document

304 NATURE MEDICINE • VOLUME 7 • NUMBER 3 • MARCH 2001

ARTICLES

The glaucomas, characterized by cupping of the optic nerve headand irreversible loss of retinal ganglion cells, affect approximately70 million people worldwide1. Elevated intraocular pressure (IOP)due to reduction in aqueous outflow facility is a major causal riskfactor. The main aqueous outflow pathway of the eye consists of aseries of endothelial-cell–lined channels in the angle of the anteriorchamber comprising the trabecular meshwork (TM), Schlemm’scanal, the collector channels and the episcleral venous system. Inclosed-angle forms of glaucoma, elevated IOP is due to anatomicobstruction of this pathway2,3. Factors causing reduced outflow fa-cility in the open angle glaucomas may include accumulation ofextraneous material or cells within the TM, alterations in junc-tional structures between cells of the TM or Schlemm’s canal, accel-erated TM cell death and collapse of trabecular beams2–7.

Chronic sublethal injury due to cellular stress is a commontheme in the pathogenesis of diverse diseases including atheroscle-rosis, glomerulonephritis and pulmonary fibrosis8,9. Previous stud-ies indicate that the glaucomas are part of this disease group, withsublethal damage to the outflow pathways being the result of accu-mulated oxidative stress arising from the environment, vasculardysregulation, aging and/or the disease process itself3,10–12. Thepathophysiology of diseases involving sublethal cell injury is deter-mined largely by the cells surviving the damaging insult. Thesecells mount a protective response involving expression of newgenes and other molecular changes8,9, dependent on the nature ofthe damaging stimulus and on tissue type13,14. A few studies havecompared the molecular phenotypes of glaucomatous with normalTM (refs. 4,15); however, a diagnostic marker of diseased TM hasnot yet been identified.

ELAM-1 is a disease marker for glaucomatous TMCell-adhesion molecules (CAMs) are key functional components of

biological structures for fluid containment and transport16.Moreover, upregulation of selectin- and immunoglobulin-typeCAMs is a common response to cellular stress agents implicated invascular disease17,18. To characterize CAMs in the aqueous outflowpathways of normal eyes, and to determine whether CAM expres-sion might be altered in glaucoma, we performed an immunohisto-chemical screen using a battery of vascular endothelial CAMprobes. Six of the eight vascular CAM probes consistently stainedthe TM and Schlemm’s canal of both normal and glaucomatouseyes. A representative experiment with the N-CAM-16 (neural celladhesion molecule-16) marker is shown in Fig. 1. Staining forICAM-1 (intercellular adhesion molecule-1), VCAM-1 (vascular celladhesion molecule-1) and integrin α3 was also easily detected, andtrace staining for integrin α2 and integrin α5 was apparent. A sev-enth vascular marker, PECAM-1 (platelet/endothelial cell adhesionmolecule-1), stained only Schlemm’s canal. The level of immunos-taining for each of the vascular CAMs was similar in normal andglaucomatous eyes. Normal and glaucomatous TM-cell lines re-tained these markers, as shown by N-CAM-16 staining (Fig. 1).

Of these, only the probe for ELAM-1 differentiated the glauco-matous from the normal eye aqueous outflow pathway (Fig. 1), re-gardless of glaucoma subtype or severity (Table 1), andindependent of prior glaucoma therapy (four eyes had no treat-ment prior to surgical intervention). ELAM-1 staining was absent inthe outflow pathway of normal eyes, but was clearly present inglaucomatous eyes in the region of Schlemm’s canal and the sur-rounding TM. The differential expression of ELAM-1 was retainedin subcultured TM cells. ELAM-1 was consistently present in allglaucomatous tissue specimens and TM-cell lines examined. Incontrast, ELAM-1 was consistently absent from the TM of all nor-mal tissue specimens and TM-cell lines. This screen defines a mole-cular marker specific for glaucomatous TM cells that is retained

Activation of a tissue-specific stress response in the aqueousoutflow pathway of the eye defines the glaucoma disease

phenotype

NAN WANG, SHRAVAN K. CHINTALA, M. ELIZABETH FINI & JOEL S. SCHUMAN

Vision Research Laboratories, New England Eye Center, Tufts University School of Medicine, Boston, Massachusetts, USA

N.W. and S.K.C. contributed equally to this study.Correspondence should be addressed to J.S.S.; email: [email protected]

