BritishJournal ofOphthalmology 1994; 78: 917-920 917

ORIGINAL ARTICLES - Laboratory science

Effect ofinterleukin 1 receptor antagonist on the blood-aqueous barrier after intraocular lens implantation

Okihiro Nishi, Kayo Nishi, Yasuichi Ohmoto

Abstract patients with proliferative diabetic retinopathy," 1 (IL-1) possesses as an intercellu- and in the aqueous humour following intraocular lar signal a wide spectrum of inflammatory, lens implantation in rabbits. 2 metabolic, haematopoetic, immunological, Interleukin 1 receptor antagonist (IL-lra) has and reparative properties and can be a media- recently been identified as the first described tor not only ofhost defence but also ofdisease. naturally occurring, endogenous receptor antag- Reduction of IL-1 can decrease the inflamma- onist ofany . It has a molecular weight of tory host response. A human recombinant IL-1 17000-25000 and inhibits IL-1 activities by receptor antagonist (IL-Ira) was used to block competitive binding to a specific receptor."' It is IL-1 after intraocular lens implantation in produced by monocytes, macrophages, neutro- rabbits. Seventeen rabbits underwent intercap- phils, fibroblasts, keratinocytes, and other sular phacoemulsification and posterior cham- epithelial cells."' It has been shown in various ber lens implantation. A 100 ig dose (0.1 ml) of animal models of disease'516 as well as in IL-lra (1 mg/ml) was injected into the anterior humans'7-'9 to block the activity of IL-1. It chamber at the end ofsurgery in seven rabbits. effectively reduced the inflammatory response of The 10 rabbits serving as the controls received anterior uveitis caused by intravitreal injection of no IL-lra. Postoperatively, all rabbits were IL-i 20 IL-ira may be effective for reducing observed with a slit-lamp, and the aqueous postoperative inflammation after intraocular lens flare intensity was measured with a laser flare implantation. cell meter at 12 hours, 1, 2, 3, and 4 days and We performed an experiment in rabbits to thereafter at 1, 2, 3, and 4 weeks. Aqueous flare examine whether IL-lra is effective in reducing intensity was significantly lower on days 2 and postoperative inflammation after cataract sur- 3, and fibrin deposition much less marked in gery. the eyes treated with IL-Ira, compared with the controls. The results suggest that IL-1 is involved in the postoperative inflammation that Materials and methods occurs after intraocular lens implantation and the use of the IL-lra would be valuable for- SURGICAL PROCEDURE reducing this problem. Seventeen rabbits weighing 1i5 to 2 kg were (BrJf Ophthalmol 1994; 78: 917-920) anaesthetised by the ketamine hydrochloride (5 mg/kg) and xylazine hydrochloride (2 mg/kg). Before surgery the pupil of the rabbit used was The cytokine interleukin 1 (IL-1) functions as an dilated with 1-0% tropicamide and 2 5% phenyl- intercellular signal and regulates locally and, at ephrine. Surgery was performed on one eye only. times, systemically the immunological, inflam- Intercapsular phacoemulsification following a matory, and host response to linear Nishi Eye Hospital reparative injury.' small upper anterior capsulotomy was 4-14-26, Nakamichi, The current studies with recombinant IL-1 in performed. To each 500 ml of balanced salt Higashinari-ku, Osaka human subjects revealed that IL-1 can be a solution (Alcon Inc, Fort Worth, Texas, USA), 1 537, Japan mediator not only of host defence but also of ml ofadrenaline hydrochloride at a concentration O Nishi K Nishi disease. Its over or prolonged production in of 1 mg/ml and 1000 units of sodium heparin either situation, however, will diminish or even were added to facilitate pupillary dilatation and Research Center of impair the normal host functions. Therefore, reduce the fibrinous reaction during surgery. Ohtsuka Pharmaceutical, 463-10, Kagasuno, control of IL-I synthesis or its effects becomes a After a posterior chamber lens with modified C Kawauchi-cho, target oftherapy in many diseases.2 loop and 6-5 mm optic was implanted, the axial Tokushima 777-01, In the ophthalmic field, the intravitreal injec- anterior capsule was removed in such a manner Japan Y Ohmoto tion of IL-I is known to cause an acute anterior so that the remaining anterior capsule covered uveitis in Interleukin 1 