Graefes Arch Clin Exp Ophthalmol DOI 10.1007/s00417-013-2508-z

RETINAL DISORDERS

Increased intravitreal angiopoietin-2 levels associated with rhegmatogenous retinal detachment Sirpa Loukovaara & Kaisa Lehti & Alexandra Robciuc & Timo Pessi & Juha M. Holopainen & Katri Koli & Ilkka Immonen & Jorma Keski-Oja

Received: 4 March 2013 / Revised: 8 August 2013 / Accepted: 21 October 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract Purpose To explore factors related to pathogenesis of rhegmatogenous retinal detachment (RRD) and development of proliferative vitreoretinopathy (PVR), vitreous levels of angiopoietin-1 and −2 (Ang-1 and −2), previously undefined in RRD, transforming growth factor-(TGF) β1, vascular endothelial growth factor (VEGF), erythropoietin (EPO) and proteolytic mediators of extracellular matrix remodelling (MMP-2 and −9) were compared in eyes with RRD and eyes with idiopathic macular hole or pucker. Methods Vitreous samples were collected from 117 eyes with RRD (study group) and 40 eyes with macular hole or pucker

This study was partially presented at the XXVIIIth Meeting of Club Jules Gonin, in Reykjavik, Iceland, June 23rd 2012. S. Loukovaara (*) : I. Immonen Unit of Vitreoretinal Surgery, Department of Ophthalmology, Helsinki University Central Hospital, Haartmaninkatu 4 C, 00290 Helsinki, Finland e-mail: [email protected] K. Lehti Genome-Scale Biology, Haartman Institute, Biomedicum Helsinki, University of Helsinki, Helsinki, Finland K. Koli : J. Keski-Oja Molecular Cancer Biology, Haartman Institute, Biomedicum Helsinki, University of Helsinki, Helsinki, Finland A. Robciuc : J. M. Holopainen Helsinki Eye Laboratory, Department of Ophthalmology, University of Helsinki, Helsinki, Finland A. Robciuc National Institute of Health and Welfare, Institute for Molecular Medicine, Biomedicum Helsinki, Helsinki, Finland T. Pessi Datawell Oy, Espoo, Finland

(control group). Growth factors were measured by ELISA and matrix metalloproteinases (MMPs) by gelatin zymography. Results The mean vitreous concentrations of Ang-2, MMP-2, and MMP-9 were higher (all p <0.01), whereas concentration of VEGF was lower (p =0.01) in eyes with RRD relative to controls. Logistic regression analysis identified Ang-2 concentration as a novel marker of RRD (p =0.0001, OR 48.7). Ang1, EPO, and total TGF-β1 levels were not significantly different between the groups. However, TGF-β1 and MMP-2 were increased in eyes with total RRD compared to those with local RRD (p ≤0.05). In eyes with PVR, no differences were observed in any studied marker as compared with non-PVR eyes. Conclusions Current results reveal Ang-2 as a key factor upregulated in RRD. It may co-operate with fibrosisassociated factors and contribute to vascular complications such as breakdown of blood–eye barrier and PVR development. Keywords Angiopoietin-1 . Angiopoietin-2 . Endothelial cell . Erythropoietin . Matrix metalloproteinases . Rhegmatogenous retinal detachment . Proliferative vitreoretinopathy . Transforming growth factor-β1 . Vascular endothelial growth factor

Introduction Rhegmatogenous retinal detachment (RRD) is characterized by separation of the neurosensory retina and retinal pigment epithelium (RPE), with fluid accumulation in the intervening space. A common event in RRD is the breakdown of the blood–eye barrier and leakage of the tissue remodelling factors in the vitreous. Proliferative vitreoretinopathy (PVR) is an inflammatory, proliferative response to retinal detachment (RD) that results in the formation of tractional fibrocellular membranes, often resulting in failure in retinal detachment

Graefes Arch Clin Exp Ophthalmol

surgery. Previous studies have described signs of activated synthesis and degradation of the extracellular matrix (ECM) and proliferation, migration and apoptosis of several cell types including glial, vascular, and inflammatory cells in the pathogenesis of PVR [1–5]. Angiopoietins and VEGF are important factors in the physiology/pathophysiology of endothelial cells (ECs), whether or not angiogenesis is present [6–11]. Angiopoietins are key regulators of endothelial cell–cell and cell–extracellular matrix (ECM) interactions both in angiogenesis and inflammation [12]. VEGF has been found in the vitreous and epiretinal membranes in eyes with PVR [13], but the role of angiopoietins in human RRD has not been thoroughly examined. In proliferative vitreoretinal diseases, matrix metalloproteinases (MMPs), and transforming growth factor (TGF)-β1are involved in ECM processing and remodelling [14, 15]. Previous studies have revealed increased concentrations of MMP-2 and -9 either in plasma, vitreous, or fibrovascular tissues in eyes with proliferative diabetic retinopathy (PDR) [16–24]. The role of MMPs has also been characterized in eyes with RRD and PVR [25–27]. TGF-β is a multifunctional cytokine produced mainly by RPE cells and pericytes, and overexpressed in the vitreous of patients with PDR and PVR [28–30]. To date, the role of erythropoietin (EPO) is less conclusive in the pathogenesis of RRD. In addition to activating red blood cell synthesis, EPO provides angiogenic and neuroprotective signals against ischemia-reperfusion injury in many diseases [31–34]. EPO levels in vitreous are increased in ischemic retinal diseases, such as in branch retinal vein occlusion and PDR [35]. The aim of the current study was to identify novel candidate molecules that could be upregulated in eyes with RRD and potentially be related to PVR process. We consider this topic highly relevant because of the paucity of current pharmacologic strategies that would aid treatment of these patients.

