`iii
`:13 AMERKAN ACADEMY
`or OPHTHALMDLOGV
`
`The Eyre M.D. Assuciatiun
`
`
`
`Sebaflemm Carcinuma 0f the Eyelid.
`
`.
`
`Cemtml Carneal Thicknfisx
`
`Cimrse {if Cymmegalnvinm Retinitis
`
`Retinfipathy 0f Prematurity
`
`
`
`Bphthaimulugy
`BEL annex
`UC Ban fliega
`Received an:
`
`lfl-figwgé
`
`‘Tiumelll NumhcrEE f Decemberlfifi4 E ELSEViER 188N4Hfi1-6420
`
`Argentum Pharm. LLC V. Alcon Research, Ltd.
`CaseIPR2017-01053
`
`ALCON 213 1
`
`
`
`Ophthahnology
`
`Journal of the
`American Academy of Ophthalmology
`Aims and Scope: Ophthalmology, the official journal of the American Academy of Ophthalmology, publishes
`original, peer-reviewed reports of research in ophthalmology, including basic science investigations and clinical
`studies. Topics include new diagnostic and surgical techniques, treatment methods, instrument updates, the latest
`drug findings, results of clinical trials, and research findings. Ophthalmology also publishes major reviews of specific
`topics by acknowledged authorities.
`
`Editor-in-Chief
`Andrew P. Schachat*
`
`Associate Editor-in-Chief
`Henry D. Jampel*
`
`EDITORIAL BOARD
`Douglas R. Anderson*
`Miami, FL
`
`James D. Brandt
`Sacranrl'llto, CA
`Wiley A. Chambers, II
`Wasl,ingto11, DC
`Thomas S. Chang
`Los Angeles, CA
`Susan H . Day
`Sm, Frc111risro, CA
`William R. Freeman
`L1jolla, CA
`Glenn J. Jaffe
`D1.1rha111, NC
`Yoshiaki Kitazawa
`Giji1, Japa11
`Ronald Klein
`Madiso,,, WT
`
`Associate Editor, Genetics
`Richard Alan Lewis
`Housto11, TX
`
`Maureen G. Maguire
`Pl,iladelphia, PA
`
`Steven A. McCormick
`New York, NY
`
`Joel S. Mindel*
`New York, NY
`
`Arthur J. Mueller*
`M11e11d1e11, Gem1a,1y
`
`David C. Musch
`A 11n Arl,or, lvfl
`
`Steven A. Newman
`CJ,,,rlorrewille, VA
`
`Joan M. O'Brien*
`Sa11 Francisr,1, CA
`
`Denis M . O'Day
`Naslwille, TN
`
`Roger F. Steinert
`Bvsto11, MA
`
`Richard A. Stone
`Philadelphia, PA
`
`Alan Sugar
`Ann Arlwr, MI
`
`Rohit Varma
`Los Angeles, CA
`Thomas A. W eingeist
`Iowa City, IA
`Richard W ormald
`Lo11do11, England
`
`EDITORIAL STAFF
`Veronica Doyle
`Ma11agi11g Editor
`Miranda Walker
`Assistant Ma11agi11g Editor
`Tom Juhasz
`Editorial Assistant
`
`ACADEMY STAFF
`Richard A. Zorab
`VP, Ophtlrn/11,ic K110111/edge
`Hal Straus
`Director of Publications
`
`* t\Jlc111bl'r, EdiMrial Advisory Co111111irtec
`
`[ll(zuiries or wmments regardi11g tl,e j o11rnc1I may be directed tv the Editorial Advisory Board, a11011y111ously if so desired.
`Addresses of Board 111embers may be obtai11ed in th£' AA O Member Directory or from tl,e editorial ~{]ice.
`
`Op/1t/1a/111,>i,i~Y (ISSN 111 6 1-6420) is published monthly for th< American Academy ofOphthalmologv. Inc. (For Post Office use only: Volume 111 issue 12 of 12) by Elsevier Inc.,
`Y,O Park Avenue South. New York. NY f(J/Jlll- 17H>. © ~0114 American Ac.1demy ofOphrhalmolob"'· Inc. All righc; reserved. Printed in the USA. Periodicals postage paid .lt New
`York, NY :md addition.ii m.tiling ollict:'s. Pm tmaster: Send .tddress changes co Elst:'vie-r, Journals Cui.comer Service, 6277 St'J Harbor Dr. , O rl:mdo, FL 32887-4800. M ember ch ange
`o f address or claim for issues not received : American Academy of Ophthalmology, 655 Beach St., P.O. Box 7424, San Francisco, CA 94120; E- mail:
`Journal_ lnquiries@aao.org.
