8
Abstract Purpose. Elevated intraocular pressure in those with glau- coma appears to be a function of increased resistance to movement of aqueous humor through the conventional outflow pathway. The majority of resistance in both normal and glaucomatous eyes is generated in the region between the juxtacanalicular trabecular meshwork and the inner wall of Schlemm’s canal. To accommodate transient elevations in pressure, we hypothesize that conventional outflow increases rapidly due to changes in complexity of intercellular junc- tions between cells of the inner wall of Schlemm’s canal. Methods. To test this hypothesis we examined specifically the effects of hydrostatic pressure gradients and the calcium chelator, Na 2 EDTA, on permeability of cultured human Schlemm’s canal cell monolayers in isolation. Human Schlemm’s Canal cells were isolated, cultured and then seeded onto permeable supports and maintained in culture to allow intercellular junctions to mature. With a minimum net transendothelial electrical resistance of 10 Ohm cm 2 , cells were placed into an Ussing-type chamber and hydraulic con- ductivity was calculated from pressure and flow measure- ments that were continuously recorded. Simultaneously, transendothelial electrical resistance was measured manually at fixed intervals. In parallel experiments, cell margins were monitored in real time by videomicroscopy. Results. During the baseline measurement period when cells were exposed to pressure but not Na 2 EDTA, hydraulic con- ductivity was constant but transendothelial electrical resis- tance decreased continuously at rate of 0.24 W cm 2 /minute. After Na 2 EDTA treatment, no significant change in transendothelial electrical resistance was measured while, hydraulic conductivity of Schlemm’s Canal monolayers increased significantly by 125%; corresponding to noticeable intercellular separations. Restoration of cell-cell contact was observed by videomicroscopy 30 minutes following washout of Na 2 EDTA and functionally after 2 hours. Conclusions. Responses of Schlemm’s Canal cells to pres- sure and calcium chelators in vitro are consistent with a role for calcium sensitive junctions in outflow resistance in vivo. Keywords: glaucoma; aqueous humor; outflow facility; intercellular junctions; Ussing Introduction Glaucoma, the second leading cause of blindness in the United States, is a group of disorders characterized by loss of vision that in most cases is coincident with an increase in intraocular pressure (IOP). 1–3 While the underlying patho- physiological parameters that directly contribute to elevated IOP in most glaucomatous eyes are presently unknown, elevated IOP appears to be a function of defective cellular mechanisms in the conventional outflow pathway, resulting in decreased outflow facility. 4–7 Defects likely occur at the primary site of resistance in the conventional outflow pathway, at or near the inner wall of Schlemm’s canal. 8,9 Residing in this region are two cell types, trabecular mesh- work and Schlemm’s canal, that mediate resistance to outflow. Thus, resistance is likely a function of extracellular Received: July 22, 2003 Accepted: December 5, 2003 Correspondence: W. Daniel Stamer, PhD, Associate Professor, Departments of Ophthalmology and Pharmacology, The University of Arizona, 655 North Alvernon Way, Suite 108, Tucson, AZ 85711-1824, USA. Tel: 520-626-7767; Fax: 520-321-3665; E-mail: [email protected] Effect of hydrostatic pressure gradients and Na 2 EDTA on permeability of human Schlemm’s canal cell monolayers A.G. Burke 1 , W. Zhou 1 , E.T. O’Brien 3 , B.C. Roberts 4 and W.D. Stamer 1,2 Departments of 1 Ophthalmology and 2 Pharmacology, University of Arizona, Tucson, AZ; 3 University of North Carolina, Chapel Hill, NC; 4 Department of Pathology, Duke University, Durham, NC, USA Current Eye Research 2004, Vol. 28, No. 6, pp. 391–398 0271-3683/04/2806-391$22.00 © 2004 Taylor & Francis Ltd. Curr Eye Res Downloaded from informahealthcare.com by University of California Irvine on 10/29/14 For personal use only.

Effect of hydrostatic pressure gradients and Na 2 EDTA on permeability of human Schlemm's canal cell monolayers

  • Upload
    wd

  • View
    213

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

Abstract

Purpose. Elevated intraocular pressure in those with glau-coma appears to be a function of increased resistance tomovement of aqueous humor through the conventionaloutflow pathway. The majority of resistance in both normaland glaucomatous eyes is generated in the region betweenthe juxtacanalicular trabecular meshwork and the inner wallof Schlemm’s canal. To accommodate transient elevations inpressure, we hypothesize that conventional outflow increasesrapidly due to changes in complexity of intercellular junc-tions between cells of the inner wall of Schlemm’s canal.

Methods. To test this hypothesis we examined specifically the effects of hydrostatic pressure gradients and the calciumchelator, Na2EDTA, on permeability of cultured humanSchlemm’s canal cell monolayers in isolation. HumanSchlemm’s Canal cells were isolated, cultured and thenseeded onto permeable supports and maintained in culture toallow intercellular junctions to mature. With a minimum nettransendothelial electrical resistance of 10Ohm cm2, cellswere placed into an Ussing-type chamber and hydraulic con-ductivity was calculated from pressure and flow measure-ments that were continuously recorded. Simultaneously,transendothelial electrical resistance was measured manuallyat fixed intervals. In parallel experiments, cell margins weremonitored in real time by videomicroscopy.