The glaucomas are a group of optic neuropathies comprising the leading cause of irreversible blind-ness worldwide. Elevated intraocular pressure due to a reduction in normal aqueous outflow is amajor causal risk factor. We found that endothelial leukocyte adhesion molecule-1 (ELAM-1), the ear-liest marker for the atherosclerotic plaque in the vasculature, was consistently present on trabecularmeshwork (TM) cells in the outflow pathways of eyes with glaucomas of diverse etiology. We deter-mined expression of ELAM-1 to be controlled by activation of an interleukin-1 (IL-1) autocrine feed-back loop through transcription factor NF-κB, and activity of this signaling pathway was shown toprotect TM cells against oxidative stress. These findings characterize a protective stress response spe-cific to the eye’s aqueous outflow pathways and provide the first known diagnostic indicator of glau-comatous TM cells. They further indicate that common mechanisms contribute to thepathophysiology of the glaucomas and vascular diseases.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co

m© 2001 Nature Publishing Group http://medicine.nature.com

NATURE MEDICINE • VOLUME 7 • NUMBER 3 • MARCH 2001 305

ARTICLES

Fig. 1 Immunohistochemical screen for CAMs altered in outflow path-ways of glaucomatous eyes. Cross-sections through the angle of the ante-rior chamber of glaucomatous or normal eyes (left panels), or sub-culturedTM-cell lines from glaucomatous or normal eyes (right panels) werestained with N-CAM-16 or ELAM-1 antibody probes. Note the strong stain-

ing of Schlemm’s canal (arrow) and the surrounding TM (arrowhead) inglaucomatous eyes and TM cells from glaucomatous eyes. Tissue sectionswere negative for the inflammation marker LFA-1 and the fibrosis markerHPCA-2, and both tissue sections and cultured cells were positive for theTM cell marker HLA class I antigen (data not shown).

even when the cells are subcultured.

ELAM-1 expression is not due to inflammationELAM-1 expression is stimulated in normal vascular endotheliumby inflammatory cytokines such as IL-1 (ref. 19). Many of the glau-comatous eyes that we examined had forms of disease associatedwith an inflammatory component. In addition, laser therapy orthe preservatives in glaucoma medications could induce inflamma-tion20. On gross examination, however, there was no evidence ofinflammation or trauma in the glaucomatous cadavereyes. Trabeculectomy was performed only on eyes inwhich inflammation was subdued or absent, andnone of these eyes had received laser therapy withinthree months of surgery. The prostaglandin analogdrugs might induce inflammatory cytokines21; how-ever, only one of the cadaver eyes and only five of thetrabeculectomy eyes had been treated with suchdrugs. One cadaver eye came from a patient whose dis-ease was diagnosed only two months prior to death, sothere was little opportunity for drug therapy. One tra-beculectomy patient with primary open-angle glau-coma and three with angle-closure glaucoma hadreceived no treatment whatsoever for their disease.Routine hematoxylin and eosin staining of tissue sec-tions revealed no evidence of inflammation.Moreover, immunostaining revealed that all normaland glaucomatous eyes were positive for the TM-cellmarker used in this study (HLA class I antigen), butnegative for both the leukocyte marker and the fibro-sis marker (data not shown). These results indicatethat the commonality among all of the glaucomatouseyes leading to ELAM-1 expression is the diseaseprocess itself.

ELAM-1 expression and IL-1-NF-κB signalingAutocrine cytokine feedback loops can be activated intissue repair, disease and cellular aging, and serve as amechanism for amplifying and sustaining expressionof specific genes22. Exogenous IL-1 stimulated expres-sion of ELAM-1 mRNA (Fig. 2a) and protein (Fig. 2c) innormal TM cells. The mRNA for IL-1α was unde-

tectable in normal TM cells, and the mRNAs for IL-1β and IL-6 werefound only at low levels (Fig. 2b). Exogenous IL-1 stimulated ex-pression of each of these mRNAs (Fig. 2b). In contrast, IL-1α, IL-1βand IL-6 mRNAs were present at easily detectable levels in un-treated TM cells derived from glaucomatous specimens (Fig. 2b),and correlated positively with ELAM-1 expression (Fig. 2a).Treatment with IL-1 receptor antagonist (IL-1ra), a naturally occur-ring analog of IL-1 that binds to IL-1 receptors but does not trans-duce a signal23, significantly downregulated ELAM-1 expression

Table 1 Clinical findings in surgical specimens

Patient (age/sex) Diagnosis IOP (mmHg) Cup/Disc ratio Visual field Analyses73/M POAG 29 0.4 × 0.4 INS I58/M PXFG 25 0.8 × 0.6 INS, SAS I49/F POAG 50 0.95 × 0.95 SAS, INS I88/F PXFG 30 0.9 × 0.9 SNS, SPC I80/F PXFG 23 0.9 × 0.9 Full I75/F POAG 16 0.8 × 0.8 SAS I78/F Inflammatory 14 0.9 × 0.9 Nonspecific I