Patients and methods The study design was a prospective controlled observational study. Patients were admitted for primary vitrectomy due to RRD or macular hole or pucker in the unit of vitreoretinal surgery, Helsinki University Central Hospital, Helsinki, Finland. Recruitment of study patients (117 eyes with RRD) and controls (40 eyes with macular hole or pucker) was ongoing from September 2006 to December 2008. The eyes were operated during the same or the following day by the recruiting surgeon. We included only eyes with no previous posterior segment surgery. The duration of RRD was not analyzed. None of the study patients or controls had diabetes.

The study was conducted according to the tenets of the Declaration of Helsinki, and it was approved by the Institutional Review Board of Helsinki University Central Hospital. Signed informed consent was obtained from each participant before sampling. Confidentiality of the patient records was maintained when the clinical data were entered into a computer-based standardized data entry for analysis. Surgery Undiluted vitreous samples (500–1,000 μl) were collected at the start of the pars plana vitrectomy (20- or 23-gauge, Accurus, Alcon Instruments, Inc., Fort Worth, TX, USA) without an infusion of artificial fluid. The samples were collected by manual aspiration into a syringe via the vitrectome with the cutting function activated. Samples were transferred into sterile 1.5 ml Eppendorf tubes (Fremont, CA, USA) and immediately frozen at −70 °C until laboratory analysis. Venous sample collection Venous blood samples from the antecubital vein were also collected. The samples were clarified by centrifugation at 3, 000 rpm (1,590 g) at room temperature for 10 min; the plasma was immediately frozen and stored at – 70 °C until assayed. Patients and controls Results of the analysis of the samples from 157 patients and controls are reported here. Of the 117 patients with RRD, 19 had a total retinal detachment, and in 98 patients, the retina was partially detached. PVR of Grade C was observed in 22 eyes. PVR was graded according to the classification of the Retina Society Terminology Committee (1991) [36]. The control group consisted of 40 eyes with a quiescent idiopathic macular hole or pucker. Patients operated due to RRD were younger (mean age 62.0±12.1, median 60 years) than control patients operated due to macular hole or pucker (mean age 69.3±6.6, median 68.5 years) (p <0.001). The gender distribution was also unequal between the groups. Of RRD patients, 69 (58.5 %) were males and 49 (41.5 %) females, compared to 13 (32.5 %) males and 27 (67.5 %) females operated due to macular hole or pucker (p =0.004). Determination of the vitreous and plasma biomarkers All vitreous samples were melted and clarified by centrifugation in September 2009. To ensure equal sample quality for all analyses, the supernatants were divided into ten aliquots in sterile tubes and re-frozen at –70 °C. Fresh aliquots were subjected for the measurement of the biomarkers of interest.

Graefes Arch Clin Exp Ophthalmol

Determination of Ang-1, Ang-2, VEGF, TGF-β1, and EPO levels by enzyme immonoassay For Ang-1 and Ang-2 measurements in vitreous samples we used commercial solid-phase ELISA kits for both proteins and employed the assay protocol developed by the manufacturer (R&D Systems, Minneapolis, MN, USA). The plates were analyzed using the microplate reader Victor 2 Multilabel Counter (Wallac, Turku, Finland) at wavelength 450 nm. Frozen vitreous humor samples were thawed and clarified by centrifugation at 13,000 rpm at room temperature for 5 min and diluted ×1.25 (40 μl of vitreous sample/well) in the sample dilution buffers provided with the kits. A reference serum sample was used to calculate intra- and inter-assay variation. Standard curves were constructed by plotting the concentration and standards mean absorbance at 450 nm. To calculate the concentration, the standards were fitted with a second order polynomial equation. Concentrations have been expressed here as pg/ml. Intra-assay variance was calculated based on values measured for the reference sample present on each ELISA plate in duplicates of two dilutions: ×50 and ×100 for serum, ×15 and ×30 for plasma in the Ang-1 assay, and respectively ×5 and ×10 for both serum and plasma in the Ang-2 assays. All four concentrations measured for the reference sample were used to calculate the intra-assay variance coefficient of 7.26 for Ang-1 and 6.40 for the Ang-2 ELISA method. Inter-assay variance was calculated based on the same reference samples, and values of all ELISA plates were taken into consideration. In our hands, Ang-1 ELISA reported an 8.44 variance coefficient, while for the Ang-2 method the inter-assay variance was of 4.16. VEGF and total TGF-β1 levels in the vitreous were measured using Quantikine ELISA kits (R&D Systems, Minneapolis, MN, USA). Latent forms of TGF-β1 were acidactivated before the assay according to manufacturer’s instructions. The detection level was 15 pg/ml. Erythropoietin levels (pg/ml) in the vitreous were measured with commercial ELISA kits (Absorbance @ 450 (1.0 s), Photometry Wallac, Turku, Finland). For statistical analyses, Ang-1/Ang-2 ratio were calculated separately. The rationale for that was the recent finding that especially Ang-1/Ang-2 ratio could be a critical switch controlling inflammatory processes [37]. Gelatin zymography and quantification of MMP-2 and −9 activities Gelatin zymography was performed to evaluate the relative levels and activation ratios of the MMP-2 zymogen (proMMP-2) and -9 (proMMP-9) in the vitreous samples. To analyze the gelatinolytic proteins, aliquots of vitreous samples were subjected to gelatin zymography essentially as described