Results. During the baseline measurement period when cellswere exposed to pressure but not Na2EDTA, hydraulic con-ductivity was constant but transendothelial electrical resis-tance decreased continuously at rate of 0.24 Wcm2/minute.After Na2EDTA treatment, no significant change in

transendothelial electrical resistance was measured while,hydraulic conductivity of Schlemm’s Canal monolayersincreased significantly by 125%; corresponding to noticeableintercellular separations. Restoration of cell-cell contact wasobserved by videomicroscopy 30 minutes following washoutof Na2EDTA and functionally after 2 hours.

Conclusions. Responses of Schlemm’s Canal cells to pres-sure and calcium chelators in vitro are consistent with a rolefor calcium sensitive junctions in outflow resistance in vivo.

Keywords: glaucoma; aqueous humor; outflow facility;intercellular junctions; Ussing

Introduction

Glaucoma, the second leading cause of blindness in theUnited States, is a group of disorders characterized by lossof vision that in most cases is coincident with an increase inintraocular pressure (IOP).1–3 While the underlying patho-physiological parameters that directly contribute to elevatedIOP in most glaucomatous eyes are presently unknown, elevated IOP appears to be a function of defective cellularmechanisms in the conventional outflow pathway, resultingin decreased outflow facility.4–7 Defects likely occur at theprimary site of resistance in the conventional outflowpathway, at or near the inner wall of Schlemm’s canal.8,9

Residing in this region are two cell types, trabecular mesh-work and Schlemm’s canal, that mediate resistance tooutflow. Thus, resistance is likely a function of extracellular

Received: July 22, 2003Accepted: December 5, 2003

Correspondence: W. Daniel Stamer, PhD, Associate Professor, Departments of Ophthalmology and Pharmacology, The University of Arizona, 655 North Alvernon Way, Suite 108, Tucson, AZ 85711-1824, USA. Tel: 520-626-7767; Fax: 520-321-3665; E-mail: [email protected]

Effect of hydrostatic pressure gradients and Na2EDTA on

permeability of human Schlemm’s canal cell monolayers

A.G. Burke1, W. Zhou1, E.T. O’Brien3, B.C. Roberts4 and W.D. Stamer1,2

Departments of 1Ophthalmology and 2Pharmacology, University of Arizona, Tucson, AZ; 3University of North Carolina,Chapel Hill, NC; 4Department of Pathology, Duke University, Durham, NC, USA

Current Eye Research2004, Vol. 28, No. 6, pp. 391–398

0271-3683/04/2806-391$22.00 © 2004 Taylor & Francis Ltd.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 2: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

392 A.G. Burke et al.

matrix turnover by TM cells, intercellular junction complex-ity between SC cells and/or dynamic interactions between SCand TM cells.7,10–13

Several recent studies using a variety of differentapproaches have provided six lines of evidence that supportthe idea that the majority of resistance to outflow resides atthe level of junctions between Schlemm’s canal cells. First,preferential occlusion of intercellular junctions by cationicbut not anionic ferritin in anterior chamber perfusions dra-matically decrease outflow facility.14 Second, the complexityof intercellular junctions of SC’s inner wall decrease inresponse to increased intraocular pressure.12 Third, openingsbetween SC cells are more apparent under high rather thanlow intraocular pressure in vivo.15 Fourth, intracellular levelsof a second messenger, cAMP, change in a manner that is proportional to hydrostatic pressure gradients across SCmonolayers in vitro and dependent upon mature and intactcell-cell contacts.16 Fifth, expression levels of zonula occlu-dens-1, – an associated protein of calcium-sensitive junc-tions, influence the permeability of SC monolayers in vitro.17

Sixth, calcium chelating agents increase outflow facility in areversible manner in both cat and monkey eyes in vivo;18,19

however, because only intact outflow pathways were evalu-ated, the specific contribution of SC cells to outflow resis-tance is unknown.

The purpose of the present study was to test the hypothe-ses that Na2EDTA effects in vivo are at the level of SC andthat disruption/reformation of cell-cell junctions at level of SC represent a likely mechanism for outflow tissues torespond to transient changes in pressure. To examine thesepossibilities, we measured the effects of hydrostatic pressuregradients and Na2EDTA on intercellular junctions of primarycultures of SC cell monolayers in real time in vitro. Primarycultures of human SC cells on permeable supports weremounted in a Ussing-type chamber and two physiologicalparameters, transendothelial electrical resistance (TEER) andhydraulic conductivity, were measured in real time before,during, and after exposure to hydrostatic pressure or hydro-static pressure plus Na2EDTA and compared to changes incell-cell associations using videomicroscopy.