Correspondence to: rabbits." is synthesised the intraocular lens margin. Okihiro Nishi, MD, Nishi by various ocular cells such as retinal pigment A 0-I ml aliquot (100 rig) of human recombi- Eye Hospital, 4-14-26, corneal Muller and Nakamichi, Higashinari-ku, epithelial cells,6 cells,7 cells,8 nant IL-Ira, (Ohtsuka Pharmaceutical, Osaka 537, Japan. lens epithelial cells of human cataracts.9 Inter- Tokushima, Japan, 1 mg/ml) was injected into Accepted for publication leukin 1 was detected in the subretinal fluid of the anterior chamber at the end of surgery in 13 September 1994 retinal detachment in humans,'" in the vitreous of seven rabbits, and 0-1 ml of BSS was injected 918 Nishi, Nishi, Ohmoto

into 10 rabbits serving as the controls. A 30 tion increased. Compared with the controls, in gauge disposable needle attached to a 1 ml the eyes treated with IL-Ira, the mean value of syringe was pushed from the limbus across the aqueous flare intensity was lower at every the corneal stroma to make a long channel to measurement, with the exception of the value at prevent leakage of the aqueous humour after week 1. The mean value rather increased towards withdrawal ofthe needle. An antibiotic ointment the end of week 1 and then decreased in the eyes and 1% atropine sulphate ointment were then treated with the IL-Ira injection. The difference instilled. in the mean value of the aqueous flare intensity between both groups was statistically significant on day 2 (p<001) and day 3 (p<005) (multiple POSTOPERATIVE LASER FLARE CELL METRY AND comparisons according to Scheffe's method) SLIT-LAMP EXAMINATION (Fig 1). The anterior segment was carefully observed with a hand held slit-lamp (SL-14, Kowa Co, Japan), and aqueous flare was measured with a SLIT-LAMP EXAMINATION laser flare cell meter (FC-1000, Kowa Co) at 12 Fibrin deposition was seen in every control eye hours, 1, 2, 3, and 4 days, and thereafter 1, 2, 3, on days 1-4, and it needed 2 to 4 weeks to and 4 weeks. At each measurement, one of three disappear. In contrast with the untreated eyes, masked observers measured the flare intensity of fibrin deposition was much less marked and each eye five to 10 times. After the higher or appeared later in the treated eyes, as determined lower readings were omitted, the median value by the slit-lamp examination. among at least five remaining readings was All control eyes showed obvious secondary determined, and the values from seven treated or cataract formation with capsular opacification 10 untreated eyes were then average for each and synechiae on slit-lamp examination 6 months group. The mean value obtained was expressed after surgery. Red reflex from the fundus could as mean flare intensity for each group at each hardly be detected. Two of seven treated eyes measurement in photon counts/millisecond. The showed obviously less secondary cataract forma- regression equation between photon count and tion and less capsular opacification, so that the protein concentration (albumin) determined by red reflex was evident. We submitted these two the manufacturer is Y=-1-55+1-07X; Y=log eyes and one control eye chosen randomly, to y, and X=log x; y is protein concentration in histopathological examination to observe mg/ml, and x is photon count/ms. Coefficient of specifically the capsular bag in those eyes. correlation r=0-95. Accordingly, a reading of 457 photons/ms equals 20 mg albumin/ml. Six months after surgery, all rabbits under- HISTOPATHOLOGICAL EXAMINATION went slit-lamp examination again to observe In the capsular bag of the two treated eyes, LEC secondary cataract formation. proliferation was generally much less marked than that in the control eye. In the former, a few lens epithelial cells were observed on the pos- Results terior capsule, whereas lens epithelial cells were thickly layered in the latter. LASER FLARE CELL METRY In the untreated eyes, the mean aqueous flare was at its highest on day 2. At the same time, the Discussion remaining anterior capsule came in contact with Though our results are based only on one experi- the posterior chamber lens, and the lens epithe- ment, they showed that human IL-Ira decreased postoperative inflammation after intraocular lens Figure I Aqueous flare lial cells (LECs) under the capsule began to