Glaucoma loss59/M Inflammatory 30 0.3 × 0.3 IAS, SNS I

Glaucoma60/M Inflammatory 16 0.9 × 0.8 SNS I

Glaucoma40/M POAG 40 0.9 × 0.9 Central island I73/M POAG 18 0.95 × 0.95 SAS, IAS I85/M POAG 15 0.9 × 0.9 SAS, IPC I+N+R60/F POAG 18 0.1 × 0.1 INS, IPC I+N+R88/F POAG 25 0.8 × 0.8 IAS, SAS I68/F PXFG 22 0.9 × 0.9 SAS I69/F PXFG 15 0.9 × 0.9 IHD, OS I16/M JOAG 27 0.95 × 0.95 Central island I+N47/M POAG 18 0.95 × 0.95 Central island I+N+E41/M OAG 19 0.8 × 0.8 Full I81/F PXFG 30 0.8 × 0.8 Full I58/F ACG 42 0.3 × 0.3 NA I66/F ACG 50 0.3 × 0.3 NA I63/F ACG 38 0.3 × 0.3 NA I

The analyses performed are: immunohistochemistry (I); northern-blot (N); RT-PCR (R); EMSA (E). ACG:angle-closure glaucoma, IAS: inferior arcuate scotoma, IHD: inferior hemifield defect, INS: inferior nasalscotoma, IPC: inferior paracentral scotoma, JOAG: juvenile open-angle glaucoma, POAG: primary open-angle glaucoma, PXFG: pseudoexfoliation glaucoma, SAS: superior arcuate scotoma, SHD: superiorhemifield defect, SNS: superior nasal scotoma, SPC: superior paracentral scotoma, NA: not available

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co



ARTICLES

Fig. 2 Regulation of ELAM-1 expression byexogenous and endogenous IL-1. Normaland glaucomatous TM cells were plated intoreplicate cultures and were untreated (–) ortreated with IL-1 for 2 h (+). In some experi-ments, cells were also treated with receptor antagonist (IL-1ra) for 24 h. a,Northern blotting with ELAM-1 cDNA (ref. 42). The GAPDH probe35 served todetermine the equality of RNA loading among lanes. b, RT-PCR analysis for IL-1α,IL-1β and IL-6 mRNA. The cDNA for GAPDH was amplified in parallel reactions toassess cDNA-loading equivalence among samples42. c, Indirect immunofluores-cent staining with antibody to ELAM-1. NTM: normal TM cells. GTM-A: glauco-matous TM cells. GTM-B: glaucomatous TM cells, specimen-B. GTM-C:glaucomatous TM cells, specimen-C.

(Fig. 2c). These findings indicate that sustained ELAM-1 expressionin glaucomatous TM cells is controlled largely by autocrine IL-1, al-though expression may be amplified by other autocrine cytokinessuch as IL-6.

Expression of ELAM-1 (ref. 17), IL-1α, IL-1β and IL-6 (ref. 21) isdependent on activity of the NF-κB family of dimeric DNA-bindingcomplexes24. Electrophoretic mobility shift assay (EMSA) revealedthat exogenous IL-1 stimulated NF-κB DNA-binding activity in nor-mal TM cells (Fig. 3a). This binding activity was present constitu-tively in glaucomatous TM cells, and was further stimulated byexogenous IL-1. Antibodies to NF-κB family member p50 (NF-κB1)completely super-shifted the inducible complexes from normalcells and the constitutive complex from glaucomatous cells,whereas antibodies to p65 (Rel A) super-shifted a distinct subcom-ponent of these complexes (Fig. 3a). This is consistent with theidentities of the complex of faster mobility as p50–p50 homod-

imers and the complex of slower mobility as p50–p65 het-erodimers. Interestingly, p50 antibodies did not alter the elec-trophoretic mobility of the inducible complexes fromglaucomatous cells, although these complexes were shifted com-pletely with p65 antibodies (Fig. 3a). This indicates that disease al-ters the expression of NF-κB family members. IL-1ra treatmenteliminated the constitutive NF-κB DNA-binding complex fromglaucomatous cells (Fig. 3b). Treatment with the NF-κB antagonist,SN50 peptide, reduced ELAM-1 expression in glaucomatous TM

cells, whereas treatment with control peptide againhad no effect (Fig. 3c). These findings indicate thatIL-1 sustains ELAM-1 expression in the TM cells ofglaucomatous eyes through a signaling pathway thatculminates in activation of NF-κB.

