Analysis of multiple cytokines in aqueous humor of patients with idiopathic macular hole

Background Idiopathic macular holes are common ophthalmic manifestations with unknown pathogenesis. Thus far, there has been minimal research regarding the causes of idiopathic macular holes, especially with respect to the underlying immune mechanism. To provide clarity regarding the treatment and prognosis of idiopathic macular holes, specifically regarding the levels of cytokines in affected patients, this study examined and analyzed multiple cytokine levels in aqueous humor from patients with idiopathic macular holes. Methods This comparative cross-sectional study included 38 patients in two groups: a cataract control group (n = 17) and an idiopathic macular hole group (n = 21). The levels of 48 cytokines in aqueous humor were detected by multiplex analysis with antibody-coupled magnetic beads. The Kolmogorov–Smirnov test was used to check whether the data were normally distributed; Student’s t-test and the Mann–Whitney U test were used to assess differences in cytokine levels between the two groups. Spearman correlation analysis was used to assess relationships among cytokine levels in the experimental group. Signaling pathways containing cytokines with significantly different expression in the experimental group were identified. Results There were significant differences in aqueous humor cytokine levels between patients with idiopathic macular holes and patients in the cataract control group. Notably, hepatocyte growth factor (p = 0.0001), GM-CSF (p = 0.0111), and IFN-γ (p = 0.0120) were significantly upregulated in the experimental group, while TNF-α (p = 0.0032), GRO-α (p < 0.0001), and MIF (p < 0.0001) were significantly downregulated in the experimental group. Furthermore, the GM-CSF level showed significant positive correlations with levels of IL-1 (r = 0.67904, p < 0.001), IL-4 (r = 0.76017, p < 0.001), and IFN-γ (r = 0.59922, p = 0.004097) in the experimental group. Moreover, the levels of nerve growth factor and hepatocyte growth factor showed a significant positive correlation (r = 0.64951, p = 0.001441) in the experimental group. Conclusions Patients with idiopathic macular holes showed significant variation in aqueous humor immune response after the onset of hole formation, including the recruitment of immune cells and regulation of cytokine expression. Our findings also suggest that it is not appropriate to use patients with macular holes as the control group in studies of aqueous humor cytokine levels in ophthalmic diseases.


Background
Macular holes constitute tissue defects that extend from the retinal inner membrane to the photoreceptor layer of the macula, seriously impairing central vision in affected patients. Kuhnt first reported non-traumatic macular holes in 1900 [1]. Since then, various causes of macular holes have been recognized [2]. The overall prevalence of macular holes is 0.33%; affected patients most commonly exhibit agnogenic idiopathic macular holes (approximately 83% of patients with macular holes). Moreover, macular holes most frequently occur in healthy women aged > 50 years (mean age, 65 years; female: male ratio of 2:1); 6-28% of affected patients have bilateral macular holes [3].
Idiopathic macular holes (IMHs) have no obvious systemic or ocular local causes, and do not involve disease in the fundus itself. Current research suggests that IMHs are mainly associated with age-related degenerative changes in the vitreoretinal interface. In recent years, IMHs have become increasingly important in clinical practice because of the large number of older people in many countries, and the morbidity of this condition has gradually increased [4].
The pathogenesis of IMHs has been unclear. In 1995, Gass proposed a theory of posterior vitreous tangential traction [5]. Based on this theory, vitrectomy combined with internal limiting membrane stripping became a common surgical procedure for the treatment of IMHs, although this theory was unable to explain the molecular mechanism underlying IMH formation. Investigation of this mechanism is important for preventing the occurrence of IMHs [6].
In prior studies involving histological analysis of the internal limiting membrane in patients with IMHs, components of Müller cells, glial cells, and fibroblasts were identified [7,8]. The present analysis of cytokines in aqueous humor was performed to investigate whether intraocular cell proliferation is mediated by inflammatory and immune cells, as well as cytokines secreted by these cells, during the occurrence, development, and repair of IMHs.

Participants and ethics approval
The experimental group in this study included 21 eyes of 21 patients (1 man and 20 women) with stage ≥ 3 IMHs from April 2018 to April 2019. The control group comprised 17 eyes of 17 non-diabetic patients who underwent cataract surgery at the same time, including 12 men and 5 women. The inclusion criteria in both groups were no other retinal or optic neuropathy, except diffuse retinopathy. Exclusion criteria included (1) any other eye disease (e.g., glaucoma or uveitis), (2) history of ocular surgery, and (3) history of ocular inflammation.
The baseline characteristics of patients with IMHs (n = 21) and controls (n = 17) are shown in Table 1. Participants were considered to have hypertension if their blood pressure was above 140/90 mmHg or they were taking any antihypertensive medications. Participants were considered to have hypercholesterolemia if their total fasting plasma cholesterol level was > 200 mg/dl. Participants were considered to have hypertriglyceridemia if their fasting plasma triglycerides level was > 200 mg/dl. Participants were considered to be nonsmokers (no smoking for at least 1 year) or current smokers. The study was approved by the Ethics Committee of the Fourth People's Hospital of Shenyang, People's Republic of China, and followed the tenets of the Declaration of Helsinki. All patients provided written informed consent prior to participation in the study.

