Plasma endothelin-1 and endothelin-A receptor concentrations in patients with primary open-angle glaucoma

ABSTRACT Endothelin-1 (ET-1) is a potent vasoconstrictor and is considered to have a key role in the regulation of ocular perfusion and glaucoma pathogenesis. High ET-1 and ETA-receptor levels are reported in patients with primary open-angle glaucoma (POAG). We compared the mean plasma ET-1 and ETA-receptor concentration of controls and patients with early and advanced POAG stage, and assessed the correlation with the retinal nerve fibre layer (RNFL) thickness. The study included a total of 75 participants, aged 45–83 years: 25 (controls), 22 (early glaucoma) and 28 (advanced glaucoma). The plasma concentration of ET-1 and ETA-receptor was determined by enzyme-linked immunosorbent assay. The RNFL thickness was evaluated with spectral-domain optical coherence tomography. The mean ET-1 concentration was lower in the control group (4.88 ± 1.75 pg/mL) than in the early and advanced POAG group (6.33 ± 2.38 and 6.34 ± 1.56 pg/mL). Statistically significant difference was found between the mean ET-1 concentrations in controls and glaucoma patients (p = 0.029 – early glaucoma, p = 0.018 – advanced glaucoma), and no significant difference was observed between the two POAG groups (p = 0.998). The mean ETA-receptor concentration was highest in the control group (1209.28 ± 314.48 pg/mL) and the differences between the three groups were significant. Significant relationship was found only between ET-1 and RNFL. This study demonstrated the role of ET-1 in glaucoma pathogenesis based on the observed significant high ET-1 and ETA-receptor plasma levels in POAG patients. A new therapeutical approach needs to be considered in some patients.


Introduction
Endothelin-1 (ET-1) is one of the three isoforms of endothelin. This hormone is among the most powerful vasoconstrictors [1]. It was discovered by Yanagisawa et al. in 1988 [2] and for the first time was isolated, described and cloned using pork aortic endothelial cells.
It is a peptide consisting of 21 amino acids and originates from the vascular endothelial cells with the help of endothelin-converting enzyme (ECE). EG-1 affects the target cells by connecting with two receptors, type A (EG A ) and type B (EG % ) [3]. Vascular endothelial cells are the main source of endothelin, but a variety of other cell types also participate in its production, like the endothelium of renal tubules, glomerular mesangium, cardiac myocytes, glia, pituitary gland, macrophages, mast cells, etc. [4]. There are two types of EEE: EEE-1 and EEE-2. EEE-1 is a membrane-linked metalloproteinase, which releases both intra-and extra-proendothelin-1 in basic pH medium and structurally refers to the neutral endopeptidases. EEE-2 has acidic pH optimum and acts as an intracellular enzyme [5].
EG A receptor is expressed on the smooth muscle cells of the vessels and causes their contraction through an influx of Ca 2+ followed by vasoconstriction. Two subtypes of EG % receptor are distinguished. EG B1 is expressed on the vascular endothelial cells and mediates direct vasodilatation through NO release and its transit to the smooth muscle cells with subsequent relaxation. The other subtype, EG %2 , is situated on the smooth muscle cells and mediates direct vasoconstriction [3,6].
Lungs and kidneys are mainly responsible for the EG-1 plasma clearing, so in cases of lung and kidney diseases, the plasma levels of ET-1 increase. Besides the local paracrine function in the vascular tonus regulation, ET-1 has an impact on the development of atherosclerosis and cardiovascular diseases [5,7]. The plasma ET-1 concentration suffers dynamics also in many adrenal diseases, like primary aldosteronism, pheochromacytoma, hypocorticism and hypercorticism [8].
EG-1 is normally found in the eye, where it is produced by the non-pigmented ciliary epithelium and is released into the intraocular liquid [10]. Its receptors are expressed upon various structures, like iris, ciliary muscle, trabecular meshwork (TM), vasculature, retinal and optic nerve astrocytes, which determines the physiological role of ET-1 in the eye [11][12][13].
After establishing increased levels of ET-1 in the anterior chamber fluid of glaucoma patients, ET-1 has also been considered to play a role in the pathogenesis of glaucoma [14]. ET-1 is associated with retinal ganglion cells death. It has been proved that the intravitreal application of ET-1 in rats has a direct effect upon the anteroand retrograde axonal transport in the ganglion cells and causes dose-and time-dependent death [15][16][17]. These effects can partly be explained by the caused vasoconstriction of the optic disc and retinal microvessels. Some studies, using different animal models, could successfully provoke an optic disc impairment and enlargement of the excavation through ET-1 induced ischemia [18,19]. Another damaging mechanism of ET-1 on the optic disc is proposed by Prasanna et al. [12,20]. It states that the increased proliferation of astrocytes under the influence of ET-1 causes hypoperfusion, which makes the optic disc more vulnerable to the intraocular pressure (IOP).