[38]. The polypeptides of the samples were dissolved in nonreducing Laemmli sample buffer and separated by SDSPAGE using 10 % polyacrylamide gels containing 1 mg/ml of gelatin. After electrophoresis, the gels were washed twice with 50 mM Tris–HCl, pH 7.6, containing 5 mM CaCl2, 1 μM ZnCl2, 2.5 % Triton X-100 (v/v) for 15 min to remove SDS, followed by a brief rinsing in washing buffer without Triton X-100. The gels were then incubated at 37 °C for 12– 24 h in 50 mM Tris–HCl buffer containing 5 mM CaCl2, 1 μM ZnCl2, 1 % Triton X-100, 0.02 % NaN3, pH 7.6. The gels were stained with Coomassie Brilliant Blue R250 followed by destaining with 10 % acetic acid, 5 % methanol. The zymogen gels were later scanned by an image scanner and the areas of clear bands corresponding to MMP-2 and -9 activity were calculated using ImageJ software (NIH free software v1.42q software, USA, http://rsb.info.nih.gov/nihimage/) on non-altered original TIFF files in the 8-bit dynamic range of signal intensities. Additionally, total MMP-2 and total MMP-9 levels were calculated. These values are reported in arbitrary units (AU). Protein measurement Average protein concentrations (mg/ml) in the vitreous samples were measured using bicinchonic acid (BCA) protein assay kit (Pierce, Thermo Scientific, Rockford, IL, USA) according to the manufacturer’s instructions for a microplate procedure. Each vitreous sample was diluted with sterile water before the measurement, thus leading to sample dilutions between 1:30 and 1:50 in the BCA reaction mix that minimized any effects from blood (iron) or other interfering substances. All data were normalized per protein content for each sample. Results were analyzed both with and without protein standardization. Our data has been normalized to protein content. The correction for total protein in the vitreous was performed by dividing the original vitreous concentration of all the markers of interest separately by measured average protein concentration in the vitreous. Statistical analysis NCSS 2007 (NCSS, LLC, Kaysville, UT, USA) was used for statistical analysis. Power transformations were used for those continuous non-normally-distributed variables that could be transformed to an approximately normal distribution. Other non-normally-distributed continuous variables were analyzed using non-parametric tests. Normally distributed continuous variables were compared using two-sample t -tests. For non-normal variables, the Mann–Whitney U test was used instead. Spearman rank correlation was used to find dependencies between continuous variables. A multivariate logistic regression analysis was

Graefes Arch Clin Exp Ophthalmol

performed. A p-value less than 0.05 was considered statistically significant.

Results The mean vitreous Ang-2 concentration (107.7±68.8 pg/ml) was significantly higher in eyes with RRD compared to control eyes (43.0±29.5 pg/ml) (p <0.000001) (Table 1), and Ang1/Ang2-ratio was significantly lower (0.33±0.21) in eyes with RRD compared to control eyes (0.68 ± 0.47) (p < 0.000003). Intravitreal VEGF level was found to be lower in eyes with RRD (14.3±49 pg/ml) as compared with controls (40.3±165.8 pg/ml; p =0.01) (Table 1), but no differences were found in the intravitreal levels of Ang-1, EPO or total TGF-β1 between the two groups. Neither were any differences observed in plasma levels of Ang-1 or −2 between the groups. Concentrations of MMP-2 and -9 were significantly higher (p =0.002 and p ≤0.006) in eyes with RRD than in control eyes (Table 1).

Table 1 Intravitreal angiopoietin (Ang)-1 and −2, vascular endothelial growth factor (VEGF), erythropoietin (EPO), transforming growth factor (TGF)-β1 and matrix metalloproteinases (MMP-2 and −9) in eyes with rhegmatogenous retinal detachment (RRD group) and idiopathic macular hole/pucker (control group)

Intravitreal Ang-1 Ang-2 Ang1/Ang2-ratio Plasma Ang-1 Ang-2 Intravitreal VEGF EPO Total TGF-β1 Pro MMP-2 Total MMP-2 Pro MMP-9 Total MMP-9

RRD group (n =117)

Control group (n =40)

