Acta Ophthalmologica 2014

Conjunctival matrix metalloproteinases and their inhibitors in glaucoma patients Minna Helin-Toiviainen,1 Seppo R€ onkk€ o,2 Tuomo Puustj€arvi,1,3 Petri Rekonen,1,3 Minna Ollikainen3 4,5 and Hannu Uusitalo 1

Department of Ophthalmology, Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland School of Pharmacy, Pharmaceutical Technology, Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland 3 Department of Ophthalmology, Kuopio University Hospital, Kuopio, Finland 4 Department of Ophthalmology, University of Tampere, Tampere, Finland 5 Tampere University Hospital Eye Center, Medical School, University of Tampere, Tampere, Finland 2

ABSTRACT. Purpose: Chronic conjunctival inflammation, caused by various reasons, for example long-term use of topical drugs and/or their preservatives, affects the outcome of glaucoma surgery by interfering with wound healing. Matrix metalloproteinases (MMPs) remodel extracellular matrix (ECM) and are involved in the wound healing process. This study was designed to evaluate the conjunctival expression of MMPs and their tissue inhibitors (TIMPs) in the normal eye, primary open-angle glaucoma (POAG) and exfoliation glaucoma (ExG) and whether there is an association between staining intensities and deep sclerectomy outcome. Methods: Immunohistochemical procedures were performed on conjunctival samples which were obtained from POAG (n = 11) and ExG (n = 14) patients as well as normal (n = 7) subjects. Antibodies against MMPs (MMP-1, -2, -3 and -9) and TIMPs (TIMP-1, -2 and -3) were used. Results: In conjunctival stroma, expression levels of MMP-2 (p = 0.047), MMP-3 (p = 0.009), MMP-9 (p < 0.001), TIMP-1 (p = 0.003), TIMP-2 (p < 0.001) and TIMP-3 (p < 0.001) in ExG and MMP-9 (p = 0.008), TIMP-2 (p = 0.02) and TIMP-3 (p = 0.002) in POAG were significantly increased compared to control. We further found correlations between expression of MMP-1 and MMP-3 and the length of pilocarpine treatment. Conclusion: The expression of MMPs and TIMPs is increased in the conjunctiva of POAG and ExG patients having a long history of topical antiglaucoma drops. Antiglaucoma agents and/or their preservatives alter the remodelling balance of ECM in conjunctiva of POAG and ExG eyes. The balance between MMPs and TIMPs may play a crucial role in the conjunctival wound healing process and the outcome of glaucoma surgery. Key words: conjunctiva – deep sclerectomy – exfoliation glaucoma – primary open-angle glaucoma – scarring

Acta Ophthalmol. ª 2014 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd

doi: 10.1111/aos.12550

Introduction Matrix metalloproteinases (MMPs) are a family of zinc peptidases that degrade

and remove extracellular matrix (ECM) proteins and thus are involved in tissue remodelling, repair, wound healing, growth and development

(Birkedal-Hansen et al. 1993; Nagase et al. 2006). MMPs, also called matrixins, can be divided in collagenases (MMP-1, MMP-8, MMP-13, MMP18), gelatinases (MMP-2, MMP-9), stromelysins (MMP-3, MMP-10, MM P-11), matrilysins (MMP-7, MMP-26), membrane-type MMPs (MMP-14, MMP-15, MMP-16, MMP-24, MMP17, MMP-25) and others on the basis of structure similarities and substrate specificity (Wong et al. 2002; Iyer et al. 2012). The action of all of MMPs is regulated by a group of endogenous tissue inhibitors of metalloproteinases (TIMPs) (Nagase et al. 2006). The balance between MMPs and TIMPs in the ECM is of critical importance for the maintenance of the tissue (Kerrigan et al. 2000), and dysfunction in this balance can lead to alterations in extracellular matrix (ECM) that are shown during various conditions, such as inflammation, tumour invasion, angiogenesis and chronic degenerative diseases (Gomez et al. 1997; Mohammed et al. 2003; Seiki 2003; Ellenberg et al. 2010). MMPs have been reported to exist in almost all tissues of the eye, and they play a significant role in corneal wound healing and a variety of diseases in the retina (Sivak & Fini 2002). In the conjunctiva, studies have been conducted of the involvement of MMPs in wound healing (Chintala et al. 2005; Mizoue et al. 2006), ocular surface disorders (Sakimoto et al. 2004), allergy (Abu El-Asrar et al. 2001; Fuk-

1

Acta Ophthalmologica 2014

uda et al. 2006; Leonardi et al. 2007) and as an indicator of the influence of glaucoma medication in human conjunctiva (Terai et al. 2009). The state of the conjunctiva is known to be an important factor for the success of trabeculectomy (Broadway et al. 1994b) and deep sclerectomy (DS) (Helin et al. 2011). Therefore, the effects of chronic topical treatment and other factors that may interfere with the normal function and structure of the conjunctiva during the course of glaucoma are of vital importance. This study was designed to evaluate MMPs (MMP-1, MMP-2, MMP-3, MMP-9) and TIMPs (TIMP-1, TIMP-2, TIMP3) in the conjunctival stroma and epithelium of primary open-angle glaucoma (POAG) and exfoliation glaucoma (ExG) patients who underwent mitomycin C-augmented DS with implant and to compare the results to that seen in normal eyes. Our objective was to find out whether there is an association between staining intensities and surgery outcome.

