Effects of triamcinolone acetonide on vessels of the posteriorsegment of the eye

Fatemeh Valamanesh1234 Marianne Berdugo1234 Florian Sennlaub123 Michegravele Savoldelli5Cyndie Goumeaux123 Marianne Houssier123 Jean-Claude Jeanny123 Alicia Torriglia123

Francine Behar-Cohen12345

1INSERM UMRS 872 Physiopathology of ocular diseases Therapeutic innovations Centre de Recherche des Cordeliers 15 ruede LrsquoEcole de Meacutedecine 75006 Paris France 2Universiteacute Paris Descartes Paris France 3Universiteacute Pierre et Marie Curie ParisFrance 4Laboratoire drsquoInnovations Theacuterapeutiques Fondation Rothschild 48 rue Mathurin Moreau 75019 Paris France5Assistance Publique-Hocircpitaux de Paris Hocirctel-Dieu de Paris Department of Ophthalmology 1 place du Parvis Notre-Dame 75001Paris France

Purpose This study investigates the effects of triamcinolone acetonide (TA) on retinal endothelial cells in vitro andexplores the potential vascular toxic effect of TA injected into the vitreous cavity of rats in vivoMethods Subconfluent endothelial cells were treated with either 01 mgml or 1 mgml TA in 1 ethanol Control cellswere either untreated or exposed to 1 ethanol Cell viability was evaluated at 24 h 72 h and five days using thetetrazolium 3-(45-dimethylthiazol-2-yl)-25 phenyltetrazolium bromide test (MTT) and lactate dehydrogenase (LDH)assays Cell proliferation was evaluated by 5-bromo-2-deoxyuridine (BrdU) test Apoptosis was evaluated by terminaldeoxynucleotidyl transferase dUTP nick end labeling assay (TUNEL assay) annexin-binding and caspase 3 activationCaspasendashindependent cell deaths were investigated by immunohistochemistry using antibodies against apoptosis inducingfactor (AIF) cytochrome C microtubule-associated protein (MAP)-light chain 3 (MAP-LC3) and Leukocyte ElastaseInhibitorLeukocyte Elastase Inhibitor-derived DNase II (LEIL-DNase II) In vivo semithin and ultrathin structureanalysis and vascular casts were performed to examine TA-induced changes of the choroidal vasculature In additionouter segments phagocytosis assay on primary retinal pigment epithelium (RPE) cells was performed to assesscyclooxygenase (COX-2) and vascular endothelial growth factor (VEGF) mRNAs upregulation with or without TAResults The inhibitory effect of TA on cell proliferation could not explain the significant reduction in cell viabilityIndeed TA induced a time-dependent reduction of bovine retinal endothelial cells viability Annexin-binding positivecells were observed Cytochrome C was not released from mitochondria L-DNase II was found translocated to the nucleusmeaning that LEI was changed into L-DNase II AIF was found nuclearized in some cells LC3 labeling showed theabsence of autophagic vesicles No autophagy or caspase dependent apoptosis was identified At 1 mgml TA inducednecrosis while exposure to lower concentrations for 3 to 5 days induced caspase independent apoptosis involving AIF andLEIL-DNase II In vivo semithin and ultrathin structure analysis and vascular casts revealed that TA mostly affected thechoroidal vasculature with a reduction of choroidal thickness and increased the avascular areas of the choriocapillariesExperiments performed on primary RPE cells showed that TA downregulates the basal expression of COX-2 and VEGFand inhibits the outer segments (OS)-dependent COX-2 induction but not the OS-dependent VEGF inductionConclusions This study demonstrates for the first time that glucocorticoids exert direct toxic effect on endothelial cellsthrough caspase-independent cell death mechanisms The choroidal changes observed after TA intravitreous injectionmay have important implications regarding the safety profile of TA use in human eyes

Glucocorticoids are commonly used in the treatment ofocular pathologies associated with vascular leakage vesselsabnormalities and ocular neovascularization Intravitreousinjections of Triamcinolone Acetonide (TA) are currentlyused for the treatment of macular edema [12] They induce adramatic reduction of macular thickness but this isinconsistently correlated with long-term functional recoveryWhen used for the treatment of choroidal neovascularizations

Correspondence to Francine Behar-Cohen INSERM UMRS 872team 17 Centre de Recherche des Cordeliers 15 rue de lrsquoEcole deMeacutedecine 75006 Paris France Phone +33 1 40 46 78 46 FAX +331 40 46 78 55 email francinebehargmailcom

associated with Age-related Macular Degeneration (AMD) orother ocular neovascularisations TA decreases the vascularleakage on the short-term [3] However its long-term effecton visual acuity remains controversial [4-6] For the treatmentof retinal neovascularization TA has been used mostly incombination with other antivascular strategies [78] Oncapillary eyelid hemangiomas glucocorticoids seem to havebeneficial effects through a reduction in the size of thevascular lesion [9]

The vascular effects of TA have been confirmed in animalmodels of choroidal (CNV) and retinal neovascularization Inthe rat model of laser-induced CNV TA not only decreasedfluorescein angiography (FA) leakage [1011] but it also

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gtReceived 28 May 2009 | Accepted 21 November 2009 | Published 8 December 2009

copy 2009 Molecular Vision

2634

reduced the CNV membrane diameter and thickness [1112]and transiently inhibited CNV formation [13] In the rat modelof retinopathy of prematurity TA reduced the retinalneovascularization in a dose-dependent manner [14-17] andreduced the vascular leakage [16]

In vitro TA has been shown to downregulate theexpression of tumor necrosis factor α vascular endothelialgrowth factor (VEGF) interleukin 1 and matrixmetalloproteinases and subsequently influenceneovascularization [16] TA has also been reported to havedirect effects on endothelial cells through an antiproliferativeeffect [17] and a reduction in the interferon-inducedpermeability of human choroidal endothelial cells [18]

Beside their known effects on vascular permeability thein vivo antiangiogenicangiostatic activity of corticosteroidsresults from many factors including their anti-inflammatoryactivities However it remains unknown if TA can directlyinduce vascular endothelial cell death In this paper weinvestigated the mechanisms of the effects of TA on retinalendothelial cells in vitro and in the rat retinal and choroidalvascularization in vivo