P-value

18.9±17.3 107.7±68.8 0.33±0.21

19.1±25.4 43.0±29.5 0.68±0.47

0.57 <0.000001* <0.000003*

4,139±3,217 2,058±938

4,035±3,546 2,623±2,142

0.80 0.13

14.3±49.0 12.8±9.4 5.8±10.2 1,779±931 1,864±988

40.3±165.8 27.4±68.0 7.8±12.8 1,262±958 1,294±980

0.01* 0.61 0.50 0.002* 0.002*

53.8±56.6 100.3±129.8

18.4±34.5 27.8±51.9

0.006* 0.003*

Subgroup analysis To investigate further whether the levels of the measured soluble factors in the vitreous correlated with the severity of RRD, we divided the study eyes according to the following parameters: 1) the extent of the RRD (Table 2), 2) the presence of vitreous haemorrhage, and 3) the PVR formation observed at the time of surgery (Table 3). In eyes with total RRD, the total TGF-β1 level was higher (p =0.05), while Ang-2 showed only a trend (p =0.056) towards increased level in total RRD eyes compared to eyes with partial RRD (Table 2). The mean vitreous concentration of total MMP-2 was also higher in eyes with total RRD (2,359± 1,197 AU) than in eyes with partial RRD (1,786±935 AU, p = 0.047). However, in pro MMP-2 levels, only a trend towards higher value in total RRD was observed (2,203±1,112 AU compared to 1,712±888 AU, p=0.07). A minor vitreous haemorrhage from an accompanying retinal tear was observed in eight eyes in the RRD group. In these eyes, the mean Ang-1 level was higher compared to those without haemorrhage (37.1±35.9 and 17.6±14.5 pg/ml respectively, p =0.004). Additionally, the total MMP-9 level was higher (224.8±161.5 AU) than in eyes without haemorrhage (86.4±119.5 AU, p =0.02). PVR formation was observed in 22 of the 117 eyes with RRD. In PRV eyes, no differences were observed in any marker of interest as compared with the non PVR eyes (Table 3).

Table 2 Intravitreal levels of angiopoietin (Ang)-1 and −2, vascular endothelial growth factor (VEGF), erythropoietin (EPO), transforming growth factor (TGF)-β1 and matrix metalloproteinases (MMP −2 and −9) in eyes with partial or total RRD Partial RRD (n =98) Total RRD (n =19) P-value

Data are given as mean ± SD. Ang-1 and −2, EPO, VEGF and TGF-β1 are given in pg/ml. Values of MMPs are given in arbitrary units (AU). Student’st-test was used in Ang-1, Ang-2, VEGF, EPO, Pro MMP-2 and Total MMP-2 analysis. Mann–Whitney test was used in TGF-β1, Pro MMP-9 and Total MMP-9 analysis. Additionally, plasma Ang-1 and −2 levels were measured. *Denotes a statistically significant difference between the groups

Ang-1 Ang-2 Ang1/Ang2-ratio VEGF EPO Total TGF-β1 Pro MMP-2 Total MMP-2 Pro MMP-9 Total MMP-9

17.7±13.5 101.3±64.3 0.35±0.42 11.5±45.1 12.7±9.8 5.6±10.8 1,712±888 1,786±935 51.6±57.2 92.8±118.1

25.1±14.7 142.5±97.0 0.21±0.15 30.6±65.6 13.1±6.8 6.9±6.4 2,203±1,112 2,359±1,197 64.7±54.8 136.4±178.0

0.41 0.056 0.2 0.1 0.44 0.05* 0.07 0.05* 0.32 0.32

Data are given as mean ± SD. Ang-1 and −2, EPO, VEGF and TGF-β1 are given in pg/ml. Values of MMPs are given in arbitrary units (AU). Student’st-test was used in Ang-1, Ang-2, VEGF, EPO, Pro MMP-2 and Total MMP-2 analysis. Mann–Whitney test was used in TGF-β1, Pro MMP-9 and Total MMP-9 analysis. *Denotes a statistically significant difference between the groups

Graefes Arch Clin Exp Ophthalmol Table 3 Intravitreal angiopoietin (Ang)-1 and −2, vascular endothelial growth factor (VEGF), erythropoietin (EPO), transforming growth factor (TGF)-β1 and matrix metalloproteinases (MMP −2 and −9) in eyes with rhegmatogenous retinal detachment with or without PVR

Ang-1 Ang-2 Ang1/Ang2-ratio VEGF EPO Total TGF-β1 Pro MMP-2 Total MMP-2 Pro MMP-9 Total MMP-9

PVR (n =22)

Non-PVR (n =95)

P-value

17.3±14.8 91.8±80.4 0.30±0.29 12.6±26.2 14.3±9.4 9.8±14.5 1,705±812 1,809±897 52.8±55.1 98.7±186.6

19.2±17.6 109.4±132.1 0.30±0.41 14.6±52.0 12.5±9.4 4.5±8.5 1,772±961 1,851±1,014 53.6±56.9 98.5±115.7

0.61 0.67 0.97 0.64 0.30 0.10 0.98 0.95 0.83 0.60

Data are given as mean ± SD. Ang-1 and −2, EPO, VEGF and TGF-β1 are given in pg/ml. Values of MMPs are given in arbitrary units (AU). Student’st-test was used in Ang-1, Ang-2, VEGF, EPO, Pro MMP-2 and Total MMP-2 analysis. Mann–Whitney test was used in TGF-β1, Pro MMP-9 and Total MMP-9 analysis. *Denotes a statistically significant difference between the groups

Correlation and logistic regression analysis Correlation analysis revealed that intravitreal concentrations of Ang-2 correlated significantly with Ang-1 (r =0.493, p < 0.0001), EPO (r =0.516, p <0.0001) and both pro MMP-2 and total MMP-2 levels (r =0.565, p <0.0001 and r =0.566, p < 0.0001 respectively). Additionally, EPO levels correlated with both pro MMP-2 and total MMP-2 levels (r =0.536, p < 0.0001 and r =0.527, p <0.0001). A multivariate logistic regression analysis suggested Ang-2 concentration in the vitreous being the key marker of RRD (p =0.00001, OR=48.7). According to logistic regression, low VEGF seemed to be less significant (p =0.02, OR=7.0).