General endothelial dysfunction with increased plasma ET-1 concentration has also been considered to have a negative impact on the eye perfusion and to be a risk factor for normal tension glaucoma (NTG). Emre et al. [21] report a group of patients with normal or compensated IOP whose progression in the visual field defect could be associated with increased plasma concentration of ET-1.
The role of ET-1 in the physiology and pathophysiology of the IOP control is still contradictive. Indirect evidence in experimental cell cultures shows that TM has contractility similar to that of smooth muscles. EG-1 is capable of causing contraction of TM and, thus, a decrease in the intertrabecular space with an increase in the resistance of the intraocular fluid outflow. EG-1 also causes contraction of the ciliary muscle and tension on the TM [22]. Since the elucidation that TM and ciliary body express both receptor subtypes for ET-1 and that EG-1 is present in the intraocular fluid, it has been suggested that ET-1 has influence on the IOP by affecting the balance between the functionally antagonistic contractile forces of these tissues. A mechanism of control on IOP has also been considered, namely inhibition of the Na + /K + -ATPase in the non-pigmented ciliary epithelial cells in cultures. In vivo this could mean decreased production of intraocular fluid and, hence, a decrease of IOP [22].
ET-1-induced glaucomatous optic neuropathy has been reported in the eyes of a variety of model animals: rabbits, primates, mice and rats [23]. Intraocular ET-1 injections induce chronic ischemia of the optic nerve that causes retinal ganglion cell-specific death and glaucomatous cupping of the optic nerve head (ONH). A recent meta-analysis [24] shows that statistically significantly high ET-1 plasma levels in glaucoma patients are associated with significantly higher risk for NTG and POAG.
The aim of this study was to compare the mean plasma concentration of ET-1 and its ET A -receptor between a control group and patients with early and advanced stage of primary open-angle glaucoma (POAG) with raised IOP and to evaluate the correlation and the degree of relationship with circumpapillary and macular retinal nerve fibre layer (cpRNFL and mRNFL) thickness changes.

Materials and methods
In this prospective study, which was approved by the Ethics Committee at the Medical University of Sofia, 75 patients were investigated for ET-1 and ET A -receptor plasma levels. We obtained informed consent forms from all participants included in the clinical research. The patients' age ranged from 45 to 82 years. The gender distribution was 21 men and 54 women. They were divided into three groups: In the control group, healthy volunteers without eye or general diseases were selected. Patients with early and advanced POAG were selected according to the following inclusion criteria: best corrected visual acuity 0.2; refraction error in the following range §4.00 dsph and §1.00 dcyl; IOP above 21 mmHg IOP measured with a Goldmann tonometer; anterior chamber angle III-IV grade of the Shaffer classification; glaucoma damages in the eye fundus; visual field defects typical for glaucoma and corresponding to that in the ONH. Patients with POAG were divided into early and advanced stage glaucoma group according to the severity of the visual field defect in the Hodapp-Parrish-Anderson classification recommended by EGS (European Glaucoma Society). All pathological conditions outside the inclusion criteria were excluded, especially general diseases evidenced by elevated plasma levels of ET-1. Arterial hypertension was not defined as an exclusion factor, because of the wide age range of the participants in this study.
All patients underwent a standard ophthalmic examination, which included: detailed case and family history of ocular and general diseases; refraction and best corrected visual acuity measurement; biomicroscopy; contact central corneal thickness measurement; Goldmann tonometry; indirect gonioscopy using Goldmann lens; indirect fundus biomicroscopy using lens -90 dpt; Standard automated perimetry (SAP) with Humphrey Field Analyzer-HFA II (Carl Zeiss Meditec, Dublin, CA, USA), algorithm SITA Standard, pattern 24-2. Only reliable perimeter results were included. We used the Hodapp-Parrish-Anderson perimetry classification to stage the damages.
Spectral domain-optical coherence tomography (SD-OCT) was performed in all patients, using Topcon 3D OCT 2000 (FA+) (Topcon Corporation, Japan). Only scans with quality higher than 50%, no artefacts or pathology except glaucoma were considered. We used the following programs and protocols.