Materials and methods

Cell types and culture

SC cells

Human cadaveric eye tissue was obtained from DonorNetwork of Arizona, San Diego Lions Eye Bank or NorthCarolina Lions Eye Bank within 48 hours of death for wholeeyes stored in moist chambers, and within 96 hours for non-transplantable corneal anterior segments stored in Optisol(Chiron Vision, Clairmont, CA). Human SC cells were iso-lated from the cadaveric eye tissue using gelatin-coated cannulas, as described previously20 or by differential dissec-tion.21 Cells used in perfusion studies were isolated by both

methods while only cells isolated by differential dissectionwere used in morphology experiments. Cells were grown andmaintained in Dulbecco’s Modified Eagle Medium (DMEM)and maintained in humidified air containing 5% CO2 at 37°C,as described previously.20 Human SC cells were seeded at100,000 cells/cm2 onto 1-cm2 Snapwell Filters (Snapwell;Costar, Acton, MA) with 0.4-mm pore diameter in 10%FBS/DMEM. Primary cultures of eight SC cell strains wereisolated from eight different cadaveric eyes without a historyof glaucoma and used in the present study (SC3, SC6, SC10,SC11, SC41, SC42, SC 57 and SC68).

MDCK cells

Madin-Darby Canine Kidney cells (MDCK), were used as acontrol to characterize perfusion system and for comparisonsto SC monolayers. Responses of calcium-sensitive junctionalcomplexes between MDCK cells are well documented22 andwere the most consistent in preliminary feasibility experi-ments that evaluated the resolution of the perfusion systemto changes in TEER and HC (compared to bovine aorticendothelial cells and calf pulmonary aortic endothelial cells).MDCK cells thus served as a positive control for compar-isons with SC cells in subsequent experiments. MDCK Cellsused in the present study were a gift from Dr. Ronald Lynch(University of Arizona) and were used between passages14–16. Cells were grown in DMEM and maintained inhumidified air containing 5% CO2 at 37°C for at least threeweeks after reaching confluence before experimentation.Cells that were past five weeks post-confluence were not usedbecause HC and TEER were not responsive to EDTA duringthe time course of the experiments.

Measurement of transendothelial electrical resistanceand hydraulic conductivity

Transendothelial electrical resistance (TEER) of cell mono-layers cultured on Snapwell filters was measured using aTEER measurement chamber (ENDOHM-24; World Preci-sion Instruments) in conjunction with an epithelial voltohm-meter (EVOM; World Precision Instruments). Followingthree rinses with prewarmed 20mM HEPES (Sigma, St.Louis, MO) buffered DMEM (serum-free, pH 7.4; HEPES-DMEM), filters were carefully transferred to the TEER measurement chamber filled with HEPES-DMEM. Thebackground electrical resistance of medium and insert wasmeasured (typically 12–13 Wcm2) and subtracted from mea-surements of filters with cells in medium. Cell monolayerswere excluded from the present study if they did not exhibitmature intercellular junctional complexes, as evidenced by aminimum net TEER of 10 Wcm2 after three weeks at conflu-ence. We chose this level of maturity because monolayersless resistive consistently failed to withstand control chamberexchanges and/or pressure gradients of 5mmHg. Further,cell monolayers cultured on filters greater than five weekshad greater TEER, but consistently lifted off of filters during

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 3: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

SC cell permeability 393

control chamber exchanges. Based upon experience withhuman umbilical and bovine aortic endothelial cells onsimilar filters, weak adherence was likely the result of cellshaving of a greater reliance on cell-cell associations ratherthan cell-matrix associations over time (Ronald Heimark,personal communication).

The Snapwell filters containing the cell monolayers werethen carefully placed into an Ussing-type chamber (Ussingsystem CHM5; World Precision Instruments, Sarasota, FL)filled with HEPES-DMEM and maintained at 37°C in an airincubator for 15 minutes prior to pressure application. Filterswere oriented such that the apical surface of cell monolayersfaced the upstream chamber. Flow through cell monolayerswas a function of the pressure gradient between the upstreamand downstream compartments of the chamber. Pressure inthe upstream compartment was generated by an elevatedreservoir of HEPES-DMEM, with pressure recorded using a pressure transducer (AH 60-3002; Harvard Apparatus, Holliston, MA), while the downstream compartment wasvented to atmospheric pressure. The rate of fluid flow throughthe chamber was calculated from the weight change per unittime of the reservoir measured using an isometric transducer(AH 60-2994; Harvard Apparatus). Pressure and weight datawere recorded simultaneously at 30 times/sec onto hard driveof computer using a Data Acquisition System (MP100WSW;Biopac Systems, Inc., Santa Barbara, CA).

Simultaneously, TEER was recorded manually at fixedintervals using the epithelial voltohmmeter and 3M KCl-filled electrodes (DRIREF-L; WPI) positioned in measure-ment ports on either side of cell monolayers in theUssing-type chamber. Cell monolayers were excluded if theydid not exhibit 15 minutes of stable TEER measurement priorto exposure to pressure.