intensity after intraocular proliferate and opacify. The aqueous flare implantation in the rabbit eye when it was lens implantation in rabbit decreased gradually thereafter, while the anterior injected into the anterior chamber at the end of eyes. * and ** indicate capsule opacification caused by LEC prolifera- surgery. Its significant (decreasing) effect on statistically significant after 2 to 3 differences at p<005 and aqueous flare intensity was manifested p<001, respectively. days. The gradual increase in flare intensity thereafter suggests that IL-ira disappeared grad- E ually. IL-Ira may have been metabolised or 0C cleared in those eyes. 0 Yokoyama'2 reported that the IL-1 B concen- 0 tration in the anterior chamber at day 7 after 0 0 intraocular lens implantation in rabbits was 10 12 (SD 3 96) ng/ml aqueous humour; Rosen- ._ ax baum et a120 injected 75 tg of human recombi- Z._ nant IL-lra intravitreally in rabbit eyes to assess C e the potential activity of the IL-Ira in ocular inflammation induced by the intravitreal injec- C,0) 0) tion of IL-1. In the present study, we chose the e dose of IL-Ira, because it appeared sufficient to 0 block IL-I in the anterior chamber of the rabbit 0) 0~ and, therefore, to assess the effect. 0 I I Even though the IL-Ira used was the human Preop 12h 1 d 2d 3d 4d 1 w 2w 3w 4w antagonist, it effectively blocked IL-1 in the Postoperative period rabbit eye. There are parallel results20 that Effect ofinterleukin I receptor antagonist on the blood-aqueous bamrer after intraocular lens implantation 919

human IL-Ira reduced rabbit ocular inflamma- effects ofPGE2. This can be supported by the fact tion induced by exogenous human IL-1. Inter- that inhibitors of eicosanoid synthesis inhibited leukin 1 is known to induce IL-1 synthesis the ocular inflammatory effects of exogenous during proliferation of cell production.62 These IL-1 ca.5 However, Kulkarni and Mancino34 have facts suggest that human IL-Ira can bind com- reported that ocular inflammation, induced by petitively to the rabbit IL-1 receptor. Although surgical paracentesis or intravitreal endotoxin the aqueous flare intensity reached almost the injection, was not associated with PGE2 same level as that of the control animals, the accumulation in the aqueous humour, so that the significant effect at days 2 and 3 seemed to result eicosanoid inhibitors are not effective for in less marked fibrin deposition in the eyes reducing inflammation, and the presence of IL-I- receiving the IL-Ira injection. Sustained release like activity could not be detected. Our results of IL-lra using a proper drug delivery system and these reports suggest that, with respect to the might work more effectively to reduce disruption pathogenetic process involved, surgically ofthe blood-aqueous barrier. induced inflammation with proliferating residual From the histopathological examination, we LECs shares a common aspect with inflammation were uncertain whether the less marked second- induced by exogenous but not with ary cataract in two of seven eyes was actually that induced by surgical paracentesis or exogen- caused by the effect of IL-lra injection. IL-1 is ous endotoxin in an eye with an uninjured known to increase epithelial cell division in crystalline lens. Proliferating residual LECs may general.' It increased the tritiated thymidine play an important role here. Residual LECs can incorporation into cultured lens epithelial cells of disrupt the blood-aqueous barrier after intra- human cataracts in our vitro study. (Symposium ocular lens implantation,35 which suggests a on Cataracts and Refractive Surgery, 11 April, parallel with the fact that human cataract LECs 1994, Boston) Because only one eye in each synthesise IL-I and PGE2.9 Thus, in the present rabbit was treated surgically in this study, this study, the IL-I and PGE2 synthesis by residual might be due to differences in proliferation of LECs, apart from IL-1 production by other cells lens epithelial cells in the individual rabbit. as a result of surgical trauma, may have been Moreover, not all eyes underwent histo- blocked or decreased by IL-Ira, which contri- pathological examination. Further study is buted to the decrease in disruption of the blood- needed to confirm whether IL-Ira has a