`
`
`
`Conjunctiva! Epithelial Cell Expression
`of Interleukins and Inflammatory Markers
`in Glaucoma Patients Treated over the
`Long Term
`
`Christophe Baudouin , MD, PhD ,1•2 Pascale Hamard , MD, PhD,1 ·2 Hong Liang, MD, 2
`Catherine Creuzot-Garcher, MD, PhD ,3 Larry Bensoussan , MD, 1 Fran'roise Brignole, MD, PhD2 •4
`
`Purpose: To compare the conjunctiva! epithelial cell expressions of inflammatory cytokines in normal
`subjects and in glaucoma patients treated over the long term.
`Design: Case-control study.
`Participants: A total of 69 glaucoma patients treated over the long term and 15 normal subjects with no
`ocular abnormc!lity or topical treatment.
`Methods: Amongst the 69 glaucoma patients, 27 were treated with preserved {3-blockers, 24 with unpre(cid:173)
`served 0.5% timolol, and the other 18 patients with an association of 2:2 preserved drugs. All patients were
`treated for more than 1 year with the same treatment, with no significant differences between groups for mean
`ages and durations of treatment at the time -of the study. Impression cytol9gy specimens were taken and
`processed for immunofluorescence techniques. Conjunctiva! cell expressions of HLA DR, as a standard for
`inflammatory level, and the interleukins IL-6, IL-8, and IL-10 were obtained and quantified using flow cytometry.
`Main Outcome Measures:
`Immune markers and proinflammatory cytokines in impression cytology
`specimens.
`Results: We found a significantly increased expression of all immunoinflammatory markers and mediators in
`the conjunctiva! epithelium of glaucoma patients compared with normal eyes. Human leukocyte antigen DR was
`significantly higher in the 2 groups receiving preserved drugs than in the unpreserved timolol group. The 3
`interleukins were similarly overexpressed in all glaucoma groups, with no significant between-groups differences
`except for the expression level of IL-8, which was significantly higher in the multitreatment group than in the
`preservative-free one.
`Conclusions: The present study confirms the increased expression of immunoinflammatory markers by the
`conjunctiva! epithelium of glaucoma patients treated over the long term. The development of nontoxic preser(cid:173)
`vatives or preservative-free solutions is therefore of great interest. Ophthalmology 2004;111 :2186-2192 © 2004
`by the American Academy of Ophthalmology.
`
`There is now growing evidence from experimental and
`clinical studies that the long-term use of antiglaucoma drugs
`may induce ocular surface changes, causing ocular discom(cid:173)
`fort upon instillation,1 tear film instability,2 conjunctiva!
`inflarnmation,3 subconjunctival fibrosis,4 conjunctiva] epi(cid:173)
`thelium apoptosis,5 corneal surface impairment, 1•6 and po(cid:173)
`tential risk for failure of further glaucoma surgery. 7•8 Sub-
`
`inflammation has also been described with
`clinical
`significant infiltration of the conjunctiva] epithelium and
`substantia propria by inflammatory cells,3 •9 • 10 as has con(cid:173)
`junctiva] epithelial cell expression of inflammatory mark(cid:173)
`ers, 11 • 12 in patients receiving antiglaucoma treatments for
`long periods. However, the respective roles of the active
`compound and the preservative in inducing toxic and/or
`
`Originally received: March 17, 2004.
`Manuscript no. 240204.
`Accepted: June 16, 2004.
`1 Departments of Ophthalmology, Quinze-Vingts National Ophthalmology
`Hospital and Ambroise Pare Hospital, APHP, Paris-Ouest School of Med(cid:173)
`icine, University of Versailles, Paris, France.
`2 INSERM U598, Cordeliers Biomedical Institute, Paris, France.
`3 Department of Ophthalmology, University Hospital of Dijon, Burgundi
`University, Dijon, France.
`4 Department of Toxicology, Faculty of Biological and Pharmacological
`Sciences, University Paris 5 Rene Descartes, Paris, France.
`
`2186
`
`© 2004 by the American Academy of Ophthalmology
`Published by Elsevier Inc.
`
`Presented at: American Academy of Ophthalmology annual meeting, No(cid:173)
`vember 17, 2003; Anaheim, California.
`This study was supported by University Paris 6, Paris, France (grant no.:
`EA3123).
`Correspondence to Christophe Baudouin, MD, PhD, Quinze-Vingts Na(cid:173)
`tional Ophthalmology Hospital, 28 rue de Charenton, 75012 Paris, France.
`E-mail : baudouin @quinze-vingts.fr.