Once stable TEER was achieved, a valve between thereservoir and chamber was opened to expose the upstreamchamber to pressure. TEER, flow, and pressure measure-ments were recorded for 20 minutes to establish baselinevalues. Cell monolayers were excluded if the flow rate wasgreater than 30ml/min or increased dramatically between the baseline measurements, indicating a compromised cellmonolayer. The medium in the upstream chamber was then exchanged with 5.0 mL HEPES-DMEM at a rate of 2.5ml/min, and a second 10-minute baseline period wasmeasured. An upstream chamber exchange using 5.0ml ofNa2EDTA in HEPES-DMEM was then performed, followedby a five-minute measurement period during which TEER was measured once per minute. The upstreamchamber was then rinsed with 5.0ml HEPES-DMEM, andTEER was measured once per minute until no change wasobserved for three successive measurements. Hydraulic con-ductivity was measured continuously and TEER measure-ments were then taken every five minutes for the duration of the trial. Following completion of TEER and hydraulicconductivity measurements, TEER of cell monolayers wasmeasured using a TEER measurement chamber for compar-ison with pre-trial TEER.

Calculation of hydraulic conductivity

Hydraulic conductivity was calculated in real time by Biopacdata acquisition software using pressure and flow measure-ments. The formula used to compute hydraulic conductivitywas:

HC = Q/(P * A)

Where Q is the volumetric flow rate across cell monolayers(mL/min), P is pressure (mmHg), and A is the area of a Snapwell filter (1.13cm2).

Since pressure and area are constant in the experimentalparadigm, HC is dependent upon Q. Pressure applied to themonolayer is contingent upon the height of a reservoir thatempties as fluid moves across cell monolayers. The initialpressure for all experiments is 6mmHg and decreases grad-ually during the experiment. Q is calculated from change inweight of column of fluid over time.

SC monolayer morphology

Images were taken from a time-lapse video recording of SCor MDCK cells just before and after the flow of buffered 10mM Na2EDTA reduced the extracellular calcium concen-tration. A 25¥ phase contrast objective, 4¥ relay magnifica-tion, real time noise reduction and contrast enhancement,Dage videocamera and a 36° microscope chamber allowedus to visualize cell responses to calcium removal in real time.A flow chamber allowed the rapid exchange of buffermedium with or without EDTA with only a brief (5–8 sec)loss of imaging. The microscope was enclosed in an insu-lated chamber and warmed to a constant 36° with infraredheat lamps.

The initial image was obtained with a Dage Newviconvideo camera, and then processed to remove noise and toincrease contrast using a Hamamatsu real time image processing system (Argus 10). The limit of resolution for this technique was ~300nm. Images were then stored on aPanasonic time-lapse video recorder at about a 60-fold time-lapse factor.

Statistical analyses

Paired t-tests were used to compare rate of TEER changeduring the baseline measurement period versus rate of TEERchange during Na2EDTA treatment for SC cell monolayersand for MDCK cell monolayers. A paired t-test was also usedto determine whether the percent change between the base-line measurement of HC (mean of two measurements) andthe measurement of HC following Na2EDTA treatment wasgreater than the percent change in HC between the first andsecond baseline measurements. Similarly, a paired t-test wasused to determine whether the percent change between thebaseline measurement of HC (mean of two measurements)and the measurement of HC following washout of Na2EDTAwas greater than the percent change in HC between the firstand second baseline measurements.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 4: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

394 A.G. Burke et al.

Results

Preliminary experiments were conducted to optimize themeasurement of Na2EDTA effects and pressure gradients oncell monolayers. Titration experiments revealed that a that aminimum of 25mM Na2EDTA was required to effect bothhydraulic conductivity and TEER of MDCK cell monolayersin the time course of the experiments, while only 5mMNa2EDTA was needed for SC monolayers (n = 13–15, datanot shown). Lower concentrations of Na2EDTA were tested(0.5–2mM) and found not to impact significantly HC orTEER of SC monolayers in the time course of the experi-ment (n = 6). The optimal amount of time that was requiredto observe consistent changes was 5 minutes. The optimalrange of pressure differentials across cell monolayers toresolve changes in hydraulic conductivity was 14–16mmHgfor MDCK and 4–6mmHg for SC. Higher pressures for eachcell type consistently compromised integrity of cell mono-layers and produced erratic data (n = 5–8).

TEER effects hydrostatic pressure gradients

The effects of hydrostatic pressure gradients (4–6mmHg forSC and 14–16mmHg for MDCK) on TEER were analyzedand compared. As shown in Figure 1, TEER in both SC and MDCK cell monolayers was stable at atmospheric pressure. Upon exposure to a hydrostatic pressure gradient,TEER in both cell monolayers decreased. Figure 2a showsaverage TEER changes for both SC and MDCK cell mono-layers during a 30-minute measurement period. TEER of SCcell monolayers decreased at an average rate of 0.24 (±0.05SEM) Wcm2/min (n = 5) while TEER of MDCK cell mono-layers decreased at a rate of 0.3 (±0.12 SEM) W/min (n = 4).TEER changes due to exposure to hydrostatic pressure forboth SC and MDCK cell monolayers were significantly dif-ferent from baseline TEER measurements taken in theabsence of pressure (p < 0.05).