METHODSReagents All culture reagents were obtained from Gibco(New York NY) Corticosteroids were purchased from Sigma(Saint-Quentin-Fallavier France) except for Kenacort Retardwhich was purchased from Bristol-Myers Squibb (ParisFrance)Culture of bovine retinal endothelial cells and treatmentsBovine retinal endothelial cells (BRECs) were seeded inDulbeccorsquos modified Minimal Essential Medium (DMEM)glutamax supplemented with 10 calf serum and 20 ng ofVEGF (Sigma St Louis MO) at 37 degC in a humidifiedatmosphere containing 5 CO2 After 24 h the cells weretreated for either 24 72 h or 5 days with 01 or 1 mgml TAthat had been dissolved in 1 ethanol (TA-E 999 ethylalcohol VWR Paris France) Control cells were either nottreated or treated with 1 ethanol in medium Beforetreatment cells were maintained in medium supplementedwith 2 of fetal calf serum (FCS) for 24 h For treatment cellswere cultured in medium supplemented with 2 FCS withoutVEGF In some experiments cells were treated afterconfluence was reachedAnalysis of TA effects on cultured cells

Cell viability evaluation using MTT testmdashViabilitywas assessed by the tetrazolium 3-(45-dimethylthiazol-2-yl)-25 phenyltetrazolium bromide test (MTT SigmaChemical) using 2105 cells per well in 24-well plates Brieflyculture medium was removed and 250 microl of MTT (1 mgmlin 1times PBS 137 mM NaCl 27 mM KCl 10 mM Na2HPO42 mM KH2PO4 pH 74) were added to each well The wellswere then incubated for 1 h at 37 degC lysed with 250 microl ofisopropanol and assessed by measuring absorption at 570 nm

versus 630 nm using a microplate reader (BioRad San DiegoCA) Using a standard curve for each experiment wecorrelated the color intensity observed in each well to thenumber of viable cells assessed by the trypan blue exclusionassay

Evaluation of TA-induced viability reduction lactatedehydrogenase cytotoxicity assay In 24-well plates 2105

BRECswell were treated with TA-E for 24 h 72 h or fivedays Released lactate dehydrogenase (LDH) in culturesupernatants was measured with a cytotoxicity detection kit(Roche Diagnostics Mannheim Germany) Dye absorbancewas measured at 490 nm versus 630 nm using a standardmicroplate reader Positive control was obtained by thecomplete lysis of control wells with 02 Triton X100Because treatment of BRECs with TA-E at 1 mgml inducedmostly necrosis resulting from the direct contact of theunsolved crystals with the cells all further analysis were onlyperformed with TA-E at 01 mgmlEvaluation of cell proliferation measuring DNA synthesisrate For proliferation assay TA was used at 01 mgml and05 μgml BRECs (105 cellswell) were seeded in 24-well cellculture plates and allowed to attach for 24 h Thereafter thecells were incubated with TA-E for 24 h and 5-bromo-2-deoxyuridine (BrdU) was added into the medium for a finalconcentration of 10 microM and allowed to incubate for 18 h Theincorporation of BrdU into the genomic DNA was determinedusing the ELISA BrdU labeling and detection Kit III (RocheMolecular Biochemicals Germany)

For other experiments BRECs were seeded at 5x104

cellswell in the Lab-Tek chambered Coverglass and grownat 37 degC in a humidified atmosphere containing 5 CO2 and95 air for 24 h They were then treated with 01 mgml TA-E which is a concentration that allows further analysisControl cells were run using the culture medium alone or theculture medium containing 1 ethanol All experiments wereperformed in triplicates and repeated twice The Lab-Tekslides were mounted in Gel Mount (Biomeda BurlingameCA) before they were examined and photographed using anOlympus IX70 fluorescent microscope coupled to a digitalcamera The following excitationemission wavelengths wereused 365ndash420 nm 470510 nm 510560 nmDetection of apoptosis using terminal deoxynucleotidyltransferase dUTP nick end labeling assay After treatmentwith 01 mgml TA-E for 24 h 72 h or five days the cellswere rinsed air-dried and fixed with 4 paraformaldehyde(Merck Eurolab Fontenay sous-bois France) one hour atroom temperature The cells were rinsed during 5 min in PBSand permeabilized with 03 Triton X-100 in PBS for 20 minCells were incubated for 30 min with 50 microl of terminaldeoxynucleotidyl transferase dUTP nick end labeling assay(TUNEL Roche Diagnostics Mannheim Germany) reactionmixture at 37 degC The cells were then rinsed once with PBSstained with DAPI (Sigma) and rinsed with PBS five times

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2635

for 5 min each time Positive controls were obtained byinducing apoptosis with 1 microM staurosporine

Annexin V binding Annexin V binding was assayed with theAnnexin V Apoptosis detection kit (Santa CruzBiotechnology Inc Santa Cruz CA) according to themanufacturer instructions Briefly after treatment cells wererinsed with PBS washed once for 5 min with 500 microl per wellof assay buffer and incubated 15 min at room temperature with2 microgml of Fluorescein isothiocyanate (FITC)ndashlabeledAnnexin V Cells were then rinsed with PBS fixed with 4paraformaldehyde rinsed with PBS stained with DAPI andrinsed five more times for 5 min with PBS

Western blot analysis Total protein extracts were obtained bylysis of BRECs in Laemmli sample buffer Extracted proteinswere separated by SDSndashPAGE immobilized onnitrocellulose membrane (Millipore Billerica MA) andblotted with affinity purified rabbit anti-active Caspase-3

polyclonal antibody in a 1500 dilution (Calbiochem SanDiego CA) The secondary goat anti-rabbit IgG (VectorLaboratories Burlingame CA) was used in a 15000 dilutionThe amount of total protein extracts analyzed was 30 μglanePositive controls for caspase-3 activation were run on HL-60cells cultured in RPMI 1640 supplemented with 10 Fetalcalf serum treated for 24 h with 50 μM etoposide [19]Immunocytochemistry After the different treatments BRECcells were stained with the following antibodies anti-cytochrome C [20] (Sigma Steinheim Germany) diluted1250 in PBS containing 1 BSA (BSA 98 SigmaSteinheim Germany) anti-apoptosis-inducing factor (AIF)[21] (Sigma Saint-Quentin Fallavier France) diluted 1100in PBS containing 1 BSA anti-microtubules-associatedprotein and light chain 3 (MAP-LC3) [22] (Santa CruzBiotechnology) diluted 150 in PBS containing 1 fat-freemilk and polyclonal anti- L -DNase II antibody [2324]

TABLE 1 PRIMERS USED IN QPCR

Name Sequence (5prime-3prime)18S sense TGCAATTATTCCCCATGAACG18S antisense GCTTATGACCCGCACTTACTGGVEGF sense TGGGATGGTCCTTGCCTCVEGF antisense TCGCTGGAGTACACG GTGGTCOX2 sense TGCTACCATCTGGCTTCGGGAGCOX2 antisense ACCCCTCAGGTGTTGCACGT