`
`ISSN 0161-6420/04/$-see front matter
`doi: I 0.10! 6/j.ophtha.2004.06.023
`
`
`
`Baudouin et al · Conjunctiva! Cytokines and Antiglaucoma Drugs
`
`Table l. Demographic Characteristics of Normal Subjects and the 3 Groups of Glaucoma Patients
`
`n (eyes)
`Age (yrs)
`Mean:±: SD
`Range
`Gender
`Male
`Female
`Type of treatment
`
`Preservative-Free
`Timolol
`
`24
`
`62.7 :±: 15.5
`45-84
`
`10 (42%)
`14 (58%)
`0.5% timolol
`
`Preserved
`Timolol
`
`27
`
`60.0 :±: 7.5
`46-72
`
`12 (44%)
`lJ (56%)
`0.5% timolol
`
`Multi treatments
`
`18
`
`56.6 :±: 16.9
`28-79
`
`10 (55%)
`8 (45%)
`0.5% timolol (n = 18)*
`Prostaglandin (n = 11)*
`CAI (n = 9)*
`a.-2 agonist (n = 5)*
`
`Normal
`Subjects
`
`15
`
`49.3 :±: 20.7
`37-66
`
`7 (46%)
`8 (54%)
`NA
`
`Total duration of treatment (yrs)
`Mean:±: SD
`Range
`
`4.7 :±: 2.5
`2-7
`
`4.6 :±: 2.8
`1-10
`
`5.9 :±: 6.6
`2- 24
`
`NA
`NA
`
`CAI = carbonic anhydrase inhibitor; NA = not applicable; SD = standard deviation.
`No statistical difference between each group for age and duration of treatment (Mann-Whitney U test).
`*Eleven patients had 2 medications; 7 received 3 different drugs.
`
`proinflammatory effects of antiglaucoma ophthalmic solu(cid:173)
`tions are still being debated. The most frequently used
`preservative, benzalkonium chloride, has widely demon(cid:173)
`toxic effects in laboratory, 13· 14 experimen(cid:173)
`strated its
`tal,9· 15· 16 and clinical studies. 1·17·18 As a quaternary ammo(cid:173)
`nium, this compound is much more than an excipient and
`has been shown to cause directly or has shown some evi(cid:173)
`dence of causing tear film instability,2 loss of goblet
`cells, 17·19 conjunctiva! squamous metaplasia and apopto(cid:173)
`sis,5·20 disruption of the corneal epithelium barrier,21 severe
`loss of endothelial cells if accidentally introduced ·in the
`anterior chamber,22 and even blood-aqueous barrier disrup(cid:173)
`tion in the early phase of pseudophakia,23 leading to the new
`concept of pseudophakic preservative maculopathy.24
`In a previous series of reports on the ocular surface, our
`group developed new objective and reliable techniques for
`exploring the conjunctival epithelium and assessing the
`inflammatory and apoptotic status of conjunctival cells by
`using
`flow cytometry
`in
`impression cytology speci(cid:173)
`mens.25-27 In a recent article, 12 we used this technique in
`patients who had received antiglaucoma drugs over an ex(cid:173)
`tended period and demonstrated for the first time in a
`case- control study that patients treated with preserved
`J3-blockers exhibited high levels of HLA DR class II anti(cid:173)
`gens and intercel\ular adhesion molecule (ICAM) l, as well
`as a dramatic decrease in MUC-5AC- expressing cells,
`whereas eyes receiving unpreserved drugs showed almost
`normal patterns of inflammation and goblet cell density.
`Little is known, however, about the inflammatory pathways
`and mediators involved in ocular surface inflammation re(cid:173)
`lated to toxic side effects of ophthalmic drugs and preser(cid:173)
`vatives. In the present case-control study, we used the flow
`cytometry technique to assess the expression of 3 major
`interleukins involved in inflammation and inflammatory cell
`recruitment: IL-6, IL-8, and IL- I 0. As the overexpression of
`these 3 interleukins has previously been described in differ(cid:173)
`ent ocular surface diseases,28-30 our aims were to validate
`the technique of flow cytometry in impression cytology for
`
`these intracellular markers, investigate interleukin expres(cid:173)
`sion by conjunctiva! epithelial cells in normal eyes and
`glaucoma patients, and compare their expression in differ(cid:173)
`ent groups of patients receiving preserved or unpreserved
`drugs.