P P

EDTA

EDTA

30

35

40

45

50

55

60

65

-20 0 20 40 60

Time (min)

TE

ER

(O

hms

cm2 )

10

15

20

25

30

35

-20 0 20 40 60

Time (min)

TE

ER

(O

hms

cm2 ) A B

Figure 1. Effect of hydrostatic pressure and Na2EDTA on transendothelial electrical resistance across cell monolayers. After 15 minutes ofstable TEER at no pressure, either human Schlemm’s canal cells (panel A) or MDCK cells (panel B) were exposed to hydrostatic pressure (P= 4–6 mm Hg for SC and 14–16 mm Hg for MDCK). At the 20-minute mark, chambers containing cell monolayers were exchange with normalmedium (chamber exchange). At the 30-minute mark, chambers containing cell monolayers were exchanged with medium containingNa2EDTA (panel A, 5 mM and panel B, 25 mM). At the 35-minute mark, chambers containing cell monolayers were exchanged with normalmedium (rinse). Shown is one experiment for each cell type of 4–5 total experiments.

Cell Type/Pressure (mmHg)SC/5 MDCK/15

dR/d

t (O

hms/

min

)

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

**p < 0.05*Baseline

A

Cell Type/Pressure (mmHg)SC/5 MDCK/15

dR/d

t (O

hms/

min

)

-4

-3

-2

-1

0

Na2EDTA Treatment

B

Figure 2. Effect of hydrostatic pressure on change in transendothelial electrical resistance (TEER) across cell monolayers. Shown as his-tograms are the mean changes (±SEM) in TEER across Schlemm’s Canal (SC) and MDCK cell monolayers during exposure to hydrostaticpressure either in the absence (panel A; 15 min. baseline) or presence (panel B) of Na2EDTA (5 mM for SC and 25 mM for MDCK for 5 min;measurement period = 5 min. following exposure). Data represent combined data from 4–5 experiments for each cell type.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 5: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

SC cell permeability 395

TEER effects of Na2EDTA

Exchange of chamber with normal medium did not affect rateof TEER change. However after Na2EDTA treatment, TEERdecreased at a new average rate of 1.9 (±0.12 SEM) Wcm2/min (n = 4) in MDCK cell monolayers and 0.34 (±0.26SEM) Wcm2/min (n = 5) in SC cell monolayers (Figs. 1 and2b). Figure 1 shows effect of Na2EDTA relative to other treat-ments and 2b shows histogram of net average TEER changesfor both SC and MDCK cell monolayers after Na2EDTAtreatment. Paired t-tests showed no significant change inTEER decrease for SC cell monolayers relative to the 30-minute baseline measurement period in the presence of pres-sure. In contrast, MDCK cell monolayers showed a changein TEER that approached significance (p = 0.06). Followingrinse with normal medium, average rate of change of TEERfor both SC and MDCK cell monolayers returned to levelsnot significantly different from pre-Na2EDTA rates for theduration of the testing period.

Effects of Na2EDTA on hydraulic conductivity

Figure 3 shows values from individual experiments com-paring hydraulic conductivity before (baseline), during(EDTA), and after (rinse) treatment with Na2EDTA. SC cellmonolayers were exposed to 5mM Na2EDTA (driven by 4–6mmHg gradient), while MDCK cell monolayers were exposed to 25 mM Na2EDTA (driven by 14–16mmHggradient) for 5 minutes.

During the 30-minute baseline measurement period,MDCK cell monolayers exhibited an average stable hydraulicconductivity of 0.038 ml/min/mmHg/cm2, compared to 2.44ml/mmHg/cm2 in SC cell monolayers. This level is con-sistent with permeability measurements from SC monolay-ers in vitro by others and about 10-fold greater thancalculated in vivo.17,23,24 As expected, baseline HC measure-ments were dependent upon initial TEER measurement for

monolayer (Table 1). Following a 5-minute Na2EDTA treat-ment, hydraulic conductivity of the five SC cell monolayersfrom four different cell strains tested increased 68%, 94%,116%, 119%, and 230% relative to baseline. Hydraulic con-ductivity of the four MDCK monolayers increased 128%,285%, 1664%, and 346% relative to baseline. Followingchamber exchange with normal medium, hydraulic conduc-tivity of SC cell monolayers continued to increase (123%,156%, 406%, 505%, and 454%) relative to baseline, whileMDCK cell monolayers exhibited a decrease in hydraulicconductivity to 17%, -35%, 1137%, and 12% relative tobaseline during the 15 minute baseline period (Table 1). Intwo additional experiments with SC monolayers, HC wasexamined two hours after Na2EDTA challenge in the presence of a pressure gradient and found to return to nearbaseline levels (58% and -39% relative to baseline).