suppres- aqueous barrier. sive effect on lens epithelial cell proliferation. In conclusion, the present study suggests that Our results suggest that IL-I is implicated in IL-1 is involved in postoperative inflammation the postoperative inflammation after intraocular after intraocular lens implantation, and IL-Ira lens implantation, as it was detected in the seems to be valuable for decreasing surgically aqueous humour following intraocular lens induced inflammation in the rabbit eye. Deter- implantation in rabbits.12 Interleukin 622 was mining of the role of in ocular inflam- detected in the aqueous humour following intra- mation may lead to new treatments for the ocular lens implantation in humans. Interleukin control of the effects of these important media- 6 caused an acute inflammation subsequent to the tors. intravitreal injection22 and is known to be The authors have no proprietary interest in the methods or induced by IL-12324 in the inflammatory or products mentioned in this paper. Part of this paper was presented at the symposium on cataract reparative host response to injury. However, and refractive surgery, 11 May 1993 in Seattle, USA. Lundgren et al could not detect IL-125 in the aqueous humour after intraocular lens implanta- 1 Di Giovine FS, Duff GW. Interleukin 1: the first interleukin. Immunol Today 1990; 11: 13-20. tion in rabbits; according to the authors this 2 Dinarello CA. Interleukin-1 and interleukin-I antagonism. might be because ofinsufficient assay sensitivity. Blood 1991; 77: 1627-52. 3 Rosenbaum JT, Samples JR, Hefeneider SH, Howes EL Jr. Cytokines act at concentrations of 10-' to 1O'5 Ocular inflammatory effects of intravitreal interleukin-1. mol/l to stimulate target cell functions, and such Arch Ophthalmol 1987; 105: 1117-20. 4 Bhattacherjee P, Henderson B. Inflammatory responses to a low concentration range aggravates the detec- intraocularly injected interleukin-1. Curr Eye Res 1987; 6: tion problems.' 929-34. several acute or chronic diseases, the 5 Tilden ME, Boney RS, Goldenberg MM, Rosenbaum JT. The During effects of topical S(+)-ibuprofen on interleukin-1-induced effect ofspecific blocking ofIL-I has not yet been ocular inflammation in a rabbit model. J Ocul Pharmacol clarified. Rosenbaum and co- 1990; 6: 131-5. completely 6 Planck SR, Huang XN, Ansel JC, Robertson JE, Rosenbaum workers20 reported that IL-Ira was effective in JT. Retinal pigment epithelial cells produce interleukin-1 reducing IL-1 induced inflammation in rabbit beta and granulocyte-macrophage colony-stimulating factor in response to interleukin-l alpha. Curr Eye Res 1993; 12: eyes, but did not produce a significant reduction 205-12. in inflammation subsequent to an active Arthus 7 Shams NBK, Reddy CV, Watanabe K, Elgebaly SA, Hanninen LA, Kenyon KR. Increased interleukin-1 activity reaction or the intravitreal application of endo- in the injured vitamin A-deficient cornea. Cornea 1994; 13: toxin. A parallel result'9 was reported that intra- 156-66. 8 Roberge FG, Caspi RR, Nussenblatt RB. Glial retinal Muller ventricular injection of IL-Ira attenuated the cells produce IL-1 activity and have a dual effect on febrile response to a subsequent intracerebro- autoimmune T helper lymphocytes. J Immunol 1988; 140: 2193-226. ventricular bolus ofIL-1, but had no effect on the 9 Nishi 0, Nishi K, Imanishi M. Synthesis of interleukin- I and to prostaglandin E, by lens epithelial cells of human cataracts. response the endotoxin. Prostaglandin E2 Br]r Ophthalmol 1992; 76: 338-41. _ _ (PGE2) plays a major role in postoperative inflam- 10 Davis JL, Jalkh AE, Roberge F, Caspi R, Flynn HW Jr, Schepens CL, et al. Subretinal fluid from human retinal mation after intraocular lens (IOL) implanta- detachment contains interleukin-1. Invest Ophthalmol Vis tion.2627 The effect of IL-ira in the present study Sci 1988 (Suppl); 29: 396. be 11 Abu el Asrar AM, Maimone D, Morse PH, Gregory S, Reder may accounted for by the blockage of IL-I AT. Cytokines in the vitreous of patients with proliferative that stimulates PGE2 synthesis by activating the diabetic retinopathy. Am J7 Ophthalmol 1992; 114: 731-6. arachidonic acid 12 Yokoyama T. Interleukin-l in the aqueous humor in aphakic cascade.28`0 Recent reports3"3- and pseudophakic eyes of rabbits. Acta Soc Ophthalmoljpn showed that IL-Ira inhibits the production and 1992;96:67-73. 920 Nishi, Nishi, Ohmoto