Activation of NF-κB in response to stimulators in-volves release from an inhibitor, I-κB, exposing aprotein domain which enables translocation to thenucleus24,25. An antibody against the nuclear localiza-tion epitope which recognizes only the active form

a b c

a b

c

Fig. 3 NF-κB activity in normal and glaucomatous TM cells.a, TM cells isolated from normal and glaucomatous eyes wereleft untreated (–) or treated with IL-1 for 2 h (+). EMSA wasperformed. Arrowhead indicates the migration of NF-κB pro-tein–DNA complex. Specificity is indicated by supershift analy-sis with antibodies to the p50 or p65 subunits of NF-κB andcompetition analysis with ‘cold’ NF-κB probe (1:50). b, EMSAfrom TM cells left untreated (–), treated with IL-1 for 2 h, ortreated overnight with IL-1ra. Specificity was demonstratedusing a control probe corresponding to the E-box-like elementin the human IL-1α gene promoter35. Binding to this probewas unaffected by treatment with either IL-1 or IL-1ra (datanot shown). c, Indirect immunofluorescent staining with anti-body to ELAM-1.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co



ARTICLES

of p65 showed substantial binding to TM cells surroundingSchlemm’s canal in glaucomatous tissue specimens, but bindingwas not detectable in the outflow pathways of normal eyes (Fig. 4).The localization pattern of activated p65 antigen overlapped sub-stantially with that of ELAM-1 antigen. These data support the hy-pothesis that NF-κB activation stimulates ELAM-1 expression insitu.

IL-1-NF-κB signaling protects cells from oxidative stressIL-1 and NF-κB are reported to be protective of cells subjected tostress26,27. We performed an experiment to determine whether en-dogenous IL-1 protects glaucomatous TM cells (Fig. 5). Normal TMcells exhibited a dose-dependent apoptotic re-sponse when treated with the oxidant tBH. Incontrast, glaucomatous TM cells were resistantto the oxidant, with apoptosis apparent onlyat the highest dose in the range. When normalTM cells were treated with IL-1, the thresholdoxidant dose for eliciting apoptosis was alsoincreased. Conversely, when glaucomatousTM cells (which produce IL-1 endogenously)were treated with the IL-1 antagonist, IL-1ra,they exhibited less resistance to the oxidant.The NF-κB antagonist SN50 also decreased re-sistance, whereas control peptide had no ef-fect. These results support the hypothesis thatactivation of NF-κB through IL-1 is protectiveof glaucomatous TM cells.

DiscussionThe vascular endothelium and the TM have a common role in theformation of biological structures for containment and conductionof bodily fluids. As such, they share many properties, including theexpression of the CAM markers defined in this study. Nonetheless,the vascular endothelium and the TM can be clearly distinguishedby presence or absence of the vascular endothelial-cell marker, fac-tor VIII (ref. 28), and they are derived from different embryologicalprecursors29. ELAM-1 upregulation in the vascular endothelium hasbeen associated with a broad range of acute and chronic diseasestates in all organs of the body, including the eye30–32. A few reportsalso describe ELAM-1 expression in the inflamed corneal endothe-

Fig. 4 Active NF-κB in glaucoma-tous TM cells in situ. Double-labelimmunofluorescent localization wasperformed using antibody that bindsELAM-1 and antibody against thep65 nuclear localization signal,which binds only the activated formof this subunit free of I-κB. Sectionswere further stained with Hoechstdye to visualize nuclei. ‘Triple’ indi-cates the overlap of all images.Magnification: ×40.

Fig. 5 Resistance of normal and glaucomatous tra-becular meshwork cells to apoptosis in response to anoxidant, and protective role of IL-1 and NF-κB.Normal TM cells or glaucomatous cells were treatedwith indicated concentrations of the oxidant tBH for6 hours and stained with fluorescein-TUNEL assay .Groups of cells were pre-treated with IL-1 (10µg/ml),IL-1ra (500 µg/ml) or SN-50 peptide (NF-κB antago-nist, 50 µg/ml) then assayed for the apoptotic re-sponse. Cells were also stained with propidium iodideto stain the nucleus. Arrows indicate apoptotic cellswhich stain yellow because of the colocalization ofTUNEL staining and propidium iodide.

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co



ARTICLES

lium30, a tissue that is continuous with the TM and of similar em-bryonic origin. However, this is the first report of which we areaware that identifies ELAM-1 expression in the TM and Schlemm’scanal. We note that the juxtacanalicular location of ELAM-1 ex-pression that we observed corresponds to the site of maximal resis-tance to aqueous outflow, thought to be the location of theglaucomatous lesion in primary open angle glaucomas3–5.