Discussion The most remarkable finding of our study was the elevation of Ang-2 level and the decrease of Ang1/Ang2 ratio with simultaneous decrease in VEGF level in the eyes with RRD compared to control eyes. In fact, in multivariate analysis, Ang-2 level was the best discriminating factor between vitreous samples from eyes with RRD and control eyes, when all parameters studied were taken into account. Ang-2 is known to sensitize retinal vasculature to VEGF, but in the absence of VEGF, Ang-2 rather induces apoptosis of ECs and blood vessel regression [39], which could be relevant in eyes with RRD with increased Ang-2 and relatively low VEGF levels.

Moreover, Ang-2 has emerged as an important player in the induction of inflammation [12]. Ang-1 and -2 seem to act as “Yin Yang” antagonists of each other [36, 40]. In experimental studies, Ang-1 expressed by different cell types (pericytes, fibroblasts, vascular smooth muscle cells) has been linked to the prevention of RRD in proliferative retinopathy [41]. Ang-1 is known to suppress vascular plasma leakage in mature vessels, and to inhibit vascular inflammation, EC death, and apoptosis [42–44]. Thus, Ang-1 is thought to protect from later pathological consequences such as tissue fibrosis [36, 45]. Generally, in stable vessels Ang-1, expressed in capillary pericytes, promotes vascular integrity and stability, whereas Ang-2, released from the EC Weibel–Palade bodies, is capable of disrupting blood–tissue barriers and inducing leukocyte recruitment in response to traumatic stimuli. Ang-2 is mainly induced in ECs at the sites of vascular remodelling [39]. Increased levels of serum Ang-2 can even predict the prognosis of patients with systemic multiple traumatic injuries [46]. In addition, the Ang1/Ang2 ratio seems to be a critical switch controlling inflammatory processes, such as leukocyte transmigration in the vessel wall [12, 36]. Following EC activation (either due to inflammation, trauma, or hypoxia), Ang-2 is released rapidly and the expression of Ang-2 is upregulated strongly in ECs, shifting the local Ang-1/Ang-2 balance in favour of Ang-2 [12]. In view of the marked increase in Ang-2 levels in eyes with RRD, and even higher values in eyes with total RD, it is tempting to speculate that the blood–eye barrier loss seen in RD is somehow mediated by Ang-2 in response to the tissue trauma in RD. Taking into account the multiple actions of Ang-2, it may also be a key element in forming the conditions favourable for the PVR process. Pharmacologic inhibition of Ang-2 in eyes with RD could possibly prevent PVR formation in the future [39]. The role of Ang-1 was less clear in our study. In the subgroup analysis, we found a statistical difference in Ang-1 levels in eyes with vitreous haemorrhage compared to those without. This could favour for the protection mechanism being activated in these eyes. Tissue malfunction was observed only locally in the eye, since the systemic plasma Ang-1 or −2 levels were not altered between the study groups. Our results confirm the previous studies reporting increased intravitreal MMPs (pro MMP-2, total MMP-2, pro MMP-9 and total MMP-9) in eyes with RRD [28–30]. These MMP findings indicate that the mechanisms for tissue remodelling are activated early in the course of RRD. Our findings also indicate a novel association between the intravitreal levels of Ang-2 and MMP-2 and -9 in eyes with RRD. All together, these results suggest that the increased levels of Ang-2, MMP2 and -9 could contribute to the pathogenesis of RRD. Previously, upregulation of TGF-β pathway has been documented in several inflammatory and fibrous disorders. According to our study, the levels of total TGF-β1 were

Graefes Arch Clin Exp Ophthalmol

increased in more severe forms of RRD, supporting therefore its role in fibrosis formation [47]. Quite recently, increased levels of TGF-β1 were shown to be correlated with aqueous laser flare photometry values and with later PVR development in eyes with RRD [48]. Previously, also the elevated concentration of TGF-β2 has been reported to correlate with the severity of PVR [49]. Therefore, our results are in line with the potential development and implications for TGF-βinhibitors as a rational strategy in the prevention of vitreoretinal diseases in the future [50]. In RD, the outer retina (including photoreceptors) loses its oxygen supply from the adjacent RPE and choroid. Recent experimental and clinical studies have shown elevated levels of EPO in eyes with RD, increasing with the duration of RD, and in patients with RD and PVR [51, 52]. Thus, it was unexpected not to find any elevation of the ischemiasensitive mediators VEGF and EPO in our study groups. In fact, the mean VEGF concentration was significantly lower in eyes with RRD than in control eyes, suggesting that VEGF may be more relevant in other eye complications/conditions rather than RRD. Certainly, the duration of RD may have been longer in previous studies. One explanation to this phenomenon might also be that Müller cells, the major producers of VEGF, are still nourished from the inner retinal capillaries. Additionally, liquefied subretinal vitreous fluid could carry oxygen to the outer retina in RRD, and also the demand for oxygen may be diminished due to the loss of function of the photoreceptors (apoptosis starting within 1 to 3 days after RD) [2, 4]. Clinical neovascularization in eyes with RRD is not a common phenomenon, supporting our observations of low levels of VEGF in eyes with RD being understandable and clinically relevant. Similar explanation may apply to the low levels of EPO as well. Elevated levels of EPO have neuroprotective role in ischemic tissue. Thus, the lack of response in EPO levels in the eyes with RD may contribute to the apoptosis of retinal neuronal cells seen in RD. According to our study, EPO does not seem to play any major role early in the course of RRD. Only the correlation analysis was able to reveal highly significant correlation between EPO, Ang-2 and MMP-2 in eyes with RRD. This correlation could be true but unrelated, since in this aspect we rely only on statistical analysis. However, the intravitreal EPO may somehow be related to RRD, but we do not understand the mechanism. Perhaps the relative lack of EPO and VEGF suggests a compensatory mechanism of the neural retina in our clinical setting of RD entirely different from ischemic diseases, and the elevation of Ang-2 may reflect this mechanism. Additionally, the results could have been different if we had measured also other VEGF isoforms [53]. The strength of our study is its prospective design and larger number of eyes operated due to RRD when compared to previous studies [8–10], although larger subgroups would have been advantageous. Additional clinical data regarding