Circle program was used for evaluation of cpRNFL thickness. We analyzed the following parameters from the Circle protocol: (i) Total cpRNFL, which shows the average thickness in 360 ; (ii) Sup cpRNFL, the thickness in the superior 90 ; (iii) Inf cpRNFL, the thickness in the inferior 90 ; (iv) Nas cpRNFL, the thickness in the nasal 90 , and (v) Temp cpRNFL, the thickness in the temporal 90 .
3D Macula (V) program was used for evaluation of the inner macula layers thickness. We analyzed the following parameters from this protocol: (i) Sup mRNFL, which shows the thickness in the superior half; (ii) Inf mRNFL, the thickness in the inferior half; (iii) Total mRNFL, the average thickness in the whole investigated area.
A blood sample was taken from each patient early in the morning from Vena mediana cubiti through a closed system (holder and vacutainer for serum with yellow tops containing a special gel that separates blood cells from serum) in a fasting state, supine position and relaxed condition. After that, the blood was centrifuged, allowing the serum to be removed for testing. The plasma was separated into 1.5-mL Eppendorf tubes and stored frozen at ¡10.00 C.
Data were analyzed by Microsoft Excel (MS Office 2013) and SPSS for Windows (USA, Chicago, SPSS Inc. Version 17.0). Dispersion analysis (ANOVA test) with inter-group variance and correlation analysis were applied. Values of p < 0.05 were considered to indicate significant differences; statistically significant relationships were also considered as weak in correlation coefficient (R) values of 0-0.3, moderate in 0.31-0.5, significant in 0.51-0.7, strong in 0.71-0.9 and very strong in R > 0.91.

Results and discussion
The total number of participants included in the present study was 75, aged 45-83 years, with a mean age of 63.4 § 8.6 years. The descriptive statistics are shown in Table 1.
The mean ET-1 plasma levels ( Table 2) were higher in patients with POAG (6.33 and 6.34 pg/mL) than in healthy controls (4.88 pg/mL). What is more, the mean levels of ET-1 were almost the same in patients with glaucoma: early stage (6.33 pg/mL) and advanced stage (6.34 pg/mL). Maximum ET-1 plasma concentration was found in the advanced glaucoma group (8.44 pg/mL), and minimum concentration in the early glaucoma group (0.43 pg/mL). The mean ET A -receptor plasma concentration was found to be highest in the control group (1209.28 pg/mL) and lower in the glaucoma patients ( Table 2). The lowest mean ET A -receptor levels were found in the early stage of POAG (673.44 pg/mL), and in patients with advanced POAG they were higher (992.28 pg/mL). Maximum EG A -receptor concentration was found in controls (2276.34 pg/mL), and minimum in early stage glaucoma (261.24 pg/mL). The application of inter-group comparative analysis demonstrated a statistically significant difference in the mean ET-1 plasma concentrations between controls and patients with POAG (p = 0.029, early POAG; p = 0.018, advanced stage) ( Table 3). The statistical analysis showed no significant difference in the mean ET-1 concentrations between groups with early and advanced glaucoma stage (p = 0.998). Statistically significant differences in the mean ETA-receptor plasma levels were observed between controls and patients with glaucoma (p < 0.001, early stage; p = 0.021, advanced stage), as well as between the two glaucoma groups (p = 0.001).
In 2006, Kunimatsu et al. [25] reported their results from examination of the ET-1 plasma concentration using an immunoenzyme method in three groups of patients: control group (19 subjects), POAG (18 patients) and NTG (30 patients) aged under 60 years. The mean EG-1 plasma concentration value in our cohort was higher than that in the Japanese patients both in the control and glaucoma groups. Kunimatsu et al. [25] found a higher ET-1 plasma level in patients with POAG as compared to the control group (1.33 § 0.50 pg/mL), but with no statistical significance, unlike our results (Table 4). In 1997, Tezel et al. [14] and, in 2003, Nicolela et al. [9] also did not find significant differences in the EG-1 plasma concentration between healthy and POAG patients as well (Table 4).
In 2012, Cellini et al. [26] examined the EG-1 plasma concentration in 20 controls compared to 20 POAG patients (Table 4). Their results, similar to ours, showed not only higher ET-1 plasma concentration, but also a statistically significant difference between the concentration in healthy and glaucoma patients: 1.75 § 0.25 pg/mL vs. 2.83 § 0.28 pg/mL (p < 0.001).
All but one study of the ET-1 plasma concentration found no statistical significance in the difference between healthy and POAG patients. It is possible for the results to have been compromised by imprecise inclusion and exclusion criteria, for numerous endocrine and cardiovascular diseases could influence the ET-1 plasma concentration. The ET-1 plasma levels could also be influenced by the general condition of the patient, whether active or calm, lying down in bed or sitting in a chair.
