Effect of prostaglandin analogues on tear proteomics and expression of cytokines and matrix metalloproteinases in the conjunctiva and cornea

Abstract

The purpose of this work was to identify potential tear-film-based proteins and their effect on changes in the conjunctiva and cornea in eyes using prostaglandin (PG) analogues. Recruited subjects were individuals who had used PG for at least 1 year and comparison with eyes of normal controls and timolol using patients were done. Approximately 3–5 μL of tears were sampled from both eyes of each subject using glass microcapillaries. Proteomic analysis was done to compare the pooled tear samples from each group by Bradford assay and cytokine arrays. Impression cytology was used to gather mRNA from conjunctival epithelial cells, and target protein mRNA was quantified by PCR. Rabbits treated with PG were scarified, and changes in the corneal stroma were evaluated by immunohistochemical staining and western blot analysis. There were increased levels of IL-1β, IL-6, MMP-1, MMP-3, and MMP-9 and decreased levels of TIMP-1 and TIMP-2 in the tears of PG-treated patients. The mRNA of IL-1β, MMP-1, MMP-3, and MMP-9 was elevated and mRNA of TIMP-1 and TIMP-2 was decreased in the conjunctival epithelial cells. Rabbits treated with PG showed corneal thinning with decreased collagen type I expression. The protein of MMP-1 and MMP-9 was elevated and protein of TIMP-1 was decreased in the rabbit cornea by western blot analysis. Immunohistochemical staining showed elevated expression of MMP-1 and MMP-9 and the decreased expression of TIMP-1 in the corneal stroma. The topical use of PG analogues results in an altered balance between MMPs and TIMPs, which may be triggered by inflammatory cytokines. This results in an increase of matrix degradation and decrease of stromal collagens in the cornea by PG treatments.

Introduction

The prostaglandin (PG) analogs latanoprost, travoprost, and bimatoprost are one of the most prominent drugs used for the treatment of ocular hypertension and glaucoma (Parrish et al., 2003). Derived from PGF2α, these compounds have been proved to be highly efficient agents for lowering intraocular pressure (IOP), with few local and systemic side effects (Bito, 1997; Grierson et al., 2004; Lee and Goldberg, 2011).

The Ocular Hypertension Study Group reported longitudinal changes in central corneal thickness of study participants (Brandt et al., 2008). In their subgroup analyses, participants treated only with topical PG analogues had a greater rate of decrease in central corneal thickness per year than did participants treated only with topical β-blockers. Recently, a case of keratoconus progression associated with the use of topical latanoprost was reported (Amano et al., 2008). Moreover, progressive myopic regression or ectasia was reported with the topical use of PG analogues in eyes after refractive surgeries (Kamiya et al., 2008; Nowroozzadeh, 2009). Some corneal changes as a result of PG analogues may contribute to these clinical findings; however, the exact mechanism and contributing factors are not well known.

PG analogues are thought to show their effect of lowering IOP by upregulation of matrix metalloproteinases (MMPs) (Lindsey et al., 1996; Weinreb et al., 1997). Degradation of the extracellular matrix (ECM), including collagens within the ciliary muscle and sclera, results in decreased resistance for uveoscleral outflow (Ocklind, 1998; Sagara et al., 1999; Gaton et al., 2001). It has been postulated that upregulation of MMPs and subsequent effects on the ECM of the cornea may contribute to corneal thinning by PG analogues (Wu et al., 2005). The corneal stroma comprises 90% of the total corneal thickness and contains abundant type I collagens, which are secreted by keratocytes (BenEzra and Foidart, 1981; Lee and Davison, 1984). Increased expression of MMPs in the conjunctiva in vivo has been reported (Mietz et al., 2003). However, the effect of PG analogues on corneal stroma has been poorly investigated. Latanoprost has been reported to upregulate MMPs in cultured keratocytes (Wu et al. 2005). Proliferation, migration and viability of the keratocytes were reduced after exposure to latanoprost. However, an in vivo study that measured keratocyte density by scanning laser confocal microscopy in eyes using PG analogues showed contrary results. The keratocyte density was increased in PG-treated eyes in all layers of the corneal stroma (Bergonzi et al., 2010).

The aims of this study were to detect tear-film–based protein expression differences between PG-treated patients, normal controls, and timolol-treated patients. The different cytokines drawn from the tear proteomics were investigated in the conjunctival cells from impression cytology of PG-treated patients. After confirmation of the related cytokines that are changed by PG treatment, animal experiment was conducted to identify changes in the stromal tissue of the cornea.