the duration and outcomes of the eyes with RRD could have given us better understanding of increased Ang-2 levels in those eyes. In-vitro experiments seem to be a necessity for the future development of this research. Since a representative material of vitreous from normal eyes in vivo is impossible to obtain, we used vitreous obtained from patients with macular hole or pucker as controls. These eye conditions, especially macular pucker, have a proliferative component in the acute phase of their pathogenesis. It is unresolved to what extent this affects our results. However, possible proliferative activity in the control eyes would rather have been expected to diminish the difference in the Ang-2 cytokine values between our RRD eyes and controls. The fact that we still could detect higher Ang-2 values in RRD vitreous suggests that these Ang-2 values were specific for RRD, or that the proliferative stimulus in RRD is much higher than in the control eyes. Additionally, our study was biased because of the difference in the age distribution of the patients. It is conceivable that aging affects the levels of the measured markers in the human vitreous. In clinical setting, we have to accept this, since it is difficult to obtain age-matched controls. Another possible bias of our study is that we did not use proteomic analysis [54]. We tried to overcome this problem by doing the main analysis from the data normalized to the amount of protein in the samples. We feel that this is a more exact way of analysing the data, which potentially corrects for serumderived mediators. However, the main results of this study would be similar if the analysis had been performed from actual concentration in the vitreous. To conclude, Ang-2 was markedly elevated in eyes with RRD, whereas VEGF showed lower values than in control eyes. In fact, logistic regression analysis suggested that high Ang-2 and low VEGF are the main determinants between RRD and control vitreous, suggesting a major triggering role for Ang-2. In the future, intravitreal levels of Ang-2 could serve as a marker of EC activation or dysfunction in RRD, and specific Ang-2 receptor inhibitors could possibly play a role in preventing PVR formation in eyes with RRD. Acknowledgments This study was supported by scientific grants from the Finnish Eye Foundation, The Eye and Tissue Bank Foundation, the Mary and Georg C. Ehrnrooth Foundation, the Nissi Foundation, and HUCH Clinical Research Grants (TKK4150 and TYH1325). Disclosure None to all authors.

References 1. Lewis GP, Guerin CJ, Anderson DH, Matsumoto B, Fisher SK (1994) Rapid changes in the expression of glial cell proteins caused by experimental retinal detachment. Am J Ophthalmol 118:368–376

Graefes Arch Clin Exp Ophthalmol 2. Cook B, Lewis GP, Fisher SK, Adler R (1995) Apoptotic photoreceptor degeneration in experimental retinal detachment. Invest Ophthalmol Vis Sci 36:990–996 3. Mitry D, Fleck BW, Wright AF, Campbell H, Charteris DG (2010) Pathogenesis of rhegmatogenous retinal detachment — predisposing anatomy and cell biology. Retina 30:1561–1572 4. Lo ACY, Woo TTY, Wong RLM, Wong D (2011) Apoptosis and other cell death mechanisms after retinal detachment: implications for photoreceptor rescue. Ophthalmologica 226(suppl 1):10–17 5. Hui YH, Goodnight R, Zhang XJ, Sorgente N, Ryan SJ (1988) Glial epiretinal membranes and contraction. Immunohistochemical and morphological studies. Arch Ophthalmol 106:1280–1285 6. Adamis AP, Miller JW, Bernal MT, D’Amico DJ, Folkman J, Yeo TK, Yeo KT (1994) Increased vascular endothelial growth factor levels in the vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol 118:445–450 7. Aiello LP, Avery RL, Arrigg PG, Keyt BA, Jampel HD, Shah ST, Pasquale LR, Thieme H, Iwamoto MA, Park JE (1994) Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 331:1480–1487 8. Ogata N, Nishikawa M, Nishimura T, Mitsuma Y, Matsumura M (2002) Inverse levels of pigment epithelium-derived factor and vascular endothelial growth factor in the vitreous of eyes with rhegmatogenous retinal detachment and proliferative vitreoretinopathy. Am J Ophthalmol 133:851–852 9. Kvanta A (2006) Ocular angiogenesis: the role of growth factors. Review article. Acta Ophthalmol Scand 84:282–288 10. Rasier R, Gormus U, Artunay O, Yuzbasioglu E, Oncel M, Bahcecioglu H (2010) Vitreous levels of VEGF, IL-8, and TNFalpha in retinal detachment. Curr Eye Res 35(6):505–509 11. Ricker LJAG, Diedonne SC, Kessels AG, Rennel ES, Berendschot TT, Hendrikse F, Kijlstra A, La Heij EC (2012) Antiangiogenic isoforms of vascular endothelial growth factor predominate in subretinal fluid of patients with rhegmatogenous retinal detachment and proliferative vitreoretinopathy. Retina 32(1):54–59 12. Fiedler U, Augustin HG (2006) Angiopoietins: a link between angiogenesis and inflammation. Trends Immunol 27(12):552–558 13. Tsanou E, Ioachim E, Stefaniotou M, Gorezis S, Charalabopoulos K, Bagli H, Peschos D, Psilas K, Agnantis NJ (2005) Immunohistochemical study of angiogenesis and proliferative activity in epiretinal membranes. Int J Clin Pract 59(10):1157–1161 14. Roberts A, Sporn MB (1993) Physiological actions and clinical applications of transforming growth factor-β (TGF-β). Growth Factors 8:1–9 15. Sivak JM, Fini ME (2002) MMPs in the eye: emerging roles if or matrix metalloproteinases in ocular physiology. Prog Retin Eye Res 21:1–14 16. Brown D, Hamdi H, Bahri S, Kennedy MC (1994) Characterization of an endogenous metalloproteinase in human vitreous. Curr Eye Res 13:639–647 17. Abu El-Asrar AM, Dralands L, Veckeneer M, Geboes K, Missotten L, Van Aelst I, Opdenakker G (1998) Gelatinase B in proliferative vitreoretinal disorders. Am J Ophthalmol 125:844–851 18. De La Paz MA, Itoh Y, Toth CA, Nagase H (1998) Matrix metalloproteinases and their inhibitors in human vitreous. Invest Ophthalmol Vis Sci 39:1256–1260 19. Kosano H, Okano T, Katsura Y, Noritake M, Kado S, Matsuoka T, Nishigori H (1999) ProMMP-9 (92 kDa gelatinase) in vitreous fluid of patients with proliferative diabetic retinopathy. Life Sci 64:2307– 2315 20. Das A, McGuire PG, Eriqat C, Ober RR, DeJuan E Jr, Williams GA, McLamore A, Biswas J, Johnson DW (1999) Human diabetic neovascular membranes contain high levels of urokinase and metalloproteinase enzymes. Invest Ophthalmol Vis Sci 40:809–813 21. Salzmann J, Limb GA, Khaw PT, Gregor ZJ, Webster L, Chignell AH, Charteris DG (2000) Matrix metalloproteinases and their natural