In 2016, Li et al. [24] performed a meta-analysis in order to combine and summarize the results from several studies analyzing ET-1 plasma levels in NTG patients (7 studies, 212 NTG, 164 controls) and POAG (5 studies, 160 POAG, 174 controls), and also to clarify the association between ET-1 plasma levels and the risk for the mentioned types of glaucoma. They found statistically significant higher ET-1 plasma concentrations in glaucoma patients than in the control group, with a 0.63 pg/ mL mean difference (p = 0.02) in the POAG group, and 0.60 pg/mL (p = 0.007) in the NTG patients. This has been the first meta-analysis so far to investigate the association between ET-1 plasma concentrations and glaucoma. The results showed that statistically significantly high ET-1 plasma levels in glaucoma patients Table 2. Mean, maximum and minimum plasma concentration of ET-1 and ETA-receptor.   were associated with significantly higher risk for NTG and POAG [24].
In 2016, Kosior-Jarecka et al. [27] reported positive correlation between ET-1 plasma levels and visual field defects. Our results showed that the ET-1 plasma levels were significantly higher in patients with POAG than in the control group, although no significant difference was found between early and advanced changes in the visual field of patients with POAG. This suggested that ET-1 most probably had influence on the glaucoma pathogenesis, but not likely on the stage of POAG changes.
The analysis of the relationship between ET-1/ET Areceptor plasma levels and cpRNFL/mRNFL thickness in the present study (Table 5) showed a statistically significant negative relationship between ET-1 and mRNFL (Inf mRNFL and Total mRNFL), as well as significant correlation between EG-1 and Inf pRNFL. Close to significant correlation was observed between ET-1 levels and Total cpRNFL. No relationship between the plasma ET A -receptor concentration and any of the RNFL parameters was found. Chen et al. [28] observed higher ET-1 concentration in POAG (31 patients) and NTG groups (18 patients) as compared to a control group (37 subjects) without statistically significant difference (Table 4). Chen et al. [28] further reported a significant correlation of the EG-1 plasma concentration with structural (cpRNFL) and functional (MD) changes. Examining the correlation between EG-1 and cpRNFL, we revealed a significant negative correlation only between EG-1 and Inf pRNFL.
In 2015, Wr obel-Dudzi nska et al. [29] reported a statistically significant difference between the frequency of occurrence of specific ET-1 (K198N) and ET-A receptor (C1222T, C70G and G231A) genes polymorphisms in POAG and NTG groups, indicating that the polymorphic variants have an effect on IOP levels and systemic arterial blood pressure and its regulation mechanisms in NTG patients. Thus, it is suggested that these genetic variances have an influence on vascular factors, and they, in turn, on glaucoma pathogenesis, namely, dysfunction of autoregulation mechanisms, vascular dysregulation and even endothelial dysfunction responsible for ET-1 release [29].
In order to determine the functional role of endotelin in glaucoma, Howell et al. [30] administered bosentan, to DBA/2J mice (an antagonist of the two types of endothelin receptors). Bosentan significantly reduced the incidence of glaucoma in mice: at the age of 10.5 months, 80% of treated eyes did not have glaucoma as compared with only 39% of untreated eyes [30]. It also increased the ocular blood flow in human glaucoma patients, not affecting the blood pressure [6]. These experiments strongly support a role of the endothelin system in the early pathogenesis of glaucoma in this model. Endothelin receptor antagonists could be considered promising new treatments of glaucoma.

Conclusions
The results from this study showed statistically significant difference between the ET-1 plasma concentration in healthy and glaucoma patients. The mean ET-1 plasma concentration increased in the groups as follows: control group < early glaucoma patients < advanced glaucoma patients. The statistical analysis revealed significant difference between the ET A -receptor plasma concentrations in the three groups. The mean EG A plasma concentration increased in the groups as follows: early glaucoma patients < advanced glaucoma patients < control group. Significant correlation was found between Inf mRNFL/Total mRNFL/Inf pRNFL and ET-1. No statistical significance was revealed between the EG A -receptor concentration and RNFL. Our results supported the suggestion that ET-1 could play a role in glaucoma pathogenesis, because of the observed significantly high ET-1 and ET A -receptor plasma levels in POAG patients. Further research, including larger cohorts of patients, would be needed to explore the potential of endothelin receptor antagonists as a new approach of behaviour and treatment in some glaucoma cases.

Disclosure statement
No potential conflict of interest was reported by the authors.