`
`Patients and Methods
`
`Patients
`
`A total of 69 patients with glaucoma and 15 normal subjects were
`included in this case-control study conducted in compliance with
`the Declaration of Helsinki (Scotland amendment, 2000). All
`glaucoma patients had chronic primary open-angle glaucoma but
`no other ocular disease, as assessed after a complete ocular exam(cid:173)
`ination. They were treated ·for at least 1 year with the same
`treatment, 27 patients with preserved 0.5% timolol (containing
`0.0 I% benzalkonium chloride as a preservative), 24 patients with
`unpreserved 0.5% timolol, and the other 18 patients with an
`association of 2:2 preserved drugs (Table I). In the first 2 groups,
`no patient had received other antiglaucoma drugs before the ones
`investigated in this study or & concomitant treatment during the last
`6 months before the time of conjunctiva! imprint collection. There
`was no significant difference for age and total treatment durations
`between all groups of glaucoma patients and normal subjects. The
`multitherapy group had a longer though not significantly different
`history of glaucoma treatment and served as a control group with
`multiple and long-term glaucoma treatments. In this group, treat(cid:173)
`ments consisted of a preserved /3-blocker in all eyes associated
`with a prostaglandin analog (11 eyes), a carbonic anhydrase in(cid:173)
`hibitor (9 eyes), or an a-2 agonist (5 eyes), 11 eyes receiving 2
`drugs and 7 eyes 3 drugs. At the time of specimen collection,
`patients did not show any clinical evidence of intolerance to the
`administered drugs or ocular surface impairment. Patients with
`other diseases such as ocular or systemic inflammatory disorders
`or contact lens wear that might affect the conjunctiva! epithelium
`were excluded from the study. Fifteen normal subjects who had no
`history of ocular disease or clinical ophthalmic abnormality and
`who had not received any topical treatment for at least 6 months
`were also investigated, after approval by the ethics committee of
`
`2187
`
`
`
`Ophthalmology Volume 111, Number 12, December 2004
`
`Dijon University, France. The ethics committee ?fUniversity Paris
`6 had previously indicated that the explorat10n of the ocular
`surface of glaucoma patients using impression cytology collection
`did not require specific approval. Nevertheless, all patients had
`received specific explanations of impression cytology and gave
`informed consent for the procedure.
`
`Impression Cytology Specimens
`One eye was randomly selected for impression cytology analyses.
`Specimens were collected in the superior bulbar co~junctiva using
`0.20-µ.m polyether sulfone filters (13 mm m diameter; Supor
`Membranes, Gelman Sciences, Ann Arbor, MI), according to
`previously described techniques. 12·27 After instillation of I drop of
`0.04% oxybuprocaine, 2 pieces of filters, each 13 X 6.5 mm, were
`applied to the superior and superotemporal bulbar conjunctiva
`without the exertion of any pressure. Specimens were collected at
`least 15 minutes after the use of fluorescein eyedrops, to avoid
`interference with immunofluorescence analyses. Membranes were
`immediately suspended and fixed in cold phosphate-buffered sa(cid:173)
`line (PBS; pH 7.4, at 4° C) containing,0.05% paraformaldehyde.
`Conjunctiva! cells were further extracted by gentle agitation and
`then centrifuged at 1600 rpm for 5 minutes before processing for
`flow cytometry.
`
`lmmunofluorescence Procedure
`Four antibodies and 3 isotypic negative· controls were used in this
`study. To determine the inflammatory profile of impression cytol(cid:173)
`ogy specimens, indirect immunofluorescence was performed with
`mouse immunoglobulin GI (lgGI) anti-HLA DR a chain (clone
`T AL.1B5, 50 µ.g/ml, Dako SA, Copenhagen, Denmark) as the
`primary monoclonal antibody. The secondary antibody was fluo(cid:173)
`rescein isothiocyanate-conjugated goat antimouse immunoglobu(cid:173)
`lins (Dako). A nonimmune mouse IgGl (Dako) was used as a
`negative isotypic control. For intracellular interleukin assessment,
`after a first step of cell permeabilization with 1 % saponin for 5
`minutes (Sigma Chemical Co., St. Louis, MO), phycoerythrin(cid:173)
`conjugated anti-IL-6 mouse IgG2b, anti- lL-8 rat IgG2a, and anti-
`IL- IO rat lgG2a (BD Biosciences, San Diego, CA) were used in
`direct immunofluorescence procedures, with
`their respective
`mouse and rat isotypic negative controls (BD Biosciences). All
`antibodies were diluted in I% bovine serum albumin-containing
`PBS to obtain a final dilution of I :50, following the manufacturer's
`instructions. After 30 minutes of incubation with monoclonal
`antibodies, cell suspensions were washed in PBS, centrifuged for
`5 minutes (1600 rpm), and resuspended in 500 µ.I of PBS before
`flow cytometry analysis. For HLA DR assessment, cells were
`additionally incubated for 30 minutes with the secondary anti(cid:173)
`mouse immunoglobulins in a 1:50 dilution, centrifuged in PBS
`(1600 rpm, 5 minutes), and resuspended in 500 µ.l of PBS for flow
`cytometry processing, performed with an Epics-XL (Beckman
`Coulter, Miami, FL).