0

5

10

15

20

25

30

35

40

Baseline EDTA Rinse

HC

(uL

/min

/mm

Hg/

cm2 )

A B

0

0.1

0.2

0.3

0.4

0.5

0.6

Baseline EDTA Rinse

HC

(uL

/min

/mm

Hg/

cm2 )

SC MDCK

Figure 3. Effect of Na2EDTA on hydraulic conductivity (HC) of cell monolayers. HC was calculated from continuous recording of pressureand flow across cell monolayers. After 30 minutes of stable HC across cell monolayers both before and after a chamber exchange with normalmedium, SC (panel A) or MDCK (panel B) cells were subjected to medium containing Na2EDTA (5 mM for SC and 25 mM for MDCK).After 5 minutes of exposure, cells were rinsed with normal medium and HC was evaluated for 15 minutes. Monitoring of HC continued uninterrupted. Each line represents one independent experiment.

Table 1. Hydraulic conductivity measurements of cell monolayers.

Cell Net Baseline Post-EDTA Post-RinseLine TEER HC HC HC

SC 10 10 5.79 10.79 14.22SC 10 15 0.47 0.95 2.63SC 11 10 5.25 18.14 30.44SC 41 14 0.08 0.18 0.42SC 43 12 0.61 1.02 1.36MDCK 76 0.053 0.12 0.062MDCK 78 0.058 0.22 0.038MDCK 64 0.0062 0.11 0.077MDCK 53 0.034 0.15 0.038

Transendothelial electrical resistance (TEER; W cm2) measurementsof cell monolayers were made using EndOhm chamber immediatelyprior to exposure to pressure gradient. Units for baseline, EDTA andRinse hydraulic conductivity (HC) measurements are ml/min/mmHg/cm2.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 6: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

396 A.G. Burke et al.

Percent change in HC relative to baseline followingNa2EDTA treatment was significantly greater than percentchange of HC between the two baseline measurements in SC cells (p < 0.05) and in MDCK cells (p < 0.05). Percentchange in HC relative to baseline following post-Na2EDTArinse with normal medium was significantly greater thanpercent change between the two baseline measurements inSC cells (p < 0.05) but not in MDCK cells.

Morphological effects of Na2EDTA

In parallel experiments, effects of Na2EDTA on cell marginswere visualized en face using phase-contrast videomi-croscopy on glass coverslips. Figure 4 shows still images cap-tured from video sequences of live SC cell monolayers beforeand after Na2EDTA treatment. Panel A shows the edge of 2adjacent SC cells before exposure to Na2EDTA. The tips ofthe arrows show the margins of the cells. The cell marginsappeared to be touching to the left of arrows, and wereslightly separated (<0.5mm) at the arrows. Separations likethis were uncommon, however these areas were chosen toimage because it enabled locating and tracking of cellborders during the experiment. Panels B and C show thesame field of view 60 and 90 seconds after the addition ofNa2EDTA. The cell margins retracted about 2mm from eachother in B, and about 3mm in C. Fine “retraction fibers” areevident in the developing space between cell margins.MDCK cell monolayers were also evaluated before and afterNa2EDTA treatment. MDCK cell borders retracted from abelt-like arrangement of fibers near, but off the basal surfaceof the cells by 60 seconds (data not shown).

Consistent with perfusion studies, cell margins thatretracted as a result of exposure to Na2EDTA did not imme-diately come back together upon removal of Na2EDTA. Infact, when exposed to 10mM Na2EDTA for 5 minutes, about30 minutes were required for the margins of the cells returnto their original position after replacement of normal medium(data not shown). By time lapse, the cells appeared to follow

the retraction fibers left behind during retraction (data notshown).

Discussion

The results of the present study show that permeability ofcultured human SC cells respond to physiologically relevanthydrostatic pressure gradients and to the calcium chelator,Na2EDTA in a manner similar to that previously observed in situ12 and in vivo, respectively.18,19 In cultured SC cellmonolayers, transendothelial electrical resistance signifi-cantly decreased in response to hydrostatic pressure, andhydraulic conductivity significantly increased in response toNa2EDTA. These data support the hypothesis that calcium-sensitive intercellular junctions of Schlemm’s Canal con-tribute in part to the generation of resistance for aqueoushumor outflow and may serve to relieve transient increasesin intraocular pressure.

Previous research has shown that complexity of intercel-lular junctions between inner wall cells of Schlemm’s Canal(as indicated by fewer tight junctional strands) decreases inresponse to elevated pressure.12 Effect of pressure gradientsof 0 (atmospheric), 15 and 45mmHg were assessed. In thepresent study, a hydraulic pressure gradient of 5mmHg wasevaluated and found to significantly decrease transendothe-lial electrical resistance (TEER) measurements across SCcell monolayers. Since TEER is an indicator of junctionalcomplexity between cells, these functional in vitro results areconsistent with those found morphologically in situ. Suchrapid changes in junctional complexity may occur in combi-nation with increased giant vacuole formation serve toaccommodate transient changes in intraocular pressure thatoccur routinely due to eye rubbing or ocular pulsations.7,25

While pressure-dependent changes in giant vacuoles are welldocumented, data indicate that pore density of inner wallcells is not pressure-dependent (see Ethier for review).13 Thissuggests that intercellular pores contribute to the long-termpermeability of the inner wall while complexity of intercel-

Figure 4. Effect of Na2EDTA on cell-cell associations of Schlemm’s canal monolayers. Shown in panel A is a phase-contrast image of cul-tured SC cell monolayers obtained by video capture before treatment with Na2EDTA. Cells were exposed to 10 mM Na2EDTA for 5 minutesand rinsed with normal medium and were visualized continuously. Shown in panels B and C is the same field as in panel A, sixty and ninetyseconds after treatment, respectively. The tips of arrows indicate cell margins. Retraction fibers are seen between the arrow tips in panels Band C. Shown is one representative experiment of 10 total. Images in all panels were videocaptured at the same magnification.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 7: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

SC cell permeability 397

lular junctions and/or formation of vacuoles may contributeto short-term adaptations.

