is characterized by overproduction of diverse proinflammatory and pro-angiogenic mediators, including (but not limited to) intercellular adhesion molecule (ICAM)-1, tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6 and vascular endothelial growth factor (VEGF) [4]. The increased production of the inflammatory mediators causes degeneration of retinal capillaries via recruiting leukocytes to endothelium, leading to an increase in vascular permeability and thrombosis probability [5]. The recruited leukocytes eventually produce more proinflammatory and pro-angiogenic factors, leading to further infiltration of leukocytes in the retina, alteration of the blood retinal barrier and neovascularization to overcome the ischemic insult caused by vascular occlusion [6].

IL-12 family members, IL-23, IL-27 and IL-35, have emerged as a critical mediators of T helper (Th)1 and Th17 cell populations responses [7]. In particular, the heterodimeric cytokine IL-27 consists of the Epstein-Barr virus-induced gene 3 and IL-27p28, which comprises a receptor composed of gp130 and the IL-27Ra that transduces signals via activation of JAK/STAT (Janus kinase/signal transducer and activator of transcription) and MAPKs (mitogen-activated protein kinases) [8]. IL-27 was initially recognized as a proinflammatory cytokine, because of its enhancement of production of interferon (IFN)-γ and proliferation in CD4+ T cells [9]. However, several studies have reported the negative regulatory effects of IL-27 on Th1, Th2, and Th17 cell responses and elucidated the various functions mediated by this cytokine, including its capability to stimulate IL-10 production and block T cell production of IL-2, which directly suppresses Th2 and Th17 cell activities [10, 11]. Thus, modulating IL-27 clinically might be beneficial in managing inflammatory conditions associated with aberrant immune response.

On this background, the present study aimed to estimate IL-27 concentration in serum and aqueous humor samples of diabetic patients with or without retinopathy and to determine the degree of correlation between its concentration and other risk factors of diabetes, such as serum glucose, cholesterol, triglycerides levels and glycated hemoglobin (HbA1c).

Subjects and methods

The aqueous humor and blood samples were collected from healthy (control group) and diabetic patients. All diabetic patients were at least 10 years of type 2 diabetes mellitus. The participants were selected from Mansoura Ophthalmic Center through 2015-2016. They were categorized into 4 groups: 20 control non-diabetics, 20 diabetic patients without retinopathy, 20 diabetic patients with non-proliferative diabetic retinopathy and 20 diabetic patients with proliferative diabetic retinopathy. Non-proliferative diabetic retinopathy cases were having microaneurysm, retinal hemorrhage, edema and exudates, and in case of appearance of neovascularization either at the optic disk or elsewhere by fluorescein angiography, this was assigned as proliferative diabetic retinopathy. An informed consent was obtained from each patient subsequent to approval of the study by the ethical committee (Faculty of Medicine, Mansoura University). All employed procedures were concordant with the ethical regulations of the committee related with human experimentation (institutional and national) and Helsinki Declaration in 1975 and revise in 2008.

Exclusion criteria

Patients with uncontrolled hypertension or other systemic disease rather than diabetes mellitus, patients with any eye disease rather than cataract and/or diabetic retinopathy and patients with previous treatment for diabetic retinopathy, e.g., laser or intraocular injection was excluded from the study. Cases with dense cataract that prevent fundus examination were also excluded. All patients underwent general examination and laboratory tests to exclude systemic diseases. Complete ocular examination was performed including visual acuity testing slit lamp bimicroscopic examination of anterior segment, intraocular pressure measurement and posterior segment assessment using contact and non-contact lens. Fluorescein angiography was performed for all diabetic patients.

Sample collection

The aqueous humor samples (around 0.1 mL) were drained from the anterior chamber using insulin syringe before initiation of incisions of cataract surgery in groups 1, 2, 3 or prior anti-VEGF injection in group 4. Blood samples were then collected from each participant in the study after 12–14 h of overnight fasting. Blood samples were divided into two portions. The first portion (1 mL) of whole blood portion was collected in tubes containing EDTA and was used for detection of HbA1c. The second portion was centrifuged at 3000 g for 10 min for separation of serum. The resultant serum was further divided into two other portions. One of them was used for immediate determination of triglycerides, cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL) and glucose, whereas the other was kept at −70 °C until IL-27 assay.