Figure 1 Effect of triamcinoloneacetonide on BRECs viability andproliferation Cell viability wasevaluated using an MTT assaySubconfluent (A) or confluent BRECs(B) were exposed to 01 or 1 mgml TAin 1 ethanol for 24 h (dark-graycolumns) 72 h (red columns) or fivedays (white columns) Control cellswere either exposed to 1 ethanol (C-E) or were left untreated (C) C Cellproliferation was evaluated using BrdUlabeling detection after 24 h of treatmentwith 01 mgml or 05 microgml TAControl cells were either left untreated(C) or were treated with 1 ethanol(C-E) Results are expressed as meanplusmnstandard error with plt005 for allcolumns versus control Four animalswere used in each experiment DAPIstaining of untreated control cells areshown in (D) 01 mgml TA-treatedcells in (E) and 1 mgml TA-treated cellsin (F) all of them after five days oftreatment Scale bar represents 50 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2636

diluted 1100 in PBS containing 1 skimmed milk Afterincubation with the specific primary antibodies the cells wererinsed three times for 5 min with PBS and incubated for 1 hwith goat Alexa Fluor anti-rabbit IgG (Molecular ProbeInvitrogen NewYork NY) or rabbit anti goat Texas red dye-conjugated (Jackson Immuno Research Laboratories IncSuffolk UK) diluted 1250 in PBS containing 1 BSA PBScontaining 1 BSA or 1 skimmed milk was used instead ofthe primary antibody as negative controls

For autophagic positive controls BRECs were culturedfor 24 h in amino acid-depleted medium eg Hanks BufferedSalt Saline (HBSS Gibco New York NY) complementedwith 20 mM glucose [25] Slides were mounted in Gel Mountand examined with a fluorescent microscope Olympus IX70coupled to a digital camera The following excitation

emission wavelengths were used 365ndash420 nm 470510 nm510560 nm

Semithin and ultrathin structure analysis Lewis rats used inthis study were obtained from Roger Janvier Le Genest SaintIsle France The use of animals adhered to the ARVOstatement for Ophthalmic and Vision Research and protocolswere approved by the ethical committee of Reneacute DescartesUniversity of Paris Lewis rats (6-week-old weighing 150 g)were intravitreally injected with either 10 μl of PBS (n=3) orwith 10 μl of PBS containing 40 μg of TA (n=3) Thirty daysafter injection the rats were euthanized with a CO2 overdoseand their eyes processed for semithin and transmissionelectron microscopy analysis The eyes were fixed in 25glutaraldehyde cacodylate buffer (Na 01 M pH 74) for 1 hThey were then dissected and the posterior segments werefixed for 3 h Specimens were fixed in 1 osmium tetroxyde

TABLE 2 LDH RELEASE

of cytotoxicity of cytotoxicity of cytotoxicityTreatment 24 h p 72 h p 5 days pTA 1 mgml 237 lt005 234 lt005 243 lt005

TA 01 mgml 36 lt005 424 lt005 51 lt005C-E 0 0 0 C 0 0 0

Triton 02 100 100 100 Measure of LDH release in BRECs after 24 h 72 h and 5 days of treatment with 1 mgml or 01 mgml of TA Control cellswere either not treated (C) 1 ethanol-treated (C-E ) or 02 triton-treated No LDH release was found in control cells and asmall increase is seen in cells treated with 01mgml of TA The use of 1 mgml of TA drastically increase LDH release

Figure 2 Absence of TUNEL-positiveapoptosis in BRECs treated withtriamcinolone acetonide Control cellswere treated with the TA vehicle1ethanol (A-C) No TUNEL-positivecells could be observed after exposureto 01 mgml TA for 72 h (D-F) or fivedays (D5 G-I) As a positive control forthe TUNEL technique BRECs weretreated with 1 μM staurosporine (J-L)Nuclei were stained with DAPI (A DG J) Scale bar represents 60 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2637

in cacodylate buffer (Na 01 M pH 74) and progressivelydehydrated in graduated ethanol solution (50 70 95 and100) Each posterior segment was separated in two samplesincluded in epoxy resin and oriented Sections were cut to1 microm thickness with an ultra microtome Reichert Ultracut(Leica Biel Switzerland) and stained with toluidine blueSections cut to 80 nm thick were stained by uranyl acetate andlead citrate and analyzed with a Philips CM10 electronmicroscopeVascular corrosion casts Lewis rats (12 8-week-oldweighing 150 g) were intravitreally injected with either 10 microlof PBS (n=6) or with 10 microl PBS containing 40 microg of TA (n=6)After 30 days the animals were sacrificed by CO2 inhalationfollowed by a thoracotomy A catheter was introduced intothe aorta through the left heart ventricle and the right auriclewas cut to allow evacuation of injected products A perfusionwas performed with a mixture of red Mercox resin and catalyst

(Ladd Research Williston VT) Eyes were extracted andlenses were removed Tissues were conserved overnight at37 degC in PBS to allow complete polymerization and thendigested by 5 KOH for two weeks at 37 degC until only thevascular corrosion casts remained Distilled water was usedto remove salt and the mold was dried Only corrosion castswith completely filled iris vessels were used This was toexclude corrosion casts from incomplete perfusion Retinalvasculature was removed with forceps The specimens weremounted on Stretched Specimen Mount (SSM) stubs coatedwith gold palladium and scanned at an accelerating voltageof 117 kV To measure the thickness of the choriocapillarylumen we cut corrosion casts paracentrally (1 mm from theaperture of the optic nerve) and positioned for perpendicularviews of the choriocapillaries To analyze the intercapillaryspace (avascular area) we positioned the casts for frontalviews of the choriocapillaries Electron micrographs were

Figure 3 Annexin V binding of TA treated BRECs Annexin-V binding was performed in control cells treated with 1 ethanol (the TAvehicle A-C) or with 01 TA for 72 h (D-F) or five days (G-I) Condensed nuclei are seen in TA treated cells (D and G squares) as wellas Annexin-V positive cells (E and H squares) Arrows indicates Annexin-V positive cells Scale bar represents 50 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2638

scanned and analyzed using Image J Software The avascularareas were measured on frontal views and expressed as thepercentage of intercapillary surface (space between the plasticcapillary casts) of the whole area (n=6 rats five picturescast)Effect of TA on VEGF and COX-2 expression in primaryretinal pigment epithelium (RPE) cells