Chelation of calcium has been shown to increase aqueoushumor outflow facility in vivo in two different animalmodels.18,19,26 The concentration of Na2EDTA used with SCmonolayers in the present study (0.5–10mM) was consistentwith that used in these studies (0.5–6mM) and with the the-oretical concentration needed to buffer calcium in perfusionmedium (2.0mM) plus provide a sufficient gradient to pullcalcium out from intercellular junctions. The present studytested whether the effects that have been observed in vivo arepresent at the level of SC. The results showed that chelationof calcium by Na2EDTA significantly increased hydraulicconductivity across SC cell monolayers, presumably by lig-ating intercellular junctions. In addition, morphological find-ings showed a noticeable separation of cell-cell associationsfollowing Na2EDTA treatment, consistent with a decrease in complexity of cell-cell junctions. Given the hydraulic conductivity and morphological findings, a correspondingdecrease in TEER following Na2EDTA treatment wasexpected, but not measured. Failure to detect a consistent andsignificant change in TEER after Na2EDTA treatment in SCcell monolayers may have been the result of the initial pres-sure-driven decrease in junctional complexity, which reducedTEER to a level (basement effect) where further TEERdecreases cannot be resolved.

Increases in aqueous humor outflow due to calciumchelating agents have been shown to be reversible in vivo.18,19

In perfusion experiments, MDCK cell monolayers exhibiteda partial recovery of TEER and hydraulic conductivityshortly after washout of Na2EDTA. In contrast, hydraulicconductivity of SC cell monolayers took longer to approachbaseline levels following washout of Na2EDTA. To examinethis further, we followed SC monolayers after washout ofNa2EDTA for longer periods of time by videomicroscopy andnoticed that about 30 minutes were required for cell-cell con-tacts to begin to return visually and 2 hours functionally viaTEER measurements. The differences in recovery betweenMDCK (seconds) to SC (minutes to hours) are likely relatedto respective differences between MDCK and SC in initialjunctional complexity (60 versus 28 Wcm2), hydraulic con-ductivity (2.44 versus 0.038 ml/min/mmHg/cm2) and effectsof cell culture on the ability to form mature intercellular junc-tions. The differences observed between these two cell typeslikely reflect differing roles in vivo.

In conclusion, the results of the present study using cul-tured SC monolayers a) are similar to previous data usingintact outflow pathways and b) are consistent with the idea that calcium-sensitive intercellular junctions betweenSchlemm’s Canal cells are likely sites of action of hydrosta-tic pressure gradients, Na2EDTA and other calcium chelators.The recent documentation of the unique expression of vas-cular endothelial cadherin in human Schlemm’s canal cellsbut not trabecular meshwork cells in the human outflowpathway may explain these results.27 Vascular endothelialcadherin is an endothelial-specific, calcium-sensitive inter-

cellular junction protein that in part mediates permeability ofvascular endothelium28 and now represents a future thera-peutic target for regulation of aqueous humor outflow inpeople with glaucoma. The demonstration that the experi-mental model described in the present study can detect smallchanges in hydraulic conductivity and transendothelial elec-trical resistance supports the usefulness of this model inexamining efficacy of drugs that may interact with vascularendothelial cadherin or other potential targets in SC cellmonolayers.

Acknowledgments

The authors thank Ron Lynch, Ph.D. for providing MDCKcells; Velma Dobson, Ph.D. for helpful comments on themanuscript; and Eileen Ryan and Neil Atodaria, M.D. forassistance in data collection and characterization of model,respectively. This study was supported in part by Allergan,Inc., the Research to Prevent Blindness Foundation, andNational Eye Institute (EY12797 and EY13861).

References

1. Quigley H. Number of people with glaucoma worldwide.Br J Ophthalmol. 1996;80:389–393.

2. Schappert S. Office visits for glaucoma: U.S., 1991–1992.Hyattsville, MD: National Center for Health Statistics;1995:1–13.

3. Leske M. The epidemiology of open-angle glaucoma: Areview. Am J Epidemiol. 1983;118:166–191.

4. Rohen J. Why is intraocular pressure elevated in chronicsimple glaucoma? Anatomical considerations. Ophthalmol.1983;95:673–691.

5. Wiederholt M. Direct involvement of trabecular meshworkin the regulation of aqueous humor outflow. Curr OpinionOphthalmol. 1998;9:46–49.