Considering the structural and functional similarities betweenthe vasculature and the eye’s aqueous outflow pathway, it is notsurprising that diseases of these tissues share many points of con-vergence. For example, oxidized low-density lipoprotein8 is a majorinitiating factor in atherosclerosis, and oxidative stress is also im-plicated in glaucoma pathogenesis10. Fluid pressure gradients in theform of shear stress are implicated in stimulating ELAM-1 expres-sion in vascular disease14,17,18, and elevated blood pressure and IOPare key risk factors in the pathophysiology of vascular disease14,17

and glaucoma5, respectively. Inflammation and inflammatory cy-tokines17 are central to the stimulation of ELAM-1 expression in thediseased vasculature, and the inflammatory cytokine IL-1α is up-regulated in vascular endothelial cells isolated from the atheroscle-rotic lesion33. Only a few glaucomas are associated withinflammation; however, we found in this study that IL-1 is consis-tently upregulated by glaucomatous TM cells and acts as a key au-tocrine regulatory factor for ELAM-1 expression, much as invascular disease. This difference may be due to environment—thecellular and extracellular composition of arterial matrices are morecomplex than those of TM, and the acellular aqueous humorwhich percolates through the aqueous outflow pathways is differ-ent from the cellular blood that flows through arterial endothelialvessels. Inflammatory mediators may stimulate systemic effectswhen expressed by vascular endothelia, but effects may be locallycontained when expressed by TM.

Activation of the transcription factor NF-κB is a common endpoint for diverse stress signals, including oxidative stress34. Thegene for ELAM-1 has NF-κB response elements in its transcriptionalpromoter14,17. We show here that endogenous IL-1 (possibly aug-mented by IL-6) controls ELAM-1 expression in glaucomatous TMcells through NF-κB. NF-κB is also known to mediate expression ofthe inflammatory cytokines IL-1α, IL-1β and IL-621,24,35. This sug-gests a model in which sublethal damage to TM cells initiates NF-κB activation, which in turn activates the genes for both ELAM-1and the inflammatory cytokines. At first, ELAM-1 and IL-1/IL-6gene expression would occur only at very low levels depending onthe strength and duration of the initial damaging stimulus.However, this signal would be progressively amplified by the cy-tokines through further activation of NF-κB. Ultimately, NF-κB ac-tivation would become self-sustaining, and ELAM-1 and IL-1/IL-6expression would reach levels easily detectable by the methodsused in this study.

The genetic response to sublethal cellular stress confers protec-tion to the surviving tissue in a variety of disease states8,9. AutocrineIL-1 expression has been associated with replicative senescence incultured fibroblasts36, and treatment with exogenous IL-1 extendsthe replicative lifespan of cultured endothelial cells26. In addition, aprevious study has demonstrated that activity of NF-κB preventscells from undergoing apoptosis in response to stressful stimuli bycontrolling expression of genes encoding anti-apoptotic proteins27.Our results now place glaucoma in this group, as we show that IL-1produced endogenously by glaucomatous TM cells inhibits theapoptotic response to oxidative stress through NF-κB. IL-1 mayhave additional beneficial effects in glaucoma, as it has been re-ported to increase outflow facility37, perhaps through its ability to

stimulate expression of matrix metalloproteinases (MMPs)38.Although the tissue response to stress may be protective in the

short run, if continued on a chronic basis it has the potential to be-come the disease entity itself, a lesson already learned for vasculardisease8. In open-angle forms of glaucoma, IOP would increase asTM damage accumulated, secondarily amplifying NF-κB activa-tion12,18. In closed-angle forms of glaucoma, elevated IOP resultingfrom anatomic obstruction of the angle would be the primary ox-idative stress initiating NF-κB activation, but this would be subse-quently amplified through autocrine cytokine loops. TheIL-1–NF-κB pathway generates oxygen free radicals as signaling in-termediates12,34, which could cause cell damage10. Moreover, down-stream targets of this pathway such as MMPs could ultimatelycause irreversible damage to the trabecular beams. Thus, activationof the IL-1/NF-κB pathway may provide a unifying disease mecha-nism for pathophysiology in glaucomas of diverse etiology.

To our knowledge, ELAM-1 represents the first molecular markerof the glaucoma disease phenotype, that is, it is expressed in an all-or-none fashion, and could be used diagnostically. However, acti-vation of the NF-κB signaling pathway in TM cells may be the morecomprehensive determinant of disease, because it controls not justone gene (ELAM-1), but a whole battery24. We detected upregula-tion of three additional members of this battery (IL-1α, IL-1β andIL-6) in glaucomatous TM, although we reserve judgement on theirdesignation as true disease markers since we did not rigorously doc-ument their absence from normal TM. We predict that other dis-ease markers besides ELAM-1 will be found once additionalmembers of the gene battery stimulated by NF-κB in the TM areidentified. The possible role of ELAM-1 in protection or pathophys-iology of the glaucomas remains to be learned.