Rat RPE primary culturemdashTen-day-old Wistar ratpups were sacrificed by CO2 inhalation and the eyes wereenucleated Eyes were maintained at room temperatureovernight in DMEM (Invitrogen Cergy Pontoise France)then incubated 45 min with 2 mgml trypsincollagenase I at37 degC After trypsin inhibition with DMEM containing 10FCS the RPE layer was harvested The RPE was plated in 12-well plates at a rate of one RPE from one eye per well inDMEM containing 10 FCS 1 penicillinstreptomycinand 02 fungizone Cells were maintained for 12 days beforethe phagocytosis assay [26]Outer segments isolation and phagocytosis assay Outersegments (OS) were isolated following established protocols[27] Briefly 20 pig retinas were dissected and homogenizedin 20 sucrose buffer 20 mM Tris 2 mM MgCl2 and013 mM NaCl (pH 72) Retina samples were centrifuged ona sucrose gradient (50 27) 1h at 100000times g and OS wereharvested at the ring interface and diluted in DMEMCentrifugation for 10 min at 10000times g was performed andthe pellet was resuspended in DMEM to obtain a stocksolution of 108 OSml

Before treatment RPE cells were maintained in mediumsupplemented with 2 of FCS for 24 h Confluent rat primaryRPE cells were incubated with 01 mgml TA-E Control cellswere run using the culture medium alone After 24 h oftreatment the cells were rinsed once with PBS and incubatedwith 150 μl of 108 outer segmentsml After 1 h 850 microl ofcomplete medium was added After 6 h cells were washedand 350 μl of guanidine thiocyanate lysis buffer (RNeasy minikit Qiagen Courtaboeuf France) was added for RNAextractionReverse transcription and real-time polymerase chainreaction Total RNA was isolated with RNeasy plus Mini Kit(Qiagen Courtaboeuf France) Single-stranded cDNA was

synthesized from total RNA using random primer andsuperscript reverse transcriptase (Invitrogen) Subsequentreal-time polymerase chain reaction (RTndashPCR) wasperformed using cDNA qPCR SuperMix-UDG PlatinumSYBR Green (Invitrogen) and the and 05 pmolmicrol of theprimers described on Table 1

PCR reactions were performed in 40 cycles of 15 s at95 degC 45 s at 60 degC No product was obtained in controlreactions where reverse transcriptase was omittedStatistical analysis Data are represented as meanplusmnstandarderror of the mean (SEM) and compared using thenonparametric MannndashWhitney U test P values werecalculated for a confidence interval of 95 and p values ofless than 005 were considered significant

RESULTSEffect of TA on BREC viability The percentage of living cellswas reduced to 75plusmn11 plt005 after exposure to 01 mgmlof TA for 24 h This survival was reduce to 52plusmn29 plt005as concentration of TA increased to 1 mgml The rate ofsurvival was further reduced as exposure time increasedbeing of 58plusmn23 and 50plusmn37 for 01 mgml and 1mgmlof TA respectively after 72 h and of 55plusmn2 and 39plusmn09plt005 respectively for the same concentrations after 5 daysof treatment TA therefore decreased the number of livingcells as compared to control cells with an increased effectupon exposure duration Photomicrographs of DAPI-stainedBREC nuclei in culture showed that after 72 h (not shown) orfive days (Figure 1D-F) of treatment with 01 mgml TA-Ethe cells were not only reduced in number (Figure 1D) butalso organized in a ramified manner (Figure 1E) Aftertreatment with 1 mgml for 72 h or five days (Figure 1F) thenumber of cells was reduced and many cellular debris werepresent in the culture medium To evaluate whether thereduction in the number of living cells resulted from areduction in proliferation or in TA-induced cell toxicity weperformed the same experiments on confluent cells As shownon Figure 1B TA induced a similar reduction in the numberof living cells this effect increased upon the duration ofexposure and the concentration suggesting that the reduction

Figure 4 Caspase 3 western-blotanalysis As a positive control HL-60cells were treated with etoposide (Eto)The cleaved active form of caspase 3 isseen Expression of pro-caspase 3 isseen en BRECs in all conditions Noexpression of activated caspase 3 wasdetected in any of the treated BRECsdemonstrating that triamcinoloneacetonide (TA) did not induce caspase 3activation

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2639

in the number of living cells did not result solely from aninhibition of cell proliferationInhibition of BREC proliferation To further characterize theeffect of TA on endothelial cell proliferation we evaluatedthe DNA synthesis rate using the BrdU test As shown inFigure 1C 01 mgml TA-E significantly reduced theproliferation rate of BRECs from OD=206plusmn009 in controlcells to OD=162plusmn002 n=4 plt005 in TA-E treated cellsThis effect was not observed with a lower concentration ofTA-E 05 microgml (OD=18plusmn003)Evaluation of retinal endothelial cell death after exposure toTA

Necrosis in BREC exposed to TA Measurement ofLDH release The reduction in the number of living cells couldnot be explained only by a decrease of proliferationTherefore cell death mechanisms were explored We firstmeasured the amount of free LDH as a marker of passive celldeath (eg necrosis Table 2) LDH is a cytoplasmic enzymenot released during active cell death like apoptosis paraptosis

or autophagy but released in necrosis due to the early increaseof plasma membrane permeabilization No necrosis wasobserved in control cells exposure of BRECs to 1 mgml TA-E induced necrosis in 237 of the cells after 24 h of exposurein 234 after 72 h and in 243 of the cells after five days(plt005) After exposure to 01 mgml TA-E a significantlylower release of LDH than before was observed reaching36 after 24 h 42 after 72 h and 51 after five days oftreatment (plt005) Taking into account the inhibition ofproliferation and the percentage of necrosis occurring in01 mgml TA-E treated cells it is possible there is anothertype of cell death that could possibly explain the reducednumber of living cells after 72 h and five days

When 1 mgml TA was dissolved in 1 ethanol crystalsremained undissolved and stayed adherent to the cellmembranes resulting in the direct cytotoxic observed effectsTherefore for further analysis only cells treated with 01 mgml TA-E were used Since the TA effect on cell viability was

Figure 5 Cytochrome C release in TA treated BRECs Absence of cytochrome C release in triamcinolone acetonide (TA)-treated BRECsControl cells are shown in A to C TA treated cells for 72 h are shown in D to F and for 5 days in G to I No cytochrome C release is seen inthese cells Compare to cells treated with staurosporin (Stp) used as positive control (J-L arrow in K and L) Nuclei were stained with DAPI(A D G J) Scale bar represents 50 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2640

increased with the time of treatment cell death was analyzedafter 72 h and five days of exposureCaspase-dependent apoptosis Caspase-dependent apoptosiswas evaluated by TUNEL assay annexin-V binding caspase3 activation and cytochrome C release As shown in Figure2 no TUNEL-positive cells were detected after treatment with1 ethanol (Figure 2A-C) In addition no TUNEL-positivecells could be detected in cells treated with 01 mgml TA-Eafter 72 h (Figure 2D-F) or even after five days of exposure(Figure 2G-I) However when cells were exposed to 1 microMstaurosporine a known inducer of caspase-dependent celldeath TUNEL-positive cells with condensed nuclei wereobserved (Figure 2J-L)

