Abstract

Purpose. In our previous study, we showed that the AT1

receptor antagonist increased uveoscleral outflow (USF)when topically applied to the rabbit eye. However thisincrease was too small to demonstrate a clear physiologicalrole for ocular angiotensin II (AII). Hence, the purpose ofthis study was to determine whether ocular AII influencedUSF regulation, and if so, how this occurred.

Methods. USF was measured by the FITC-dextran perfusionmethod in albino rabbits. AII and its receptor antagonistswere directly applied into the anterior chamber by addinginto the perfusate and by perfusing with FITC-dextran. Wealso analyzed angiotensin receptors on the rabbit ciliary bodymembrane by a receptor binding assay with 125I-[Sar1, Ile8]-AII as a ligand.

Results. CS-088 (1mg/ml) increased USF by 24% while AII decreased USF in a concentration-dependent mannerbetween 10 and 500nM. Its maximum decrease of 19%occurred at 500nM. At this AII concentration the USF reduc-tion was antagonized by 1 mg/ml CS-088, an AT1-receptorantagonist, but not by the same concentration of PD-123,177,an AT2-receptor antagonist. Specific 125I-[Sar1, Ile8]-AIIbinding to the rabbit ciliary body membranes was inhibitedby CS-088 with an inhibition constant of 7.05nM, whereasinhibition by PD-123,177 was not observed.

Conclusions. Ocular AII was indicated to attenuate USF viaAT1 receptors in rabbits, however its physiological effect wasnot critical in IOP regulation.

Keywords: uveoscleral outflow; angiotensin AT1 receptor;ciliary muscle

Introduction

Intraocular pressure (IOP) is primarily regulated through the formation and drainage of the aqueous humor. Most ofaqueous humor is drained through the trabecular meshworkinto Schlemm’s canal which is connected to the episcleralvenous plexus (the conventional route). The remainingaqueous humor drains out of the eye through a route calleduveoscleral outflow (USF). Firstly, the aqueous humordirectly enters the ciliary muscle through the space betweenmuscle bundles and then flows into the space between thesclera and uvea (suprachoroidal space) toward the posteriorpart of the eye. Eventually, the fluid returns to the systemiccirculation via the lymphatic vessels in the orbit. The exis-tence and significance of this pathway was first reported byBill.1,2

The renin-angiotensin system (RAS) plays an importantrole in the control of systemic blood pressure and extracel-lular fluid volume. During the past decade the presence oflocal RAS has been shown in several tissues and its func-tional role in the local tissues has been implied. In the ante-rior part of the eye, the propeptide3,4 of AII, the enzymes4–12

necessary for AII production and receptors4,13 also have beenlocalized in the aqueous humor and surrounding tissues. Thissupports the idea that ocular RAS is present and must havea physiological role in the anterior part of the eye.

One possible role for ocular RAS could be to regulate IOPsince inhibition of RAS leads to an IOP decrease. Inhibitorsof the renin and angiotensin-converting enzyme have beendemonstrated to decrease IOP and a local but not systemicaction of the drug was indicated in some cases.14–17 Recently,non-peptide AT1 receptor selective antagonists were alsodemonstrated to decrease IOP. Losartan18 and CS-088 (in our previous report) showed ocular hypotensive effects in

Received: April 19, 2000Accepted: August 13, 2001

Correspondence: Tatsuya Inoue, Neuroscience and Immunology Research Laboratories, Sankyo Co. Ltd., 2–58, Hiromachi 1-chome, Shinagawa-ku, Tokyo 140-8710, Japan. Tel: 3-3492-3131, Fax: 3-5436-8560, E-mail: inotat@shina.sankyo.co.jp

The effect of angiotensin II on uveoscleral outflow in rabbits

Tatsuya Inoue1, Tomihisa Yokoyoma1 and Hiroyuki Koike2

Research Laboratories of 1Neuroscience and Immunology, and 2Pharmacology and Molecular Biology, Sankyo CompanyLimited, Tokyo, Japan

Current Eye Research 0271-3683/01/2302-139$16.002001, Vol. 23, No. 2, pp. 139–143 © Swets & Zeitlinger

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

he U

nive

rsity

of

Man

ches

ter

on 1

0/26

/14

For

pers

onal

use

onl

y.

140 T. Inoue et al.

humans and rabbits, respectively. In addition, CS-088showed an USF increasing effect and therefore the ocularRAS was initially considered to regulate USF. However the increased rate of USF was insufficient in determiningwhether it had a physiological role because the techniqueused to determine USF was not of satisfactory reliability forreasons discussed in our previous report.

In this report, we examined the effects of AII and itsantagonists on USF. Angiotensin receptors on rabbit ciliarybody were also analyzed by a receptor binding assay andthose were carried out to demonstrate the effects of ocularAII on USF.

Materials and methods

Chemicals and experimental animals

Male New Zealand White rabbits weighing 2.0–2.5kg wereused. The investigation was conducted in accordance with the ARVO Resolution on the Use of Animals in Research.CS-088, an AT1-receptor antagonist, and PD-123,177, an AT2 receptor antagonist, were obtained from the MedicinalChemistry Research Laboratories, Sankyo Co., Ltd. (Tokyo,Japan). 125I-[Sar1, Ile8]-angiotensin II (125I-[Sar1, Ile8]-AII,specific activity: 2200Ci/mmol) was purchased from theNew England Nuclear Corporation (Boston, MA).