`
`.
`
`Flow Cytometry Analysis
`Cells were analyzed from a 2-parameter histogram showing side
`scatter (cell size) versus forward scatter (cell granulometry) in
`logarithmic and linear modes, respectively. For each antibody, at
`least 1000 conjunctiva] cells were gated to give a logarithmic
`fluorescence histogram showing the number of cells as a function
`of fluorescence intensities. A cursor was first set at the highest
`fluorescence level obtained for each isotypic negative control,
`which determined the limit of background fluorescence and the
`threshold of positivity for the tested antibodies. Results were
`obtained in percentages of positive cells and in mean fluorescence
`
`intensities. Mean fluorescence intensity levels of HLA DR and the
`3 interleukins tested were further quantified by the Quantitative
`Indirect Fluorescence Intensity (Dako) and Dakofluorosphere
`(Dako) methods, respectively, for indirect and direct immuno~u(cid:173)
`orescence techniques. Calibration curves were then obtained, giv(cid:173)
`ing mean fluorescence intensities versus the number of molecules
`of bound antibody, thus defining arbitrary units of fluorescence
`(AUFs). The actual number of AUFs for a specific marker was
`then obtained by subtraction of that found for the corresponding
`isotypic negative control. All flow cytometric analyses were ~er(cid:173)
`formed in a masked manner for treatment groups and pal!ent
`characteristics. Statistical comparisons were performed using the
`nonparametric Mann-Whitney U test. 3 1
`
`Results
`
`The mean percentage of HLA DR-positive conjunctiva! cells (Fig
`I A) was higher in both the preserved and the multi treatment
`groups (46.4% and 63.4%, respectively) than in the control group
`(13.1 %; P < 0.001 for both groups) and the eyes receiving preser(cid:173)
`vative-free timolol (19.3%; P :S0.03 for both groups). No differ(cid:173)
`ence was found between the control and preservative-free groups.
`Although the mean percentage of HLA DR- positive cells was
`higher in the multitreatment group than in the preservative group,
`the difference did not reach significance. The mean levels of HLA
`DR expression given in AUFs showed similar results (Fig lB),
`with values in the preservative treatment and multitreatment
`groups (47 784 and 77 406 AUFs, respectively) significantly
`higher than those in the control group (5119 AUFs; P< O.O I for
`both groups) and the eyes not treated with preservatives (29 480
`AUFs; P :S0.03 for both groups). There was also a significant
`difference between the control and preservative-free groups
`(P = 0.035).
`The result of IL-6 expression in percentage of positive cells
`(Fig 2A) showed a significant difference between the control group
`and the other 3 groups (P<0.001), without a significant difference
`between the 3 glaucoma groups. Similarly, IL-6 intracellular ex(cid:173)
`pression levels assessed in AUFs (Fig 2B) showed a highly sig(cid:173)
`nificant increase in the 3 glaucoma groups as compared with
`control normal eyes (P< 0.001), with no significant difference
`between the 3 groups receiving antiglaucoma treatments, despite a
`slight nonsignificant increase in the multitreatment group.
`The preservative-free, preservative-containing, and multitreat(cid:173)
`ment groups also had IL-8 expression significantly higher than that
`of the control group (Fig 3), both in percentage of positive cells
`(P< 0.001 for the 3 groups) and in levels of expression (P< 0.00 1
`for the 2 preservative-containing groups; P = 0.01 for the preser(cid:173)
`vative-free group). Arbitrary unit of fluorescence results, however,
`showed higher levels in the multitreatment group than in the
`preservative-free one (P = 0.03).
`The mean percentage of IL-JO- positive conjunctiva] cells (Fig
`4) was higher in the preservative and multitreatment groups than in
`the control group (P:S0.006 for both comparisons), but not in the
`preservative-free group. Levels of intracellular expression simi(cid:173)
`larly reached a significant difference between the 2 preservative(cid:173)
`containing groups and normal eyes (P< 0.05 for both groups).