During the early phases of any type of apoptosisphosphatidyl serine translocates to the external surfacemembrane of the cell [28] and binds annexin-V [29] Asshown in Figure 3 while no annexin V was found to bind onthe surface of control cells (Figure 3A-C) positive stainedcells were detected after 72 h and five days of exposure with01 mgml TA-E (Figure 3D-FG-I) After 72 h of exposure to

TA annexin-V positive cells displayed condensed orfragmented nuclei (Figure 3D-F insets) while after five dayssome cells displaying normal (uncondensed) nuclei alsostarted annexin-V binding (Figure 3G-I insets)

The results discussed in the previous section suggestedthe activation of apoptosis with no activation of caspase Toassess this point we performed western blot analysis ofcaspase-3 activation The results obtained revealed thepresence of pro-caspase-3 and the absence of activatedcaspase-3 in BRECs treated during 72 h or five days witheither nothing 1 ethanol medium or in 01 mgml TA(Figure 4) In positive control HL-60 cells treated withetoposide pro-caspase-3 as well as activated caspase-3 weredetected (Figure 4) These experiments showed that anoncaspase dependent form of apoptosis could intervene

After a toxic stress caspase 3 is in most of the casesactivated though the intrinsic pathway in which the releaseof cytochorme C from the mitochondria is a key event Weinvestigated then if cytochrome C was released In BRECstreated during 72 h or five days with 01 mgml of TA-E

Figure 6 L-DNase II activation in TA treated BRECs LEIL-DNase II immuno-labeling was performed in control cells treated with the TAvehicle (1 ethanol) A-C A cytoplasmic labeling is seen In cells treated with 01 mgml TA for 72h (D-F) or 5 days (G-I) cells withcondensed nuclei (arrows) present a nuclear staining for LEIL-DNase II indicating L-DNase II activation Nuclei were stained with DAPI(A D G) Scale bar represents 50 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2641

(Figure 5D-FG-I) no release of cytochrome C frommitochondria into the cytoplasm could be observed Thefluorescent labeling appeared indeed located in mitochondriaand no diffuse or nuclear fluorescence was present indicatingthat mitochondria have not released cytochrome C Incontrast in BRECs treated with staurosporine cytochrome Cwas located in condensed nuclei of apoptotic cells (Figure 5J-L arrows)Caspase-independent apoptosis in triamcinolone acetonide-induced toxicity As the cell death induced by TA seemed tobe caspase independent we analyzed two markers of caspase-independent apoptosis in TA-treated BRECs leukocyteelastase inhibitor L-DNase II (LEIL-DNase II) and apoptosisinducing factor (AIF)

LEI and L-DNase II are two forms of one protein that arerecognized by the same antibody However while LEI is acytoplasmic anti-protease its nuclear translocation parallelsits change into L-DNase II activity [30] While LEIL-DNaseII fluorescent labeling was present in the cytoplasm of thecontrol cells (Figure 6A-C) surrounding non-fluorescent

nuclear shapes a clear nuclear staining was observed in cellstreated with 01 mgml TA for 72 h or for five days (Figure6D-FG-I arrows) Nuclei positive for L-DNase II werecondensed or fragmented indicating a caspase-independentapoptotic pathway taking place in BRECs cells treated withTA

In control cells (Figure 7A-C) anti-AIF showed acytoplasmic labeling corresponding to the normallocalization of AIF at the mitochondria After 72 h of TA-Etreatment no change in AIF localization was seen (Figure 7D-F) However after exposure to 01 mgml TA-E for 5 dayssome cells with condensed or fragmented nuclei showedpositive nuclear staining This suggested that caspase-independent apoptotic pathways were activated in TA-treatedcells (Figure 7G-I arrows)Autophagy After activation of autophagy MAP-LC3concentrated in autophagic vesicles giving a dotted pattern tothe cells After 72 h (not shown) or five days of treatment with01 mgml TA-E (Figure 8D-F) no MAP-LC3-positivevesicles could be observedmdashmeaning that autophagy was not

Figure 7 Apoptosis Inducing Factor localization in triamcinolone acetonide treated BRECs Apoptosis inducing factor (AIF) immuno-labelingwas performed in control cells treated with the TA vehicle (1 ethanol A-C) A cytoplasmic labeling is seen This is also the case in cellstreated with 01 mgml TA for 72h (D-F) In cells treated with TA for 5 days (G-I) cells with condensed nuclei (arrows in H and I) presenta nuclear staining for AIF Nuclei were stained with DAPI (A D G) Scale bar represents 50 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2642

taking place under these conditions In cells cultured in aminoacid-deprived medium a condition known to induceauthophagy cells showed large intracytoplasmic vesicles thatstained positive with anti-MAP-LC3 antibody (Figure 8G-I)Structural analysis of rat retinal and choroidal vasculatureone month after TA exposure As demonstrated the resultsobtained indicated that TA could affect endothelial cellviability We investigated this hypothesis in vivo Semithinanalysis of the retina did not show evidence of alteration ofthe retinal vessels or capillaries but did show that a reductionof the diameter of the superficial capillaries was evident onsections of TA-treated retinas (Figure 9B) versus control ones(Figure 9A) More striking was the significant thinning of thechoriocapillaries and choroid in the TA-treated rats (Figure9D) as compared to PBS-injected rats (Figure 9 C and E)Moreover as previously described [26] in TA-treated ratsthere were alterations of the outer segments and vacuolizationof the RPE cells (Figure 9D stars)Analysis of rat choroidal vascular corrosion casts one monthafter TA exposure To better analyze and quantify themodifications observed in the choroidal vascular network we

performed resin casts In rats exposed to TA for a month thecasts showed that the spaces between choriocapillaries wereincreased and that capillary diameter was reduced ascompared to PBS-injected rats Figure 10AB presents frontalviews of choriocapillary casts of a PBS-treated rat and a TA-treated rat Figure 10CD shows cross-sectional cuts ofpericentral choroidal corrosion casts of a PBS-treated rat anda TA-treated rat In the TA-treated rat eyes the avascular areasbetween choriocapillaries were significantly increased (n=6plt005 versus PBS) and the choriocapillary diametersignificantly reduced (n=6 plt005 versus PBS Figure10EF)Modulation of VEGF and COX-2 expression in RPE cellsexposed to OS phagocytosis To explore a potential indirectmechanism responsible for TA-induced choriocapillariesalterations we evaluated whether the expression of cyclo-oxygenase-2 (COX-2) and VEGF in RPE cells submitted toOS phagocytosis was modulated by TA exposure We havepreviously shown that COX-2 knockout mice developsignificant choriocapillary involution [26] So we hypothesed