Material and Methods Antibodies

For immunohistochemical staining of matrix metalloproteinases and their inhibitors in the conjunctiva, we used the following monoclonal antibodies: MMP-1 (mouse anti-MMP-1, Chemicon International, Ternecula, CA), MMP-2 (mouse anti-MMP-2, Lab Vision Corporation, Fremont, CA), MMP-3 (mouse anti-MMP-3, Lab Vision Corporation), MMP-9 (mouse anti-MMP-9, Lab Vision Corporation), TIMP-1 (mouse anti-TIMP-1, Lab Vision Corporation) and TIMP-2 (mouse anti-TIMP-2, Lab Vision Corporation). For staining TIMP-3, we used a rabbit polyclonal antibody (Lab Vision Corporation). TIMP-1, TIMP-2 and TIMP-3 all have the capability to inhibit the action of the selected MMPs. Further, MMP-1, MMP-2, MMP-3 and MMP-9 can break down a wide range of ECM collagens and other proteins. Samples

Preoperative comprehensive ophthalmic examination and the surgical procedure were carried out as described earlier (Ollikainen et al. 2010, 2011).

2

Conjunctival tissue samples were obtained from POAG (n = 11) and ExG (n = 14) patients during mitomycin C-augmented DS surgery with implant. In order to analyse the association between staining intensities and surgery outcome, we divided the patients into four groups whether surgery was successful (S) or failed (F). Sample sizes were the following: POAG S n = 6, POAG F n = 5, ExG S n = 8 and ExG F n = 6. All of our glaucoma patients used combinations of different topical eye drops preserved with benzalkonium chloride (BAK) until the day of surgery. No steroids were applied. The criteria for surgical success were intra-ocular pressure (IOP) ≤18 mmHg without the need for antiglaucoma medication and no requirement for repeated surgery. Our patients were followed postoperatively for 2.5
0.2 years. As control samples (n = 7), we used normal conjunctival biopsies from other ophthalmological patients undergoing surgery for retinal detachment, macular hole surgery and surgical correction of strabismus. The control subjects did not receive any antiglaucoma medication, and their conjunctivae were examined normal by an ophthalmologist. Surgically removed tissue samples were frozen in 70°C until used. Thin (5 lm) cryostat (Leica CM3050S, Leica Microsystems, Nussloch, Germany) sections were placed on Super FrostPlus microscope slides (Menzel GmbH & Co KG, Braunschweig, Germany). The study was conducted in compliance with the Declaration of Helsinki, and all patients gave their written consent to participate in the study. The study protocol was approved by the Ethical Committee of Kuopio University Hospital. The clinical data of glaucoma patients are presented in our previously published study (Helin et al. 2011). Immunohistochemistry

HistostainTM-Plus Mouse Primary Bulk kit (Zymed Laboratories, San Francisco, CA, USA) was used for staining MMP-1, MMP-2, MMP-3, MMP-9, TIMP-1 and TIMP-2. HistostainTM-Plus broad spectrum kit (Zymed Laboratories) was used for staining TIMP-3. Immunostaining was carried out as described earlier (R€ onkk€ o et al. 2007). For demonstrating the MMPs and

TIMPs in the conjunctiva, the antibodies were used at a dilution of 1:100. When staining negative control samples, the respective antibody was omitted from the staining process. Samples were photographed using a Nikon Eclipse E995 digital camera (Nikon, Tokyo, Japan). Image processing was carried out using Adobe Photoshop 9.0. Quantification of immunohistochemical staining

The evaluation of staining intensity was carried out as described earlier (Lehr et al. 1997; Lahm et al. 2004; R€ onkk€ o et al. 2007). From each sample, we analysed two (out of three) sections. All digital images (including negative control subjects) were first converted to 8-bit grayscale. We highlighted the histological area of interest (stroma versus epithelium) using the Magic Wand tool. Using the Histogram command, we were able to obtain an optical density plot from the selected area, and further the mean staining intensities in arbitrary units (AUs). Same procedures were used for analysing negative control samples. The staining intensities were elicited as the differences of AUs between the experimental and negative control sample. Statistical evaluation

The assessment of the demographics was conducted by analysis of variance (ANOVA) (SigmaPlot 11.0, Systat Software Inc, San Jose, CA, USA). For comparing statistical differences between immunostainings of different groups, we used the Mann–Whitney rank sum test. A probability value of