`
`Discussion
`
`In previous studies on ocular surface immunopathology,
`some evidence has suggested that conjunctiva) epithelial
`cells may play an active role in ocular inflammation.28 - 30
`Conjunctiva] epithelial cells have therefore previously been
`
`2188
`
`
`
`Baudouin et al · Conjunctival Cytokines and Antiglaucoma Drugs
`
`80
`
`60
`
`l
`et: q
`~
`:i:: 40
`
`20
`
`0
`
`**§
`
`**§
`
`Treatment
`
`1A
`
`100000
`
`a Control
`
`• nonpreserved
`
`D Preserved
`
`(II Multitreatment
`
`80000
`
`ii:'
`
`::, ! 60000
`;
`
`:i:: 40000
`
`*
`
`20000
`
`0
`
`Ht!t!t~-l
`
`Treatment
`
`18
`
`a Control
`
`• nonpreserwd
`o preserved
`B Multitreatmen
`
`Figure 1. Results of flow cytometry in impression cytology specimens (means and standard errors). A, Percentage of HLA DR-positive cells in normal
`eyes and glaucomatous patients. §P:s0.03, compared with the preservative-free group. **P< 0.001 , compared with controls. B, Level of expression,
`expressed in arbitrary units of fluorescence (AUF). §P:s0.03, compared with the preservative-free group. *P = 0.035, compared with controls. **P<0.01,
`compared with controls.
`·
`
`shown to express immune-related markers29•32 and to be a
`possible source for proinflammatory cytokines.33•34 Normal
`conjunctival epithelial cells express mRNA for IL-1, IL-6,
`IL-8, tumor necrosis factor a (TNF-a), and regulated-on(cid:173)
`activation normal T-cell expressed and secreted (RANTES)
`protein.35•36 However, ICAM-1 and HLA DR are not ex(cid:173)
`pressed by normal conjunctival epithelial cells,26•27•36 but
`are induced to be expressed at high levels on conjunctival
`epithelial cells in inflammatory conditions.29•37 In the
`present study, we used a reliable and objective technique of
`flow cytometry in impression cytology to investigate the
`expression of 3 major interleukins by conjunctival cells. We
`also analyzed HLA DR expression as a hallmark for con(cid:173)
`junctival inflammatory status and confirmed the overexpres(cid:173)
`sion of this immune marker in glaucoma patients receiving
`long-term treatments, consistent with a previous similarly
`conducted case-control study. 12 Our results also confirmed
`that preserved drugs, especially when ~2 drugs are associ(cid:173)
`ated, induced significantly higher HLA DR expression than
`did unpreserved /3-blockers. Human leukocyte antigen DR
`expression was slightly more elevated in the preservative-
`
`free group than in the nontreated eyes, which suggests a low
`level of subclinical inflammation induced by topical treat(cid:173)
`ment, even though the active compound timolol was found
`in previous in vitro studies not or almost not to be toxic. 15
`The mechanisms by which topical treatments may induce
`HLA DR expression are not fully determined, but it is likely
`that in the ocular surface proapoptotic and proinflammatory
`cytokines, including interferon y and TNF-a, may stimulate
`common pathways.26
`The conjunctival profile of interleukin expression, how(cid:173)
`ever, differed from those of HLA DR and ICAM-1. 12 Ex(cid:173)
`cept for an increase in IL-S's level of expression (Fig 3B) in
`the preservative-containing groups as compared with the
`preservative-free group, IL-6, IL-8, and IL-10 were simi(cid:173)
`larly overexpressed in all glaucoma groups, whatever the
`type of treatment received. This would suggest that inter(cid:173)
`leukin expression by conjunctiva! cells is dependent upon
`immune mechanisms and pathways other than those influ(cid:173)
`encing adhesion molecules and class II antigens.
`As HLA DR, IL-6 is known to be secreted in response to
`TNF-a. Although IL-6 can be found in the normal eye, it is
`
`** **
`
`**
`
`2A
`
`20000 ~ - - - - - - - - - -~
`**
`
`•Control
`• nonpreseNed
`Cl preseNed
`a M.Jltitreatment
`
`ii:'
`::,
`~ 10000 +-- --
`~
`
`28
`
`a Control
`• norpreservad
`o preserved
`8 Multitreatmert
`
`~ t...
`cq
`::!
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Treatment
`Figure 2. Expression of interleukin 6 (IL-6) by conjunctiva! cells. A, Percentage of positive cells. B, Levels of intracellular expression. AUF = arbirrary
`units of fluorescence. **P<0.001, compared with controls.
`
`Treatment
`
`0
`
`2189
`
`
`
`Ophthalmology Volume 111 , Number 12, December 2004
`
`-
`
`::t
`~
`~
`::!