Uveoscleral outflow determination

USF was determined in 26 rabbits according to the fluores-cein isothiocyanate (FITC)-dextran method.19 Rabbits werefirst anesthetized with urethan (1.5g/kg, iv). Indomethacinwas then administered to the rabbits both intraperitoneally(10mg/kg) and topically (Indomelol® eye drop, Senju Phar-maceutical Co., Ltd.) 3 hrs prior to the FITC-dextran per-fusion. It was administered to minimize the formation ofprostaglandins at the time of needle insertion, which mightthen influence the USF.19 Before the perfusion, one drop of0.4% oxybuprocaine was topically applied, and a 24-gaugeneedle was inserted into the rabbit eye through the cornea.After withdrawal of the aqueous humor, the needle was con-nected via a polyethylene tube to a reservoir filled with per-fusate containing 0.1M phosphate (pH 7.2) and 0.1mg/ml ofFITC-dextran (MW 70000, Sigma). The perfusate was thenintroduced into the anterior chamber for 45 min at a perfu-sion pressure of 40mmHg. When the perfusion was com-pleted, the rabbits were sacrificed with excess pentobarbital,and the eyes were enucleated. The eyes were then thoroughlywashed with physiological saline, and the conjunctiva andcornea were removed with ophthalmic scissors. The anteriorchamber was washed extensively with Dulbecco’s phosphate-buffered saline (D-PBS, Life Technologies, Inc.), and the lenswas removed. The remaining tissues were cut into smallpieces and homogenized in D-PBS with a polytron homo-genizer. The homogenate was centrifuged (3000 rpm, 4°C) and the resulting supernatant was further filtered through a

Millex-PF filter (0.8mm, Millipore Corporation). The con-centration of FITC-dextran in the filtrate was determined fluorometrically.

The USF value was calculated as follows:

Receptor binding analysis

Twenty-five rabbits were sacrificed with an excess of pento-barbital and the eyes were enucleated. The ciliary body was then isolated and homogenized with a polytron homogenizer in buffer containing 50mM Tris (pH 7.4) and4.9mM MgCl2. The fiber was then removed by centrifugalfiltration using a cell strainer (40mm, nylon, Falcon) at 3000g and 4°C for 5min. The filtrates were centrifuged at 40000g and 4°C for 15min. Then, the pellet was resuspended in the original buffer and used as a membranepreparation.

In the binding experiments, 50 mg protein/ml of the mem-brane preparation, 50–2000pM of 125I-[Sar1, Ile8]-AII, andvarious concentrations of drugs were added to the assaybuffer composed of 50 mM Tris-HCl (pH 7.4), 4.9 mMMgCl2, 2mg/ml bovine serum albumin and 1.8 mM phenyl-methylsulfonyl fluoride. Serum albumin was added into thebuffer as it does exist in the aqueous humor and is known to affect the binding of AII to its receptors. Therefore, byincluding serum albumin, the binding conditions resemblethat of physiological more. The mixture was then incubatedfor 2 hours at 37°C. When the incubation was terminated, the mixture was immediately filtrated through a GF/C filter(Whatman), presoaked with 0.3% polyethyleneimine, to sep-arate the membrane-bound radioligands from the free ones.After separation, the filter was washed with ice-cold 50mMTris-HCl (pH 7.4) in a cell harvester. The radioactivityremaining on the filter was counted using a gamma counter.Non-specific binding was determined by incubating themixture in the presence of 1mM unlabeled AII.

Statistical analysis

A paired t-test was applied for the statistical analysis of eachUSF experiment.

Results

Effects of CS-088 on uveoscleral outflow

CS-088, an AT1-receptor antagonist, added to the perfusateof the experimental eye increased USF by 24% compared tothe control (Fig. 1). The effect of CS-088 was not, however,statistically significant by a paired t-test (P = 0.06).

USF =quantity of FITC-dextran in tissue

concentration of FITC-dextran in perfusateperfusion period

( )( )

¥ ( )

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

he U

nive

rsity

of

Man

ches

ter

on 1

0/26

/14

For

pers

onal

use

onl

y.

Uveoscleral outflow and angiotensin II 141

125I-[Sar1, Ile8]-AII, was used as a ligand. In the receptoranalysis, angiotensin receptor density on the rabbit ciliarybody was determined to be 94.1 fmol/mg protein. Bindingwas thought to occur at a single-site as shown in Figure 3.CS-088 and losartan, a standard AT1-receptor antagonist,inhibited the binding of 125I-[Sar1, Ile8]-AII whereas PD-123,177 did not (Fig. 4). The inhibition constants of CS-088and losartan were determined to be 7.05 and 34.5 nM, respec-tively. Inhibition by PD-123,177 was observed only at highconcentrations of more than 100 mM. This was considered tobe a non-specific binding.

