Sim ZH, Pinard CL, et al. Cyclooxygenase-2 expression in the eyes of cats with and without uveitis. Am J Vet Res 2018;79:90-97.
Uveitis, inflammation of the middle, highly vascular layer of tissue (uvea) in the wall of the eye, is a common, painful, and potentially vision-threatening condition in cats. The uvea comprises the iris, ciliary body, and choroid, and is situated between the sclera and inner retinal layer. Feline uveitis has a wide variety of etiologies, but the etiology of endogenous uveitis in an individual patient may be difficult to determine. When uveitis occurs, regardless of etiology, it is due to the release of inflammatory mediators following cell injury. This in turn causes disruption of the blood-aqueous barrier, allowing inflammatory cells, proteins, and even more inflammatory mediators to flood into the uveal stroma, anterior chamber, and vitreous of the eye. At this point, uveitis becomes clinically apparent.
Among the ocular inflammatory mediators involved in the etiopathogenesis of uveitis is cyclooxgenase-2 (COX-2). COX-2 is a part of the arachidonic acid inflammatory cascade and at times of cell injury, catalyzes the production of prostaglandins, which in turn cause the miosis, aqueous flare, and decreased intraocular pressure associated with uveitis. Studies in rabbits and dogs have demonstrated the upregulation of COX-2 in glaucoma. In humans and dogs, COX-2 is expressed in ocular melanomas. In cats, however, information regarding COX-2 activity in ocular tissue is lacking.
In this study of archived ocular tissue specimens from 51 cats with ocular disease (cases) and 10 cats without ocular disease (controls), the authors sought to learn more about the distribution and intensity of COX-2 expression in feline eyes with and without uveitis, as well as to identify any correlation between COX-2 expression and severity of ocular inflammation. Tissues from only one eye from each cat were evaluated with H&E stain to identify the presence of inflammatory cells within the uveal tract and other portions of the eye, as well as for diffuse iris melanoma, by finding neoplastic melanocytes within the iris. The diffuse iris melanoma cases were further subcategorized as inflammatory or noninflammatory. Ocular specimens were also stained immunohistochemically for COX-2 expression.
Of the 51 case cats, signalment information was available for 45. These animals had a mean + standard deviation age of 9.4 + 4.5 years; 26 were males and 19 were females. Domestic shorthair was the most common breed (n=28), followed by domestic longhair (n=6). The 10 controls were young adult male domestic shorthair cats.
Histopathologic evaluation of the 51 diseased eyes revealed lymphocytic-plasmacytic uveitis (n=20), neutrophilic uveitis (n=13), diffuse iris melanoma-induced uveitis (n=11), and diffuse iris melanoma without uveitis (n=7). Immunohistochemical staining demonstrated COX-2 expression in 16 uveae of the 44 eyes with uveitis; this included 11 eyes with lymphocytic-plasmacytic uveitis, 4 with neutrophilic uveitis, and one eye with diffuse iris melanoma-induced uveitis. Of the 16 eyes expressing uveal COX-2, 10 had histologic evidence of severe uveitis; 5 had moderate uveitis, and one was judged to have mild uveitis. None of the control eyes had evidence of COX-2 expression, and COX-2 staining was not found in the sclera, lens, retina, and choroid of the 51 case eyes. Only one of the 11 eyes with with diffuse iris melanoma-associated uveitis had low level uveal COX-2 expression. All other eyes with diffuse iris melanoma did not express any uveal COX-2, whether or not uveitis was present.
In 22 of the 51 case eyes, COX-2 expression was identfied in the cornea; 19 of these 22 eyes had evidence of corneal inflammation. Of the 22 case eyes with corneal COX-2 expression, 21 had uveitis and one had diffuse iris melanoma without uveitis. Therefore more case eyes with uveitis had COX-2 expression in the cornea (n=21) than in the uvea itself (n=16). It is possible that, in feline uveitis, COX-1 or lipoxygenase more commonly cleaves arachidonic acid than COX-2, given that only 36% (16/44) of the case eyes with uveitis stained positively for COX-2. Another potential explanation for the relatively low percentage of case eyes with positive uveal staining for COX-2 is that this enzyme may only be transiently or intermittently expressed in the course of feline uveitis, but no information is available regarding this possibility.
When uveitis was severe in an eye, it was positively correlated with COX-2 expression in both the uvea and the cornea. In those case eyes for which disease history was available (29 of 51 cases), the COX-2 expression was not associated with any prior treatment, with corticosteroid or NSAID treatment prior to enucleation, or with uveitis sequelae. The underlying cause of the uveitis was not identified for any of the eyes in this study. In contrast to dogs, in which eyes with uveal melanoma and concurrent uveitis tend to express COX-2, only one of the 11 feline eyes with diffuse iris melanoma and concurrent uveitis evaluated in this study expressed COX-2.
Based on the results of this study, although expression of COX-2 in feline uveitis is not universal and variable in severity, COX-2 inhibition may be of therapeutic value in the treatment of this condition. Additional studies involving the measurement of COX-2 and other inflammatory mediators in the aqueous humor of affected cats are recommended by the authors; such studies would avoid enucleation and allow identification of other potentially significant inflammatory mediators in uveitis such as COX-1 and lipoxygenase. [PJS]
Maggs DJ. Feline uveitis. An ‘intraocular lymphadenopathy.’ J Feline Med Surg 2009;11:167-182.
Wiggans KT, Reilly CM, et al. Histologic and immunohistochemical predictors of clinical behavior for feline diffuse iris melanoma. Vet Ophthalmol 2016;19:44-55.