`
`120 -
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`**
`
`**
`
`**
`
`3A
`
`• Control
`liiil nonpreserved
`o preserved
`El M.Jltitreatment
`
`Treatment
`
`--- - -
`*
`
`40000
`
`30000
`
`ii:'
`:,
`~ 20000
`~
`::!
`
`10000
`
`0
`
`Treatment
`
`38
`
`**§
`
`• Control
`r,a nonpreserved
`
`D preserved
`CJ Multitreatment
`
`Figure 3. Expression of interleukin 8 (IL-8) in the conjunctiva! epithelium. A, Percentage of positi ,·e cells. B, Levels of intracellular expression . AUF
`= arbitrary units of fluorescence. *P< 0.01 , compared with controls. **P< 0.001, compared with controls. §P = 0.03, compared with the preservative-free
`group.
`
`generally considered to be an inflammatory cytokine. Jones
`et al35 and Pflugfelder et a1 38 observed a very high expres(cid:173)
`sion of mRNA encoding IL-6 and IL-8, as well as HLA DR,
`ICAM-1, TNF-a, IL-1 a, IL-1 {3, and transforming growth
`factor {31, in the conjunctiva) epithelium of Sjogren's syn(cid:173)
`drome eyes as compared with normal eyes using the reverse
`transcriptase-polymerase chain reaction method in impres(cid:173)
`sion cytology. Human leukocyte antigen DR and IL-6 gene
`expressions were also found at higher levels in Sjogren's
`syndrome conjunctiva) cells than in normaf conjunctiva.37
`Interleukin 6 and IL-8 have been shown to be present in
`normal tears39 and to be expressed abundantly in pterygium
`epithelium. 40 In this latter study, mRNAs and proteins were
`stimulated by ultraviolet radiation in a time- and dose(cid:173)
`dependent manner, thus showing the influence of external
`noninflammatory stimulations on these cytokines.40 Other
`models of ocular surface diseases, such as corneal and
`limbal debridement or vernal keratoconjunctivitis, demon(cid:173)
`strated an overexpression of proinflammatory cytokines and
`markers, including IL-6 or ICAM-1. 4 1.42
`In addition to IL-6, we found overexpression of 2 other
`
`major interleukins, IL-8 and lL-10. Interleukin 8 has pow(cid:173)
`erful chemotactic effects on T lymphocytes and neutrophils
`and may lead to inflammatory reactions. Little is known
`about IL-S's role in ocular surface di seases, but significantly
`increased levels were found in the conjunctiva) epithelium
`of Sjogren 's syndrome patients36 and in chronic allergic
`disorders/7 .4 3.44 as compared with normal eyes. Interleukin
`10, a cytokine of the Th2 profile, mainly acts as an inhibi(cid:173)
`tory factor to restrain proinflammatory cytokines produced
`by inflammatory cells.45 In ocular surface tissues, T cells
`found in atopic keratoconjunctivitis were shown to produce
`increased levels of IL-10 in comparison with those from
`vernal keratoconjunctivitis and giant papillary conjunctivitis
`groups.46
`In our study, not only the expression of HLA DR and
`inflammatory cytokines (IL-6, IL-8) but also expression of
`the theoretically inhibitory factor IL-10 were higher in
`preserved and multitreatment groups than in the control
`group composed of normal untreated eyes. Interleukin ex(cid:173)
`pression could be hypothesized as a global reaction to toxic
`drugs administered for long periods of time to the ocular
`
`4A
`
`100 ~ - - - -- -- -----
`
`80 -t-- - -- - - -;---;;;;;;l;;;=--1 ~ - - -- -~
`• Control
`lao +-----
`m nonpreserved
`o
`o preserved
`~ 40
`rn Multitreatment
`20
`
`0
`
`Treatment
`
`48
`
`-
`
`-+--
`
`I
`
`I
`• Control
`1111 nonpreserved
`i
`1--l:fmtn-l o preserved
`l!I Multitreatrnent
`
`8000 + - --
`
`ii:'
`:J
`~
`0 ....
`~ 6000 -+----
`
`-
`
`4000
`
`2000
`
`0
`
`Treatment
`
`Figure 4. Expression of interleukin JO (IL-JO) in the conjuncti va! epithelium. A, Percentage of positi ve cells. B, Levels of intracellular expression . AU F
`= arbitrary units of fluorescence. *P< 0.05, co mpared with normal eyes. **P:s0.006, compared with normal eyes.