Biochemical analysis

The concentration of triglycerides and total cholesterol in serum were estimated as previously described [12, 13] by a kit purchased from Spinreact (Spain; with a detection limit of 0.6 mg/dL). Also, HDL concentration in serum was estimated as previously mentioned [14] by a kit purchased from Spinreact (Spain; with a detection limit of 1.57 mg/dL). LDL concentration in serum was determined from the equation of Friedewald et al. [15]. Glucose concentration in serum was determined by the method of Trinder [16] using a kit purchased from Spinreact (Spain). HbA1C levels were estimated in the whole EDTA blood samples by the ion exchange resin method of Trivelli et al. [17] using kits purchased from Stanbio (USA). Serum and aqueous humor concentrations of IL-27 were measured by enzyme-linked immunosorbent assay (ELISA) kit purchased from Glory Science Co. Ltd (TX, USA) with a detection limit of 1.6 ng/L.

Statistical analysis

Continuous variables are expressed as mean ± standard deviation (SD). Data were first examined for normality and equality of distribution before performing any analysis. One-way ANOVA test was used for comparing the continuous data among the groups. The correlations between the data with the serum and aqueous IL-27 levels were performed using the Pearson correlation co-efficient test. Statistical analysis was performed by SPSS 20.0 statistical software package (Chicago, USA). The level of significance was set at P < 0.05.

Results

The contributing subjects were categorized into four groups as follows: 20 diabetic patients with proliferative diabetic retinopathy, 20 diabetic patients with non-proliferative diabetic retinopathy, 20 diabetic patients without retinopathy and 20 sex- and aged-matched control non-diabetic subjects (Table 1).

Table 1 Demographic and laboratory characteristics of the studied groups
Full size table

One-way ANOVA indicated that there were significant differences between the investigated groups in serum glucose (F = 11.483, P < 0.001) and HbA1c (F = 7.436, P < 0.001) concentrations. One-way ANOVA showed also that there were significant differences between the groups in serum cholesterol (F = 7.720, P < 0.001), triglycerides (F = 17.079, P < 0.001), HDL (F = 6.217, P < 0.001) and LDL (F = 6.094, P < 0.001) concentrations (Table 1).

The mean of aqueous humor IL-27 concentration was elevated in proliferative diabetic retinopathy (6.1 ± 2 ng/L) and non-proliferative diabetic retinopathy (6.7 ± 2.7 ng/L) patients, but not diabetic patients without retinopathy (4.6 ± 0.5 ng/L), as compared to the control group (4.1 ± 0.8 ng/L). This elevation was found to be statistically significant by one-way ANOVA (F = 9.669, P < 0.001) (Table 1).

The mean of serum IL-27 concentration was also elevated in proliferative diabetic retinopathy (4.6 ± 3.9 ng/L) and non-proliferative diabetic retinopathy (4.6 ± 4.4 ng/L) patients, but not diabetic patients without retinopathy (2.4 ± 0.8 ng/L), as compared to the control group (2.9 ± 1.2 ng/L). This elevation was also found to be statistically significant by one-way ANOVA (F = 3.020, P = 0.035) (Table 1).

Correlation analysis among proliferative diabetic retinopathy patients demonstrated significant correlations between aqueous humor IL-27 concentration and serum glucose (r = 0.499, P = 0.025), cholesterol (r = 0.453, P = 0.045), triglycerides (r = 0.461, P = 0.041) and HbA1c (r = 0.573, P = 0.008) concentrations (Table 2).

Table 2 The correlation between IL-27 concentrations in serum and aqueous humor and patients’ characteristics in the proliferative retinopathy group
Full size table

Correlation analysis among non-proliferative diabetic retinopathy patients showed also significant correlations between aqueous humor IL-27 concentration and serum glucose (r = 0.511, P = 0.021), cholesterol (r = 0.568, P = 0.009), triglycerides (r = 0.495, P = 0.026) and HbA1c (r = 0.501, P = 0.025) concentrations (Table 3).