Figure 8 Evaluation of autophagy by microtubules-associated protein light chain 3 staining No intracytoplasmic vacuoles were stained withanti-microtubules-associated protein light chain 3 (MAP-LC3) in control BRECs (A-C) This is also the case for BRECs exposed to 01 mgml TA for five days (D-F) In the amino acid (AA) deprived medium (positive control) BRECs show multiple MAP-LC3 positive cytoplasmicvacuoles (G-I arrows in I) Nuclei were stained with DAPI (A D G) Scale bar represents 50 microm

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2643

that beside a direct toxic effect of TA COX-2 downregulationcould also be induced by TA on the choriocapillary in vivo

In primary culture rat RPE cells OS phagocytosissignificantly enhanced COX-2 expression (Figure 11Aplt0001 OS versus control) TA decreased significantlythe expression of COX-2 under basal condition (Figure 11Aplt005 versus control) and also significantly reduced the OSphagocytosis-induced COX-2 upregulation (plt005 OS-TAversus OS) Interestingly while TA also significantlydownregulated basal VEGF expression (Figure 11B plt005TA versus control) it did not influence OS phagocytosis-induced VEGF upregulation (Figure 11B plt0001 OSversus control)

DISCUSSIONGlucocorticoids are known to induce vascular effects TheMcKenzie test based on skin bleaching has been the basis forthe classification of glucocorticoid potency and cutaneousabsorption [31] This test uses the visual quantification of theskin whitening after 8 to 48 h of exposure to differentglucocorticoids applied topically and under occlusionAlthough this test is widely used the exact mechanism of skinwhitening remains unknown and unexplored The effects ofsteroids on growing vessels have been described by Folkmanand Ingber [32] They showed that only a low dose ofdexamethasone (02 microgmicrol) or hydrocortisone (5ndash6 microgml)could induce the regression and prevent the growth of

Figure 9 Semithin analysis of the retina-choroid At 30 days after triamcinoloneacetonide (TA) injection the diameterof retinal vessels was reduced ascompared to control PBS injected eyes(B A respectively arrows) Thethickness of the choroids wassignificantly reduced in TA-treated (D)as compared to the control eye (C andE) Vacuoles were observed in retinalpigment epithelial (RPE) cells and outersegment in the TA-treated eyes (Dstars) Results are expressed as meanplusmnstandard error with plt0001Fours animals were used in this studyAbbreviations outer nuclear layer(ONL) inner segment (IS) outersegment (OS) Scale bar represents 20microm

Figure 10 Choroidal vascular castsFrontal view of choroidal casts showedthat 30 days after injection oftriamcinolone acetonide (TA) theavascular areas were increased (B reddouble-headed arrow) as compared tothe control PBS-treated eyes (A) Cross-sectional cuts of pericentral choroidalcorrosion cast (C D) showed a reducedthickness of the choriocapillaries in TA-treated eyes Scale bar represents 10 micromE Quantification of the avascular areasin the choriocapillaries confirmed asignificant increase in TA-treated eyesand (F) mean vessel diameter wassignificantly reduced Results areexpressed as meanplusmnstandard error withplt005 when consideringTA versusPBS Six animals were used in thisstudy

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2644

neovessels in the chorioallantoid membrane High doses over10 microgml were toxic Folkman experiments with chemicallymodified steroids demonstrated that the vascular effect wasnot related to the corticosteroid activity but to their effect onthe basal membrane of growing vessels Further studies showthat steroids inhibit human endothelial cells proliferation[33] or promote human dermal microvascular cells [34] theseeffects depending on the type of steroid the doses and theduration of exposure [35]

In many ocular models of angiogenesis and of humanpathologies the growth of neovessels takes place in a complexwound healing process where inflammation and angiogenesisare closely interlinked [36] Therefore the effects ofglucocorticoids result from both a strong anti-inflammatoryactivity and a direct but unclear effect on vessels

In vitro experiments performed in our study allowed usto analyze the direct effect of TA on BRECs We havedemonstrated that the effects of TA depend on both the TA

Figure 11 COX-2 and VEGF mRNA expression in RPE cells treatedwith TA COX-2 (A) and VEGF (B) mRNA expression (real timequantitative PCR) in untreated control primary rat RPE cells (C) orin RPE cells exposed to triamcinolone acetonide (TA) outersegments (OS) or both (OS-TA) Results are expressed as meanplusmnstandard error with plt005 plt0001 when considering allcolumns versus control in (B) and in (A) plt005 OS-TA versus OSFour animals were used in each study

solubility and the duration of exposure At 1 mgml TA didnot dissolve completely in ethanol 1 and the observednecrosis was due to the direct contact of crystals with thecultured cells as previously observed with RPE and rat Muumlllerglial (RMG) cells [25] When confluent cells were exposed to01 mgml TA-E a clear decrease in the number of living cellswas observed which resulted from noncaspase-dependentapoptotic pathways Similarly Hartnett et al found that TAinduced a dose-dependent toxic effect on human retinalmicrovascular endothelial cells without any activation of thecaspase-3 [15] suggesting that noncaspase-dependentapoptotic pathways could be activated

In our experiments LEIL-DNase II was associated withapoptosis of BRECs after 72 h of TA exposure This is animportant result because this enzyme produces 3primeP ends [37]in DNA a break that is not labeled by terminal transferaseThis explains why theses cells showing apoptotic morphologyand Annexin V staining remain TUNEL negatives Moreoveras this is the regular technique used to evaluate the inductionof cell death in a tissue many dying cells remain unrevealedIn addition to LEIL-DNase II AIF is also nuclearizedsuggesting that it is involved in the death of cells after 5 daysof TA exposure So that multiple caspase-independentmechanisms may mediate this cell death [38] Despite anabundant literature on the anti-angiogenic activity of steroidsthis is to our knowledge the first study to analyze themechanisms of TA-induced endothelial cell death At lowdose (1 microM) TA did not influence BREC proliferation whilea moderate effect on cell proliferation was observed at thedose of 01 mgml confirming the dose-dependence ofendothelial cell proliferation

The more striking observation was that after intravitreousinjection TA also induced vascular changes mostly at thelevel of the choroids We could not measure precisely in vivowhether TA induced any changes in retinal vessel diameter asdemonstrated in patients [39] but our histological analysisseemed to confirm a potential effect of TA on superficialretinal capillary diameter However no striking effects on theretinal vessel structure were observed Moreover using bothsemithin and ultrathin analysis and choroidal casts we couldobserve a reduction in the choroidal thickness a rarefactionof cells in the choroids and a rarefaction of choriocapillarieswith increased spaces between the capillaries At this stage itis not possible to determine whether the increased spacebetween capillaries results from a loss of capillaries or avasoconstriction of the vessels Further studies are undertakento evaluate potential toxic effect of TA on resting versusproliferating choroidal neovessels