`
`2190
`
`
`
`Baudouin et al · Conjunctival Cytokines and Antiglaucoma Drugs
`
`surface. However, the conjunctiva] profiles of HLA DR and
`the 3 interleukins seemed to differ, and further studies will
`be necessary better to understand the roles of immune
`markers and chemotactic, proinflammatory, or inhibitory
`cytokines in conjunctiva] immunopathology. Indeed, in
`vitro studies in conjunctiva] epithelial cells showed that
`release of proinflammatory cytokines may be stimulated by
`different mediators, as TNF-a and IL-1 {3 are potent stimu(cid:173)
`lators of IL-8, but not of ICAM-1 and HLA DR, whereas
`interferon y has the inverse profile. 32.47 Histamine also
`stimulates the secretion of IL-6 and IL-8 by conjunctiva]
`epithelial cells in a dose- and time-dependent manner,35
`which confirms the role of the conjunctiva! epithelium in
`inflammatory and/or allergic disorders and illustrates the
`complex pathways leading to cytokine expression in the
`ocular surface.
`An explanation for our results in glaucoma patients could
`therefore be that IL-6, IL-8 and IL- IO are highly sensitive to
`any proinflarnmatory stimulation, including the weakly toxic
`preservative-free {3-blockers, whereas HLA DR and ICAM-1
`may require higher levels of stimulation. Indeed, it has been
`shown that closed-eye tears contain high levels of IL-6 and
`IL-8 in comparison with open-eye tears,48 which may demon(cid:173)
`strate that these cytokines can be easily overexpressed in
`pathophysiological conditions. It is aiso possible, however, that
`these interleukins are released after inflammatory or toxic
`challenges in a soluble fonn in the tear flow and that the high
`levels of expression in the subcellular compartment do not
`reflect the actual amounts of cytokines present in tears and
`globally synthesized by conjunctiva] cells. Further studies
`combining cellular and tear concentrations would be of interest
`to try to discriminate between the different groups of glaucoma
`treatments, especially in differentiating preservative-free and
`preservative-containing topical drugs.
`Nevertheless, subclinical inflammation may raise further
`serious ocular surface issues in glaucoma patients, espe(cid:173)
`cially at the time of filtering surgery.9 Care should be taken
`to avoid preservatives in a long-term use as much as pos(cid:173)
`sible and to limit their concentration, as their toxicity is
`dose- and time-dependent, 15 or to develop new nontoxic
`preservatives49 capable of reducing allergic reactions and
`improving ocular surface tolerance.
`
`References
`
`I. Pisella PJ, Pouliquen P, Baudouin C. Prevalence of ocu lar
`symptoms and signs with preserved and preservative-free
`glaucoma medication. Br J Ophthalmol 2002;86:418-23.
`2. Ishibashi T, Yokoi N, Kinoshita S. Comparison of the short(cid:173)
`term effects on the human corneal surface of topical timolol
`maleate with and without benzalkonium chloride. J Glaucoma
`2003;12:486-90.
`3. Broadway DC, Grierson I, Hitchings R. Adverse effects of
`topical antiglaucomatous medications on the conjunctiva. Br J
`Ophthalmol l 993;77:590-6.
`4. Schwab IR, Linberg JV, Gioia VM, et al. Foreshortening of
`the inferior conjunctiva! fornix associated with chronic glau(cid:173)
`coma medications. Ophthalmology 1992;99: 197-202.
`5. Dogan AL, Orhan M, Soylemezoglu F, et al. Effects of topical
`antiglaucoma drugs on apoptosis rates of conjunctiva! epithe-
`
`lial cells in glaucoma patients. Clin Experiment Ophthalmol
`2004;32:62-6.
`6. Herreras JM, Pastor JC, Calonge M, Asensio VM. Ocular
`surface alteration after long-term treatment with an antiglau(cid:173)
`comatous drug. Ophthalmology 1992;99: 1082-8.
`7. Broadway DC, Grierson I, O'Brien C, Hitchings RA. Adverse
`effects of topical antiglaucomatous medication. IL The out(cid:173)
`come of filtration surgery. Arch Ophthalmol 1994;112:
`1446-54.
`8. Yalvac IS, Gedikoglu G, Karagoz Y, et al. Effects of antiglau(cid:173)
`coma drugs on ocular surface. Acta Ophthalmol Scand 1995;
`73:246-8.
`9. Mietz H, Niesen U, Krieglstein GK. The effect of preserva(cid:173)
`tives and antiglaucomatous medication on histopathology of
`the conjunctiva. Graefes Arch Clin Exp Ophthalmol 1994;
`232:561-5.
`10. Baudouin C, Pisella PJ, Goldschild M, et al. Ocular surface
`inflammatory changes induced by topical antiglaucoma drugs.
`Human
`and animal
`studies. Ophthalmology 1999; I 06:
`556-63