inhibitors in fibrovascular membranes of proliferative diabetic retinopathy. Br J Ophthalmol 84:1091–1096 22. Jin M, Kashiwagi K, Iizuka Y, Tanaka Y, Imai M, Tsukahara S (2001) Matrix metalloproteinases in human diabetic and nondiabetic vitreous. Retina 21:28–33 23. Noda K, Ishida S, Inoue M, Obata K, Oguchi Y, Okada Y, Ikeda E (2003) Production and activation of matrix metalloproteinase-2 in proliferative diabetic retinopathy. Invest Ophthalmol Vis Sci 44(5): 2163–2170 24. Beránek M, Kolar P, Tschoplova S, Kankova K, Vasku A (2008) Genetic variations and plasma levels of gelatinase A (matrix metalloproteinase-2) and gelatinase B (matrix metalloproteinase-9) in proliferative diabetic retinopathy. Mol Vis 14(14):1114–1121 25. Kon CH, Occleston NL, Charteris D, Daniels J, Aylward GW, Khaw PT (1998) A prospective study of matrix metalloproteinases in proliferative vitreoretinopathy. Invest Ophthalmol Vis Sci 39(8):1524– 1529 26. Symeonidis C, Diza E, Papakonstantinou E, Souliou E, Dimitrakos SA, Karakiulakis G (2007) Correlation of the extent and duration of rhegmatogenous retinal detachment with the expression of matrix metalloproteinases in the vitreous. Retina 27(9):1279–1285 27. Symeonidis C, Papakonstantinou E, Souliou E, Karakiulakis G, Dimitrakos SA, Diza E (2011) Correlation of matrix metalloproteinase levels with the grade of proliferative vitreoretinopathy in the subretinal fluid and vitreous during rhegmatogenous retinal detachment. Acta Ophthalmol Scand 89(4):339–345 28. Simo R, Carrasco E, Garcia-Remez M, Hernandez C (2006) Angiogenic and antiangiogenic factors in proliferative diabetic retinopathy. Curr Diab Rev 2:71–98 29. Spranger J, Meyer-Schwickerath R, Klein M, Schatz H, Pfeiffer A (1999) Deficient activation and different expression of transforming growth factor-beta isoforms in active proliferative diabetic retinopathy and neovascular eye disease. Exp Clin Endocrinol Diabetes 107: 21–28 30. Kita T, Hata Y, Arita R, Kawahara S, Miura M, Nakao S, Mochizuki Y, Enaida H, Goto Y, Shimokawa H, Hafezi-Moghadam A, Ishibashi T (2008) Role of TGF-beta in proliferative vitreoretinal diseases and ROCK as a therapeutic target. Proc Natl Acad Sci U S A 105(45): 17504–17509 31. Manzoni P, Maestri A, Gomirato G, Takagi H, Watanabe D, Matsui S (2005) Erythropoietin as a retinal angiogenic factor. N Engl J Med 353:2190–2191 32. Watanabe D, Suzuma K, Matsui S, Kurimoto M, Kiryu J, Kita M, Suzuma I, Ohashi H, Ojima T, Murakami T, Kobayashi T, Masuda S, Nagao M, Yoshimura N, Takagi H (2005) Erythropoietin as a retinal angiogenic factor in proliferative diabetic retinopathy. N Engl J Med 353:782–792 33. Caprara C, Grimm C (2012) From oxygen to erythropoietin: relevance of hypoxia for retinal development, health and disease. Prog Retin Eye Res 31:89–119 34. Junk AK, Mammis A, Savitz SI, Singh M, Roth S, Malhotra S, Rosenbaum PS, Ceramil A, Brines M, Rosenbaum DM (2002) Erythropoietin administration protects retinal neurons from acute ischemia-reperfusion injury. PNAS 16:10659–10664 35. Inomata Y, Hirata A, Takahashi E, Kawaji T, Fukushima M, Tanihara H (2004) Elevated erythropoietin in vitreous with ischemic retinal diseases. Neuroreport 15:877–879 36. Retina Society Terminology Committee (1991) An updated classification of retinal detachment with proliferative vitreoretinopathy. Am J Ophthalmol 112:159–165 37. Saharinen P, Alitalo K (2011) The yin, the yang, and the angiopoietin-1. J Clin Invest 121(6):2157–2159 38. Lohi J, Lehti K, Westermarck J, Kahari VM, Keski-Oja J (1996) Regulation of membrane-type matrix metalloproteinase-1 expression