Diagnosis and surgery
This study was performed using a comparative crosssectional method at the Fourth People's Hospital of Shenyang (People's Republic of China). None of the patients in the IMH group had cataracts; none of the patients in the cataract group had IMHs. All patients in the IMH group underwent complete physical and ophthalmologic examination for clinical diagnosis of IMHs, including assessments of visual acuity and relative afferent pupillary defect, as well as multifocal electroretinography, optical coherence tomography, fundus examination, and fluorescent fundus angiography. During cataract surgery, limbal puncture was performed with a sterile tuberculin syringe. An undiluted aqueous sample (0.2-0.5 ml) was drawn into the syringe, then transferred to a 2-ml centrifuge tube. Immediately after the end of the operation, 0.2-0.5 ml aqueous solution was extracted by anterior chamber puncture. All samples were snapfrozen in liquid nitrogen and stored at − 80°C until analysis.  (Table 2). All experimental measurements were performed in accordance with the kit manufacturer's instructions. Briefly, 50 µl of 1 × beads were added to each well and the beads were washed twice with 200 µl of wash buffer per wash. Fifty-microliter aliquots of standards, samples, and controls (i.e., negative controls that are provided with the test kit) were added to respective wells. Plates were incubated in a shaker at 850 rpm for 30 min at room temperature. Each well was then washed three times with 100 µl of wash buffer per wash. Twenty-five microliters of 1 × detection antibody was added to each well and the plates were incubated in a shaker at 850 rpm for 30 min at room temperature. Each well was then washed three times with 100 µl of wash buffer per wash. Fifty microliters of 1 × streptavidin-PE was added to each well and the plates were incubated in a shaker at 850 rpm for 10 min at room temperature. Each well was then washed three times with 100 µl of wash buffer. Samples were resuspended in 125 µl of assay buffer and the plates were incubated in a shaker at 850 rpm for 30 s at room temperature.
The measurements were performed in accordance with the kit manufacturer's instructions and data were acquired using the Bio-Plex TM 200 system, software version 6.0 (Bio-Rad). The standard curve for each cytokine was generated using the reference set of cytokine concentrations provided with the kit; sample concentrations of each cytokine were calculated using the multiparameter standard curve. If a sample concentration was above or below the detection limit, it was considered and outlier and removed from analysis.

Statistical analysis
Data are shown as mean ± standard deviation, range, median, and IQR. Statistical analysis was performed using R, version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria). The Kolmogorov-Smirnov test was used to determine whether the data were normally distributed. The Pearson χ 2 test was used for comparisons of qualitative variables. Student's ttest and the Mann-Whitney U test were used for comparisons of quantitative variables between two groups. Spearman correlation analysis was used to assess relationships among cytokine levels in the experimental group. p-values < 0.05 were considered statistically significant. Pathway enrichment analysis (http://metascape.org) was used to identify major signaling pathways that contained cytokines with significantly altered expression between groups.

Cytokine levels in aqueous humor
Levels of cytokines in aqueous humor samples were compared between the IMH and control groups ( Table 2). Figure 1 shows a visual comparison of all tested cytokines between the two groups. Twenty-seven cytokines with significant differences in levels are shown in Fig. 2a; of these 27, 20 were significantly upregulated and seven were significantly downregulated.  Table 3 shows the relationships among cytokines that were significantly different between the IMH and control groups. Notably, the level of GM-CSF was significantly positively correlated with the levels of inflammationrelated cytokines including IL-1 (r = 0.67904, p < 0.001), IL-4 (r = 0.76017, p < 0.001), and IFN-γ (r = 0.59922, p = 0.004097) both. In addition, the level of neuronal and tissue repair-related nerve cell growth factor (NGF) was significantly positively correlated with the level of HGF (r = 0.64951, p = 0.001441). Figure 3 shows the results of pathway enrichment analysis (http://metascape.org) involving cytokines that were significantly different between the IMH and control groups. The greatest proportion of cytokines was involved in the cytokine-cytokine receptor pathway, which is implicated in inflammatory responses. This includes both members of pro-inflammatory signaling pathways (e.g., IL-17, IL-14, and IL-13) and members of anti-inflammatory signaling pathways (e.g., IL-10). Moreover, the cytokine levels in hematopoietic lineage signaling pathways associated with tissue repair exhibited significant differences between groups.  1 Comparison of cytokine levels between idiopathic macular hole and cataract groups using a logarithmic scale. Significant differences (p < 0.05) are marked with asterisks. Mean cytokine levels (pg/ml) in the idiopathic macular hole and cataract groups are shown. The p value is obtained by Mann-Whitney U test a b Fig. 2 Differences in cytokine levels between idiopathic macular hole and cataract groups. a Differences in cytokine levels between patients with idiopathic macular holes and patients with cataracts, shown using standardized cytokine data in heat map format. The results indicated that the idiopathic macular hole and cataract groups formed two distinct clusters. b Quantitative differences between the two groups in levels of cytokines that may play an important role in idiopathic macular holes. The value represents the mean ±SD of cytokine concentration in cataract or IMH patients. The p value is obtained by Mann-Whitney U test The lower left part of the figure is a p-value and the upper-right part is an r-value Table 3 Correlation between cytokines with significant differences between the two groups