6. Lutjen-Drecoll E. Functional morphology of the trabecularmeshwork in primate eyes. Prog Retinal Eye Res. 1998;18:91–119.

7. Johnstone M, Grant M. Pressure-dependent changes instructures of the aqueous outflow system of human andmonkey eyes. Am J Ophthalmol. 1973;75:365–383.

8. Bill A, Maepea O. Mechanisms and Routes of AqueousHumor Drainage. Philadelphia: WB Saunders; 1975:206–226.

9. Grant WM. Experimental aqueous perfusion in enucleatedhuman eyes. Arch Ophthalmol. 1963;69:783.

10. Johnson M. Aqueous outflow resistance (ISER Debate).International Glaucoma Review. 2003;43:446–447.

11. Acott T. Trabecular Extracellular Matrix Regulation. Baltimore: Williams and Wilkins; 1992:125–157.

12. Ye W, Gong H, Sit A, Johnson M, Freddo TF. Interen-dothelial junctions in normal human Schlemm’s canalrespond to changes in pressure. Invest Ophthalmol Vis Sci.1997;38:2460–2468.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.

Page 8: Effect of hydrostatic pressure gradients and Na               2               EDTA on permeability of human Schlemm's canal cell monolayers

398 A.G. Burke et al.

13. Ethier C. The inner wall of Schlemm’s canal. Exp Eye Res.2002;74:161–172.

14. Ethier C, Chan D. Cationic ferritin changes outflow facil-ity whereas anionic ferritin does not. Invest Ophthalmol VisSci. 2001;42:1795–1802.

15. Epstein DL, Rohen JW. Morphology of the trabecularmeshwork and inner-wall endothelium after cationized ferritin perfusion in the monkey eye. Invest Ophthalmol VisSci. 1991;32:160–171.

16. Stamer W, Roberts B, Epstein D. Hydraulic pressure stim-ulates adenosine 3¢, 5¢-cyclic monophosphate accumulationin endothelial cells from Schlemm’s canal. Invest Ophthal-mol Vis Sci. 1999;40:1983–1988.

17. Underwood J, Murphy C, Chen J, Franse-Carman L, WoodI, Epstein D, Alvarado J. Glucocorticoids regulatetransendothelial fluid flow resistance and formation ofintercellular junctions. Am J Physiol. 1999;277:C330–C342.

18. Warner D, Chu E. The role of ionic calcium in the resis-tance to aqueous outflow in the cat. Can J Ophthalmol.1967;2:226.

19. Bill A, Lutjen-Drecoll E, Svedbergh B. Effects of intra-cameral sodium EDTA and EGTA on aqueous outflowroutes in the monkey eye. Invest Ophthalmol Vis Sci.1980;19:492–504.

20. Stamer WD, Roberts BC, Howell DN, Epstein DL. Isolation, culture and characterization of endothelial cellsfrom Schlemm’s canal. Invest Ophthalmol Vis Sci. 1998;39:1804–1812.

21. O’Brien E, Ren X, Wang Y. Localization of myocilin to the

golgi apparatus in Schlemm’s canal cells. Invest Ophthal-mol Vis Sci. 2000;41:3842–3849.

22. Nelson W, Veshnock P. Modulation of fodrin (membraneskeleton) stability by cell-cell contact in Madin-Darbycanine kidney epithelial cells. J Cell Biol. 1987;104:1527–1537.

23. Alvarado JA, Murphy CG, Franse-Carman L, Chen J,Underwood JL. Effect of beta-adrenergic agonists on para-cellular width and fluid flow across outflow pathway cells.Invest Ophthalmol Vis Sci. 1998;39:1813–1822.

24. Johnson M, Erickson K. Mechanisms and routes of aqueoushumor drainage. In: Albert D, Jakobiec F, eds. Principlesand Practice of Ophthalmology. Philadelphia: WB Saunders Co.; 2000:2577–2595.

25. Grierson I, Lee W. Changes in the monkey outflow appara-tus at graded levels of intraocular pressure: A qualitativeanalysis by light microscopy and scanning electronmicroscopy. Exp Eye Res. 1974;19:21–33.

26. Hamanaka T, Bill A. Morphological and functional effectsof Na2EDTA on theoutflow routes for aqueous humor inmonkeys. Exp Eye Res. 1987;44:171–190.

27. Heimark R, Kaochar S, Stamer W. Human Schlemm’s canalcells express the endothelial adherens proteins, VE-cadherin and PECAM-1. Curr Eye Res. 2002;25:299–308.

28. Corada M, Mariotti M, Thurston G, Smith K, Kunkel R, Brockhaus M, Lampugnani M, Martin-Padura I, Stoppacciaro A, Ruco L, McDonald D, Ward P, Dejana E.Vascular endothelial-cadherin is an important determinantof microvascular integrity in vivo. Proc Natl Acad Sci USA.1999;96:9815–9820.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Cal

ifor

nia

Irvi

ne o

n 10

/29/

14Fo

r pe

rson

al u

se o

nly.