Graefes Arch Clin Exp Ophthalmol

39.

40.

41.

42.

43.

44.

45.

46.

by growth factors and phorbol 12-myristate 13-acetate. Eur J Biochem 239(2):239–247 Lobov IB, Brooks PC, Lang RA (2002) Angiopoietin-2 displays VEGF-dependent modulation of capillary structure and endothelial cell survival in vivo. Proc Natl Acad Sci U S A 99:11205–11210 Maisonpierre PC, Suri C, Jones PF, Bartunkova S, Wiegand SJ, Radziejewski C, Compton D, McClain J, Aldrich TH, Papadopoulos N, Daly TJ, Davis S, Sato TN, Yancopoulos GD (1997) Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 277:55–60 Nambu H, Nambu R, Oshima Y, Hackett SF, Okoye G, Wiegand S, Yancopoulos G, Zack DJ, Campochiaro PA (2004) Angiopoietin 1 inhibits ocular neovascularization and breakdown of the blood-retinal barrier. Gene Ther 11(19):865–873 Thurston G, Rudge J, Ioffe E, Zhou H, Ross L, Croll SD, Glazer N, Holash J, McDonald DM, Yancopoulos GD (2000) Angiopoietin-1 protects the adult vasculature against plasma leakage. Nat Med 6: 460–463 Kim I, Moon S-O, Park SK (2001) Angiopoietin-1 reduces VEGFstimulated leucocyte adhesion to endothelial cells by reducing ICAM-1, VCAM-1, and E-selectin Expression. Circ Res 89:477–479 Kwak HJ, So JN, Lee SJ, Kim I, Koh GY (1999) Angiopoietin-1 is an apoptosis survival factor for endothelial cells. FEBS Lett 448:249– 253 Brindle NPJ, Saharinen P, Alitalo K (2006) Signaling and functions of angiopoietin-1 in vascular protection. Circ Res 98:1014–1023 Ganter MT, Cohen MJ, Brohi K, Chesebro BB, Staudenmayer KL, Rahn P, Christiaans SC, Bir ND, Pittet JF (2008) Angiopoietin-2, marker and mediator of endothelial activation with prognostic significance early after trauma? Ann Surg 247(2):320–326

47. Sieczkiewicz GJ, Herman IM (2003) TGF-beta 1 signaling controls retinal pericyte cotractile protein expression. Microvasc Res 66:190– 196 48. Hoerster R, Hermann MM, Rosentreter A, Muether PS, Kirchhof B, Fauser S (2013) Profibrotic cytokines in aqueous humor correlate with aqueous flare in patients with rhegmatogenous retinal detachment. Br J Ophthalmol 97:450–453 49. Pfeffer BA, Flanders KC, Guerin CJ (1994) Transforming growth factor beta2 is the predominant isoform in the neural retina, retinal pigment epithelium–choroid and vitreous of the monkey eye. Exp Eye Res 59:323–333 50. Gerhardinger C, Dagher Z, Sebastiani P, Park YS, Lorenzi M (2009) The transforming growth factor-[beta] pathway is a common target of drugs that prevent experimental diabetic retinopathy. Diabetes 58(7): 1659–1667 51. Xie Z, Wu X, Qiu Q, Gong Y, Song Y, Gu Q, Li C (2007) Expression pattern of erythropoietin and erythropoietin receptor in experimental model of retinal detachment. Curr Eye Res 32:757–764 52. Wang ZY, Shen LJ, Zhao KK, Song ZM, Qu J (2009) Elevated erythropoietin in vitreous of patients with rhegmatogenous retinal detachment and proliferative vitreoretinopathy. Ophthalmic Res 42(3):138–140 53. Woolard J, Wang WY, Bevan HS, Qiu Y, Morbidelli L, PritchardJones RO, Cui T-G, Sugiono M, Waine E, Perrin R, Foster R, DigbyBell J, Shields JD, Whittles CE, Mushens RE, Gillatt DA, Ziche M, Harper SJ, Bates DO (2004) VEGF165b, an inhibitory vascular endothelial growth factor splice variant: mechanism of action, in vivo effect on angiogenesis and endogenous protein expression. Cancer Res 64:7822–7835 54. Shitama T, Hayashi H, Noge S (2008) Proteome profiling of vitreoretinal diseases by cluster analysis. Proteomics Clin Appl 2(9):1265–1280