The continued capacity to distinguish normal and diseased TMcells on the basis of ELAM-1 expression even after subculture indi-cates that TM cells retain a memory of their disease history whenremoved from the glaucomatous conditions in the eye. This meansthat TM cells can be amplified in culture to provide material formolecular characterization of the disease lesion by application ofdifferential cloning and gene profiling strategies. All current thera-peutic modalities for glaucoma are aimed at reducing IOP by de-creasing aqueous humor production or by increasing outflowfacility through alternative pathways5. Greater understanding ofthe mechanisms leading to TM-cell damage in the glaucomatouseye will facilitate development of new management strategies thattarget the primary disease lesion.

MethodsTissue specimens, embedding and cell culture. Sixteen normal (age range: 4d to 88 y) and 12 glaucomatous (age range: 67 to 86 y) cadaver eyes were ob-tained from the National Disease Research Interchange and the New EnglandEye Bank. Eyes were enucleated within 2–4 h after death and processed for ex-periments within 20–36 h. Anterior segments were isolated, lenses were re-moved and the specimens were halved; one portion was then embedded forfrozen sectioning and the TM was isolated from the other portion and ex-planted to culture. Surgical specimens from 23 glaucoma patients (age range16–88 y) were used for either frozen section or culture (Table 1). Cultures fromsingle individuals were maintained as separate lines and used before fourthpassage. In some experiments, cells were treated with IL-1α at 10 ng/mland/or IL-1ra at 100–1000 ng/ml (R&D Systems, Minneapolis, Minnesota). Inother experiments, cells were treated with the NF-κB antagonist, SN50 or con-trol peptide (Biomol, Plymouth Meeting, Pennsylvania) at 50 µg/ml.

Immunolocalizations. Nine different CAM monoclonals were used as primaryantibodies for indirect immunolocalization on frozen tissue sections (4 µM) orcells cultured on glass slides. These were the selectin family member ELAM-1(E-selectin; CD62E); the immunoglobulin superfamily members N-CAM-16

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co



ARTICLES

(CD56), ICAM-1 (CD54), PECAM-1 (CD31) and VCAM-1 (CD106); and the in-tegrin family members integrin α2 (VLA-2; CD49b), integrin α3 (VLA-3;CD49c) and integrin α5 (VLA-5; CD49d). Control antibody probes includedthe inflammation marker LFA-1α (lymphocyte function-associated antigen-1α;CD11a)39, the fibrosis marker HPCA-2 (human progenitor cell antigen-2;CD34)40, and the TM cell marker HLA class I (major histocompatibility complexreagent-1; HLA-ABC)41. Each probe was purchased from Becton Dickinson(San Jose, California) with the exception of PECAM-1 antibody, which was pur-chased from Dako (Carpinteria, California). Antibody used to detect activity ofthe NF-κB p65 subunit in tissue sections was purchased from Boehringer. Eightto 20 tissue sections and 4 coverslips of cultured cells were probed with eachprimary antibody. The avidin–biotin–peroxidase complex technique was usedto visualize antibody binding in initial experiments, but later experiments useda fluorescent secondary antibody. Appropriate controls for antibody specificitywere included. Stained slides were examined by light microscopy and positiv-ity was scored in a double-masked analysis.

RNA analysis. Northern blotting and reverse transcription (RT)-PCR were per-formed as described35,42. Sequences for construction of the IL-1α oligonu-cleotide primer pair were obtained from Genbank (primer:5′-CGGCTGCTGCATTACATAATCTGG; reverse: 5′-TGAAAGTCAGTGATA-GAGGGTGGC-3′). The IL-1β primer pairs were obtained from Clontech (for-ward: 5′-ATGGCAGAAGTACCTAAGCTCGC-3′; reverse:5′-ACACAAATTGCATGGTGAAGTCAGTT-3′). The IL-6 primer pair was pur-chased from R&D Systems (proprietary sequences).

Electrophoretic mobility shift assay (EMSA). Nuclear lysates were preparedfrom cultured TM cells and EMSA was performed using a protein equivalentamount of each lysate as described35. The radiolabeled NF-κB probe was 5′-AGTTGAGGGGACTTTCCCAGGC-3′ (Promega, Madison, Wisconsin).Supershift analysis was performed with p50 (NF-κB1) and p65 (RelA) antibod-ies purchased from Santa Cruz Biotechnology (Santa Cruz, California).

Apoptosis assay. TdT-mediated dUTP nick-end labeling (TUNEL) was per-formed using the in situ cell death detection kit with fluorescein (Boehringer).For this assay, TM cells plated in 8-well chamber slides at 1 × 105 cells/wellwere treated with 125–500 µM tert-butyl hydroperoxide (tBH, Sigma) for 6 h.