Interestingly endothelial cell damage has been reportedin veins of rabbits treated with 4 mgkg methylprednisoloneacetate for four weeks They show ultrastructural alterationsof endothelial cells and pericytes [40] suggesting that in vivosteroids may be toxic for the vasculature This observation is

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2645

in agreement with the superficial skin ecchymoses known tooccur after long-term corticotherapy [41] Finally Blaha et aldescribed the occurrence of profound choroidalhypoperfusion in patients treated with combinedphotodynamic therapy and intravitreous injections oftriamcinolone [42] It remains to be determined if thiscomplication is related to the direct potential effect of TA

The relative selective effect of TA on thechoriocapillaries as compared to the retinal vasculature invivo may result from different factors First thechoriocapillary endothelium is fenestrated and much lessprotected than retinal capillaries which are surrounded bypericytes astrocytes and retinal Muumlller glial cells So that thechoriocapillary endothelium may be more exposed to TAtoxicity than the retinal endothelium Aside from a direct toxiceffect TA could exert an indirect effect through thedownregulation of COX2 in RPE cells In a previous studywe indeed demonstrated that CD36 minusminus mice that producereduced COX-2 levels in RPE cells develop a progressivechoriocapillary involution The activation of COX-2 inducedby OS phagocytosis could be important in the homeostasis ofchoriocapillaries [26] Interestingly in our experiments whileTA in vitro prevented the upregulation of COX2 under OSphagocytosis it did not affect VEGF upregulation Whetherother pro angiogenic factors are downregulated downstreamCOX2 in RPE cells will be investigated

Other pro angiogenic factors could be downregulateddownstream of COX-2 in RPE cells Moreover whenangiogenesis develops in the choriocapillaries as in age-related macular degeneration the anti-inflammatory and anti-proliferative effects of TA may be involved in its antivasculareffects [13] These issues will be matters for futureinvestigation

In conclusion this is the first study to demonstrate thatglucocorticoids induce direct retinal endothelial cells toxicitythrough noncaspase apoptotic pathways involving LEI L-DNase II and AIF While long-term exposure to TA did notinduce any detectable toxicity for retinal vessels in vivo it didlead to choriocapillary rarefaction Taking into account thatthose changes may occur without any inflammatory reactionin the long term they may be overlooked in clinical practiceand induce deleterious consequences on visual acuityHowever because Lewis rats are albinos and because the ratretina differs from the human retina no direct extrapolationcan be made regarding toxicity in patients This study is anindication that triamcinolone may exert some effects on thechoroidal vessels and that this parameter should be monitoredparticularly in patients with preexisting chorioretinalpathologies The effects of TA injected in the suprachoroidalspace of pigmented rats are currently under investigation tocomplete the present study Further studies should beundertaken to characterize precisely whether choriocapillarychanges occur in patients treated with repeated intravitreousinjections of TA

ACKNOWLEDGMENTSWe thank Dr J Ploueumlt who generously gave us the BRECscells and we also thank Dr I Le Disquet for technicalassistance during this study particularly in using the scanningelectron microscope

REFERENCES1 Greenberg PB Martidis A Rogers AH Duker JS Reichel E

Intravitreal triamcinolone acetonide for macular oedema dueto central retinal vein occlusion Br J Ophthalmol 200286247-8 [PMID 11815359]

2 Jonas JB Kamppeter BA Harder B Vossmerbaeumer USauder G Spandau UH Intravitreal triamcinolone acetonidefor diabetic macular edema a prospective randomized studyJ Ocul Pharmacol Ther 2006 22200-7 [PMID 16808682]

3 Danis RP Ciulla TA Pratt LM Anliker W Intravitrealtriamcinolone acetonide in exudative age-related maculardegeneration Retina 2000 20244-50 [PMID 10872928]

4 Challa JK Gillies MC Penfold PL Gyory JF Hunyor ABBillson FA Exudative macular degeneration and intravitrealtriamcinolone 18 month follow up Aust N Z J Ophthalmol1998 26277-81 [PMID 9843254]

5 Lee J Freeman WR Azen SP Chung EJ Koh HJ Prospectiverandomized clinical trial of intravitreal triamcinolonetreatment of neovascular age-related macular degenerationone-year results Retina 2007 271205-13 [PMID18046226]

6 Gillies MC Simpson JM Luo W Penfold P Hunyor AB ChuaW Mitchell P Billson F A randomized clinical trial of asingle dose of intravitreal triamcinolone acetonide forneovascular age-related macular degeneration one-yearresults Arch Ophthalmol 2003 121667-73 [PMID12742844]

7 Krieglstein TR Kampik A Ulbig M Intravitreal triamcinoloneand laser photocoagulation for retinal angiomatousproliferation Br J Ophthalmol 2006 901357-60 [PMID16885191]

8 Krebs I Krepler K Stolba U Goll A Binder S Retinalangiomatous proliferation combined therapy of intravitrealtriamcinolone acetonide and PDT versus PDT alone GraefesArch Clin Exp Ophthalmol 2008 246237-43 [PMID17674018]

9 OKeefe M Lanigan B Byrne SA Capillary haemangioma ofthe eyelids and orbit a clinical review of the safety andefficacy of intralesional steroid Acta Ophthalmol Scand2003 81294-8 [PMID 12780411]

10 Falkenstein IA Cheng L Wong-Staal F Tammewar AMBarron EC Silva GA Li QX Yu D Hysell M Liu G Ke NMacdonald JE Freeman WR Toxicity and intraocularproperties of a novel long-acting anti-proliferative and anti-angiogenic compound IMS2186 Curr Eye Res 200833599-609 [PMID 18600493]

11 Kato A Kimura H Okabe K Okabe J Kunou N Nozaki MOgura Y Suppression of laser-induced choroidalneovascularization by posterior sub-tenon administration oftriamcinolone acetonide Retina 2005 25503-9 [PMID15933599]

12 Criswell MH Hu WZ Steffens TJ Li R Margaron PComparing pegaptanib and triamcinolone efficacy in the rat

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2646

choroidal neovascularization model Arch Ophthalmol 2008126946-52 [PMID 18625941]

13 Dot C Behar-Cohen F BenEzra D Doat M Jonet L May FJeanny JC Influence of triamcinolone intravitreal injectionon retinochoroidal healing processes Exp Eye Res 2007841081-9 [PMID 17408616]

14 Akkoyun I Yilmaz G Oto S Kahraman B Haberal N AkovaYA Impact of triamcinolone acetonide on retinal endothelialcells in a retinopathy of prematurity mouse model ActaOphthalmol Scand 2007 85791-4 [PMID 17488319]