13 Hannum CH, Wilcox CJ, Arend WP, Fennecke G, Joslin DT, 25 Lundgren B, Ocklind A, Holst AH, Harfstrand A. Inflamma- Dripps PL, et al. Interleukin- receptor antagonist activity tory response in the rabbit eye after intraocular implantation of a human interleukin-l inhibitor. Nature 1990; 343: with poly(methyl methacrylate) and heparin surface modi- 336-40. fied intraocular lenses. Cataract Refract Surg 1992; 18: 14 Arend WP. Interleukin-1 receptor antagonist. Adv Immunol 65-70. 1993; 54: 167-227. 26 Miyake K, Mibu H, Horiguchi M, Shirasawa E. Inflammatory 15 Ohlsson K, Bjork P, Bergenfeldt M, Hageman R, Thompson mediators in postoperative aphakic and pseudophakic RC. Interleukin-1 receptor antagonist reduces mortality baboon eyes. Arch Ophthalmol 1990; 108: 1764-7. from endotoxin shock. Nature 1990; 348: 550-5. 27 Torngren L, Roltsen W, Lundgren B. PGE, and lencocytes 16 Nast CC, Thompson RC, Dinarello CA. In vivo antiinflamma- level in aqueous humor after lens extraction and intraocular tory properties ofrecombinant interleukin-1 receptor antag- lens implantation. Ocular Immunology and Inflammation onist (IL-lra). Lymph Res 1990; 9: 307-400. 1993; 1: 151-7. 17 Rambaldi A, Torcia M, Bettoni S, Barbui T, Vannier E, 28 Chang J, Gilman SC, Lewis AJ. Interleukin-I activates Dinarello CA, et al. Modulation of cell proliferation and phospholipase A2 in rabbit chondrocytes: a possible signal by cytokine production in acute myeloblastic leukemia for IL-I action. J Immunol 1986; 136: 1283-7. interleukin-1 receptor antagonist and lack of its expression 29 O'Neill LAJ, Barett ML, Lewis GP, Induction of cyclo- by leukemic cells. Blood 1990; 76: 114-8. oxygenase by interleukin-I in rheumatoid synovial cells. 18 Guise TA, Garrett IR, Bonewold LE, Mundy GR. Inter- FEBS Lett 1987; 212: 35-9. leukin-1 receptor antagonist inhibits the hypercalcemia 30 Farrar WL, Humes JL. The role of arachidonic acid metabol- 8: mediated by interleukin-1. Bone Miner Res 1993; ism in the activities of interleukin 1 and 2.J Immunol 1985; 583-7. 135: 1153-9. 19 Coceani F, Lees J, Redford J, Bishai I. Interleukin-1 receptor Effect of treatment effectiveness against interleukin-l fever. Can 7 31 Meyers KP, Czachowski CL, Coffey JW. antagonist: with interleukin-1 receptor antagonist on the development of Physol Pharmacol 1992; 70: 1590-6. carrageenan-induced pleurisy in the rat. Inflammation 1993; 20 Rosenbaum JT, Boney RS. Activity of an interleukin 1 17: 121-34. receptor antagonist in rabbit models of uveitis. Arch 110: 547-9. 32 Conti P, Barbacane RC, Felaco M, Grilli A, Placido FC, Reale Ophthalmol 1992; M. Human recombinant interleukin-1 receptor antagonist T, Warner STC. Interleukin 1 induces 21 Dinarello CA, Ikeiima (hr IL-1 RA) inhibits prostaglandin E, generation but not interleukin 1. Induction of circulating interleukin 1 in vivo chronic- rabbits in vivo and in human mononuclear cells in vitro. alkaline phosphatase activity by in Immunol 1987; 139: 1902-10. granulomatous tissue induced by KMnO. Immunol Lett 22 Malecaze F, Chollet P, Cavrois E, Vita N, Arrie JL, Ferrara P. 1993; 37: 1-6. response after 33 Seckinger P, Klein-Nulend J, Alander C, Thomson RC, Dayer Role of in the inflammatory and recombinant human IL-1 cataract surgery. An experimental and clinical study. Arch JM, Raisz LG. Natural receptor antagonists block the effects of IL-1 on bone Ophthalmol 1991; 109: 1681-3. Immunol 1990; 23 May LT, Torua C, Cozzolino F. Interleukin 6 gene expression resorption and prostaglandin production. in human endothelial cells: RNA start sites, multiple IL-6 145: 4181-4. proteins and inhibition ofproliferation. BiochemBiophys Res 34 Kulkarni PS, Mancino M. Studies on intraocular inflammation Commun 1989; 159: 991-8. produced by intravitreal human interleukins in rabbits. Exp 24 Planck SR, Dang TT, Graves D, Tara D, Ansel JC, Rosen- Eye Res 1993; 56: 275-9. baum JT. Retinal pigment epithelial cells secrete inter- 35 Nishi 0, Nishi K. Disruption of the blood-aqueous barrier by leukin-6 in response to interleukin- 1. Invest Ophthalmol Vis residual lens epithelial cells after intraocular lens implanta- Sci 1992; 33: 78-82. tion. Ophthalmic Surg 1992; 23: 325-9.