AcknowledgmentsWe thank members of the Schuman and Fini labs for their technical assistance.Some surgical specimens were provided by the Institute of Ophthalmology & EyeHospital, Shandong Academy of Medical Science, and the Department ofOphthalmology, Wuhan Municipal First Hospital, P.R. China. The ELAM-1 cDNAwas a gift from B. Seed and the GAPDH cDNA was a gift from R. Allen. This workwas supported by the Glaucoma Foundation, New York (J.S.S.), the AmericanHealth Assistance Foundation (J.S.S.), NEI grant EY09828 (M.E.F.), theMassachusetts Lions Eye Research Fund and Research to Prevent Blindness.

RECEIVED 6 NOVEMBER; ACCEPTED 28 DECEMBER 2000

1. Quigley, H.A. Number of people with glaucoma world-wide. Brit. J. Ophthalmol. 80,389–393 (1996).

2. Grant, W.M. Experimental aqueous perfusion in enucleated human eyes. Arch.Ophthalmol. 69, 783–801 (1963).

3. Chandler and Grant’s Glaucoma 4th edn. (eds. Epstein, D.L., Allingham, R.R. &Schuman, J.S.) 18–24 (Williams and Wilkins, Baltimore, Maryland, 1996).

4. Underwood, J.L. et al. Glucocorticoids regulate transendothelial fluid flow resistanceand formation of intracellular junctions. Am. J. Physiol. 277, C330–342 (1999).

5. Tripathi, R.C. Mechanism of the aqueous outflow across the trabecular wall ofSchlemm’s canal. Exp. Eye Res. 11, 116–121 (1971).

6. Schuman, J.S., Wang, N. & Eisenberg, D.L. Leukemic glaucoma: the effects on outflowfacility of chronic lymphocytic leukemic lymphocytes. Exp. Eye Res. 61, 609–617 (1995).

7. Lütjen-Drecoll, E., Shimizu, T., Rohrbach, M. & Rohen, J.W. Quantitative analysis of“plaque” material in the inner and outer wall of Schlemm’s canal in normal and glau-comatous eyes. Exp. Eye Res. 42, 443–455 (1986).

8. Ross, R. Cell Biology of Atherosclerosis. Annu. Review Physiol. 57, 791–804 (1995).9. Dunn, C.J. Cytokines as mediators of chronic inflammatory disease. in Cytokines and

Inflammation (ed. Kimball, E.S.) 1–34 (CRC Press, London, 1991).10. Green, K. Free radicals and aging of anterior segment tissues of the eye: a hypothesis.

Ophthalmic Res. 27, 143–149 (1995).11. Flammer, J., Haefliger, I.O., Orgul, S. & Resnick, T. Vascular deregulation: a principal

risk factor for glaucomatous damage? J. Glaucoma 8, 212–219 (1999).12. Yeh, L.H. et al. Shear-induced tyrosine phosphorylation in endothelial cells requires

Rac1-dependent production of ROS. Am. J. Physiol. 276, C838–847 (1999).13. Mercurio, F. & Manning, A.M. NF-κB as a primary regulator of the stress response.

Oncogene 18, 6163–6171 (1999).14. Itoh, H. & Nakao, K. Vascular stress response and endothelial vasoactive factors for vas-

cular remodeling. Diabetes Res. Clin. Pract. 45, 83–88 (1999).15. Polansky, J.R. et al. Cellular pharmacology and molecular biology of the trabecular

meshwork inducible glucocorticoid response gene product. Ophthalmologica 211,126–139 (1997).

16. Petruzelli, L., Takami, M., Mimi, T. & Humes, H.D. Structure and function of cell adhe-sion molecules. Am. J. Med. 106, 467–476 (1999).

17. Price D.T. & Loscalzo J. Cellular adhesion molecules and atherosclerosis. Am. J. Med.107, 85–97 (1999).

18. Gimbrone, M.A., Nagel, T. & Topper, J.N. Perspective series: Cell adhesion in vascularbiology. Biomechanical activation: an emerging paradigm in endothelial cell adhesionbiology. J. Clin. Invest. 99, 1809–1813 (1997).

19. Paulson, J.C. in Adhesion: Its role in Inflammatory Disease (eds. Harlan, J.M. & Liu, D.Y.)19–42 (W.H. Freeman, New York, 1992).

20. Kawa, J.E., Higginbotham, E.J., Chang, I.L. & Yue, B.Y.J. Effects of antiglaucoma med-ications on bovine trabecular meshwork cells in vitro. Exp. Eye Res. 14, 560–565 (1993).

21. Dinarello, C.A. Biology of interleukin 1. FASEB J. 2, 108–115 (1988).22. West-Mays, J.A., Strissel, K.J., Sadow, P.M. & Fini, M.E. Competence for collagenase

gene expression by tissue fibroblasts requires activation of an interleukin 1α autocrineloop. Proc. Natl. Acad. Sci. USA 92, 6768–6772 (1995).