15 Hartnett ME Martiniuk DJ Saito Y Geisen P Peterson LJMcColm JR Triamcinolone reduces neovascularizationcapillary density and IGF-1 receptor phosphorylation in amodel of oxygen-induced retinopathy Invest Ophthalmol VisSci 2006 474975-82 [PMID 17065516]

16 Kim YH Choi MY Kim YS Park CH Lee JH Chung IY YooJM Choi WS Cho GJ Kang SS Triamcinolone acetonideprotects the rat retina from STZ-induced acute inflammationand early vascular leakage Life Sci 2007 811167-73[PMID 17881007]

17 Spandau UH Sauder G Schubert U Hammes HP Jonas JBEffect of triamcinolone acetonide on proliferation of retinalendothelial cells in vitro and in vivo Br J Ophthalmol 200589745-7 [PMID 15923512]

18 Penfold PL Wen L Madigan MC King NJ Provis JMModulation of permeability and adhesion moleculeexpression by human choroidal endothelial cells InvestOphthalmol Vis Sci 2002 433125-30 [PMID 12202538]

19 Gokhale P Patel T Morrison MJ Vissers MC The effect ofintracellular ascorbate on the susceptibility of HL60 andJurkat cells to chemotherapy agents Apoptosis 2006111737-46 [PMID 16951922]

20 Nur-E-Kamal A Gross SR Pan Z Balklava Z Ma J Liu LFNuclear translocation of cytochrome c during apoptosis JBiol Chem 2004 27924911-4 [PMID 15073175]

21 Yasugi E Kumagai T Nishikawa Y Okuma E Saeki KOshima M Susin SA Kroemer G Yuo A Involvement ofapoptosis inducing factor during dolichyl monophosphate-induced apoptosis in U937 cells FEBS Lett 2000480197-200 [PMID 11034328]

22 Tanida I Tanida-Miyake E Ueno T Kominami E The humanhomolog of Saccharomyces cerevisiae Apg7p is a Protein-activating enzyme for multiple substrates including humanApg12p GATE-16 GABARAP and MAP-LC3 J BiolChem 2001 2761701-6 [PMID 11096062]

23 Torriglia A Chaudun E Chany-Fournier F Jeanny JC CourtoisY Counis MF Involvement of DNase II in nucleardegeneration during lens cell differentiation J Biol Chem1995 27028579-85 [PMID 7499373]

24 Altairac S Zeggai S Perani P Courtois Y Torriglia AApoptosis induced by Na+H+ antiport inhibition activatesthe LEILDNase II pathway Cell Death Differ 200310548-57 [PMID 12728253]

25 Valamanesh F Torriglia A Savoldelli M Gandolphe C JeannyJC BenEzra D Behar-Cohen F Glucocorticoids induceretinal toxicity through mechanisms mainly associated withparaptosis Mol Vis 2007 131746-57 [PMID 17960113]

26 Houssier M Raoul W Lavalette S Keller N Guillonneau XBaragatti B Jonet L Jeanny JC Behar-Cohen F Coceani FScherman D Lachapelle P Ong H Chemtob S Sennlaub F

CD36 deficiency leads to choroidal involution via COX2down-regulation in rodents PLoS Med 2008 5e39 [PMID18288886]

27 Molday RS Hicks D Molday L Peripherin A rim-specific membrane protein of rod outer segment discs Invest

Ophthalmol Vis Sci 1987 2850-61 [PMID 2433249]28 Fadok VA Voelker DR Campbell PA Cohen JJ Bratton DL

Henson PM Exposure of phosphatidylserine on the surfaceof apoptotic lymphocytes triggers specific recognition andremoval by macrophages J Immunol 1992 1482207-16[PMID 1545126]

29 Andree HA Reutelingsperger CP Hauptmann R Hemker HCHermens WT Willems GM Binding of vascularanticoagulant alpha to planar phospholipid bilayers J BiolChem 1990 2654923-8 [PMID 2138622]

30 Torriglia A Perani P Brossas JY Chaudun E Treton JCourtois Y Counis MF L-DNase II a molecule that links proteases and endonucleases in apoptosis derives from theubiquitous serpin leukocyte elastase inhibitor Mol Cell Biol 1998 183612-9 [PMID 9584202]

31 McKenzie AW Stoughton RB Method comparingpercutaneous absorption of steroids Arch Dermatol 196286608-10

32 Folkman J Ingber DE Angiostatic steroids Method ofdiscovery and mechanism of action Ann Surg 1987206374-83 [PMID 2443088]

33 Cariou R Harousseau JL Tobelem G Inhibition of humanendothelial cell proliferation by heparin and steroids CellBiol Int Rep 1988 121037-47 [PMID 2852064]

34 Hettmannsperger U Tenorio S Orfanos CE Detmar MCorticosteroids induce proliferation but do not influenceTNF- or IL-1 beta-induced ICAM-1 expression of humandermal microvascular endothelial cells in vitro ArchDermatol Res 1993 285347-51 [PMID 8105755]

35 Hasan Q Dafydd H Tan ST Xu B Davis PF The optimal typedosage and timing of glucocorticoid administration fortreatment of hemangioma in vitro J Appl Res 20055360-71 [PMID 8105755]

36 Kent D Sheridan C Choroidal neovascularization a woundhealing perspective Mol Vis 2003 9747-55 [PMID14735062]

37 Torriglia A Leprecirctre C Padroacuten-Barthe L Chahory S MartinE Molecular mechanism of L-DNase II activation andfunction as a molecular switch in apoptosis BiochemPharmacol 2008 761490-502 [PMID 18761000]

38 Broughton BR Reutens DC Sobey CG Apoptotic MechanismsAfter Cerebral Ischemia Stroke 2009 40e331-9 [PMID19182083]

39 Jonas JB Rensch F Decreased retinal vein diameter afterintravitreal triamcinolone for retinal vein occlusions Br JOphthalmol 2007 911711-2 [PMID 18024817]

40 Nishimura T Matsumoto T Nishino M Tomita KHistopathologic study of veins in steroid treated rabbits ClinOrthop Relat Res 1997 33437-42 [PMID 9005894]

41 Allen DB Bielory L Derendorf H Dluhy R Colice GL SzeflerSJ Inhaled corticosteroids past lessons and future issues JAllergy Clin Immunol 2003 112S1-40 [PMID 14515117]

42 Blaha GR Wertz FD 3rd Marx JL Profound choroidalhypoperfusion after combined photodynamic therapy and

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

2647

intravitreal triamcinolone acetonide Ophthalmic Surg LasersImaging 2008 396-11 [PMID 18254344]

Molecular Vision 2009 152634-2648 lthttpwwwmolvisorgmolvisv15a281gt copy 2009 Molecular Vision

The print version of this article was created on 4 December 2009 This reflects all typographical corrections and errata to thearticle through that date Details of any changes may be found in the online version of the article

2648