Table 3 The correlation between IL-27 concentrations in serum and aqueous humor and patients’ characteristics in the non-proliferative retinopathy group
Full size table

Correlation analysis among diabetic patients without retinopathy demonstrated significant correlations between aqueous humor IL-27 concentration and serum HbA1C (r = 0.489, P = 0.028) and IL-27 (r = 0.457, P = 0.043) concentrations (Table 4).

Table 4 The correlation between IL-27 concentrations in serum and aqueous humor and diabetic patients’ characteristics without retinopathy
Full size table

Discussion

Diabetic retinopathy is a vision-threatening complication of systemic diabetes mellitus. In the present study, we assessed the levels of circulating and aqueous humor IL-27 and investigated their correlation with some conventional risk factors for retinopathy. To the best of our knowledge, studies that assess IL-27 concentration, especially in the aqueous humor of diabetic retinopathy patients, are still scarce.

The results revealed an elevation of IL-27 concentration in aqueous humor of diabetic retinopathy patients (either proliferative or non-proliferative) in comparison with diabetic patients without retinopathy or healthy individuals. Thus, diabetic retinopathy is correlated with an enhanced local synthesis and secretion of IL-27. Lee et al. [18] suggested that the retinal cells limited the intraocular inflammation via increasing the production of IL-27 and IL-10. The anti-inflammatory effect of IL-27 in retinal cells is mediated through different mechanisms. IL-27 promotes the expansion of a population of IL-10 secreting T cells [19]. Increased production of IL-10 upregulates suppressor of cytokine signaling 3 that negatively regulates proinflammatory cytokine activities, thereby making retinal cells tolerant to stimulation of these cytokines in chronic intraocular inflammatory diseases [20]. IL-27 can also suppress TH17 development and effector T cells expansion via targeting several signal transduction proteins [21,22,23].

Antunica et al. [24] did not find any significant increase in serum concentration of IL-12, which is the most influential stimulator of Th1 differentiation and IFN-γ production in diabetic subjects with retinopathy. However, Zorena et al. [25] found a marked elevation in IL-12 concentration in serum of diabetic retinopathy patients, compared with healthy subjects. The discordance in the data of both studies can be attributed to the variable rates of patients’ microangiopathy. Our results indicated that serum IL-27 concentration was elevated in diabetic retinopathy patients. Despite its retinal anti-inflammatory properties, IL-27 may have proinflammatory potential systemically. IL-27 can stimulate the differentiation of naïve CD4 T cells into Th1 and which then produce IFN-γ by the action of IL-12 [9, 26, 27].

Our results revealed a significant correlation between IL-27 in aqueous humor and some conventional risk factors for retinopathy, such as serum glucose levels, glycemic control as assessed by HbA1C, cholesterol and triglycerides levels. Our data are in harmony with those of Schram et al. [28] who demonstrated that the proinflammatory cytokines IL-6 and TNF-α were in a good correlation with some risk factors for retinopathy. On the other hand, our data were in disagreement with the studies conducted by Winkler et al. [29] and Antunica et al. [24], who found that IL-12 did not correlate with risk factors of diabetic retinopathy. According to our data, there were correlations between serum and aqueous humor IL-27 concentrations and risk factor parameters only in diabetic subjects without retinopathy. Based on this finding, diabetic patients without retinopathy possess an unaltered hemato-ocular barrier devoid of abnormalities in local synthesis of IL-27.

In conclusion, this study showed a marked elevation of both serum and aqueous humor IL-27 concentrations in diabetic retinopathy patients. Also, the aqueous humor concentration of IL-27 correlated with some diabetic retinopathy risk factors like serum glucose and HbA1C concentrations. More studies are still required to fully investigate the biological effects of IL-27 on the characteristic changes associated with retinopathy and phenotyping of the patient groups, especially in relation to aspects of diabetic control and drug-treatment regimes.