23. Dinarello, C.A. & Thompson, R.C. Blocking IL-1: interleukin 1 receptor antagonist invivo and in vitro. Immunol. Today 12, 404–410 (1991).

24. Barnes, P.J. & Karin, M. Nuclear factor-κB: a pivotal transcription factor in chronic in-flammatory diseases. N. Engl. J. Med. 336, 1066–71 (1997).

25. Bauerle, P.A. & Baltimore, D. IκB: a specific inhibitor of the NF-κB transcription factor.Science 242, 540–546 (1988).

26. Maier, J.A., Voulalas, P., Roeder, D. & Maciag, T. Extension of the life-span of humanendothelial cells by an interleukin-1 alpha antisense oligomer. Science 249, 1570–1574(1990).

27. Beg, A.A. & Baltimore, D. An essential role for NF-κB in preventing TNF-—induced celldeath. Science 274, 782–789 (1996).

28. Krohn, J. Expression of factor VIII-related antigen in human aqueous drainage chan-nels. Acta Ophthalmol. Scand. 77, 9–12 (1999).

29. Foets, B., van den Oord, J., Engelmann, K. & Missotten, L. A comparative immunohis-tochemical study of human corneotrabecular tissue. Graefes Arch. Clin. Exp.Ophthalmol. 230, 269–74 (1992).

30. Whitcup, S.M., Wakefield, D., Li, Q., Nussenblatt, R.B. & Chan, C.C. Endothelial leuko-cyte adhesion molecule-1 in endotoxin-induced uveitis. Invest. Ophthalmol. Vis. Sci. 33,2626–2630 (1992).

31. Bacon, A.S. et al. Adhesive molecules and relationship to leukocyte levels in allergic eyedisease. Invest. Ophthalmol. Vis. Sci. 39, 322–330 (1998).

32. Pappa, A., Calder, V., Fells, P. & Lightman, S. Adhesion molecule expression in vivo onextraocular muscles (EOM) in thyroid-associated ophthalmopathy (TAO). Clin. Exp.Immunol. 108, 309–313 (1997).

33. Loppnow, H. & Libby, P. Functional significance of human vascular smooth musclecell-derived interleukin 1 in paracrine and autocrine regulation pathways. Exp. Cell Res.198, 283–90 (1992).

34. Schreck, R., Rieber, P. & Baeuerle, P.A. Reactive oxygen intermediates as apparentlywidely used messengers in the activation of the NF-κB transcription factor and HIV-1.EMBO J. 10, 2247–2258(1991).

35. Cook, J.R., Mody, M.K. & Fini, M.E. Failure to activate transcription factor NF-κB incorneal stromal cells (keratocytes). Invest. Ophthalmol. Vis. Sci. 40, 3122–3131 (1999).

36. Kumar, S., Millis, A. J. T. & Baglioni, C. Expression of interleukin 1-inducible genes andreduction of interleukin 1 by aging human fibroblasts. Proc. Natl. Acad. Sci. USA 89,4683–4687 (1992).

37. Kee, C. & Seo, K. The effect of interleukin-1alpha on outflow facility in rat eyes. J.Glaucoma 6, 246–249 (1997).

38. Alexander, J.P., Samples, J.R. & Acott, J.S. Expression of matrix metalloproteinases andinhibitor by human trabecular meshwork. Curr. Eye Res. 17, 276–85 (1998).

39. Diamond, M.S., Staunton, D.E., Marlin, S.D. & Springer, T.A. Binding of the integrinMac-1 (CD11b/CD18) to the third immunoglobulin-like domain of ICAM-1 (CD54)and its regulation by glycosylation. Cell 65, 961–971 (1991).

40. Schuman, J.S., Wang, N. & Eisenberg, D.L. Expression of cell adhesion molecules in thetrabecular meshwork and aqueous outflow pathways. in Advances in OcularImmunology. (eds. Nussenblatt, R.B. Whitcup, S.M., Caspi, R.R. & Gery, I.) 47–50(Elsevier, Amsterdam, The Netherlands, 1994).

41. Lynch, M.G., Peeler, J.S., Brown, R.H. & Niederkorn, J.Y. Expression of HLA Class I and IIantigens on cells of the human trabecular meshwork. Ophthalmology 94, 851–857(1987).

42. Fini, M.E., Girard, M.T., Matsubara, M. & Bartlett, J.D. Unique regulation of the matrixmetalloproteinase, gelatinase B. Invest. Ophthalmol. Vis. Sci. 36, 622–636 (1995).

©20

01 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://m

edic

ine.

nat

ure

.co


document

Documents