Discussion

In our previous report topical application of CS-088increased USF in rabbits. This increase was small, so afurther study was required to ascertain whether the result was

80

100

120

140

160

Control CS-088

USF

(%

of

cont

rol)

P = 0.06

Figure 1. Effect of CS-088 on uveoscleral outflow in rabbits. CS-088 (1 mg/ml) was perfused into the anterior chamber of the experimental eye. The contralateral eye was perfused with thevehicle as the control. Data are represented as mean ± SEM of 8measurements.

USF (%

of control)

USF

(%

of

cont

rol)

90

100

110

120

130

500 500 + C 500 + P

70

80

90

100

110

0 10 100 500

Concentration of angiotensin II (nM)

A B

P < 0.05 P < 0.05

Figure 2. Effect of angiotensin II (AII) on uveoscleral outflow (USF) in rabbits. AII of 10–500 nM was perfused into the anterior chamberof the experimental eye. The contralateral eye was perfused with the vehicle as the control (A). AII of 500 nM was perfused into the anteriorchamber of the experimental eye with 1 mg/ml of CS-088 (shown as C) or PD-123,177 (shown as P) while the contralateral eye was perfusedwith 500 nM of AII as the control (B). Data are represented as mean ± SEM of 3–5 (A), or 7 (B) measurements.

Effects of angiotensin II on uveoscleral outflow

AII was applied directly into the rabbit anterior chamber andthe results are shown in Figure 2A. AII decreased USF in aconcentration-dependent manner between 10–500nM. Amaximal decrease of 18.8% was observed at 500nM. Theeffects of AII were significant at 500 nM (P < 0.05). CS-088or PD-123,177, AII receptor subtype selective antagonists,were co-perfused with 500 nM AII while control eyes wereperfused with AII only (Fig. 2B). USF decrease was reversedby CS-088 (P < 0.05, 15.9% of USF recovery) but not byPD-123,177.

Receptor binding analysis on the rabbit ciliary body

Angiotensin receptors on the ciliary body, which is com-posed of the ciliary muscle and ciliary process, were studiedusing a receptor binding assay. A non-selective AII analogue,

0.00

0.01

0.02

0.03

0.04

0.05

40 50 60 70 80 90 100

B (fmol/mg protein)

B/F

Bmax: 94.1

(fmol/mg prot.)

Figure 3. Scatchard plot of specific 125I-[Sar1, Ile8]-angiotensin IIbinding to rabbit ciliary body membranes.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

he U

nive

rsity

of

Man

ches

ter

on 1

0/26

/14

For

pers

onal

use

onl

y.

142 T. Inoue et al.

Drug Ki value (nM)

CS-088 7.05

Losartan 34.5

PD-123,177 n.d.0

20

40

60

80

100

120

140

10 9 8 7 6 5 4 3

Drug concentration (-log M)

CS-088

Losartan

PD-123,177

125 I

-[Sa

r1, I

le8 ]

-AII

bin

ding

(%

)

Figure 4. Effect of angiotensin II receptor antagonists on specific binding of 125I-[Sar1, Ile8]-angiotensin II to rabbit ciliary body membranes.Data are represented as mean ± SEM of 3 determinations.

of physiological importance. Furthermore the antagonist wasadministered topically, hence it is possible that it may havebeen absorbed into the systemic circulation and subsequentlyaffected systemic RAS. In this study, experiments weredesigned to clarify the action of ocular AII. To exclude thepossibility of systemic influence the antagonist was directlyapplied into the anterior chamber. Additionally, experimentswere conducted using excess drug concentration (more than100-fold of the IC50 according to our receptor binding data)to attain a maximal effect of the antagonist. As a result, CS-088 was shown to increase USF by 24% (P = 0.06). Contraryto our expectations, the USF increase was still small eventhough the antagonist was applied in excess. This smallchange was observed consistently throughout our study. Thissupports the idea that ocular AII has a minor influence onUSF in rabbits under physiological conditions. However theeffect of AII is rather small to affect physiological IOP inrabbits since the USF accounts for 3–8% to the total outflowin this species.20,21

Apart from the experiment mentioned above, the effect ofexogenous AII was investigated to further distinguish theeffect of AII. Exogenously applied AII did decrease USF andthis effect was shown to be mediated by AT1 receptors.

The concentration of AII perfused in the present USFstudy was relatively high in comparison to physiological con-centrations found in the aqueous humor. AII concentrationsin the aqueous humor have been reported to range from 5 to16 fmol/mg protein in rabbit2 or 0.5pM in normal human sub-jects.22 One of the possible explanations for these differenceswas that AII was produced locally near the ciliary muscle(components of RAS have been intensively localized in the ciliary process, a tissue that connects to the ciliarymuscle).1,6–10 Therefore, the ciliary muscle might be exposedto a higher level of AII than that existing in the aqueoushumor.

Angiotensin receptors in the rabbit ciliary muscle werefurther analyzed by a receptor binding assay. The relevanceof our results were not limited exclusively to the ciliarymuscle because we used ciliary body membrane preparationswhich were composed of both ciliary process and ciliarymuscle. As a result, a Scatchard analysis showed that binding of AII to the ciliary body occurred at a single site. The receptor subtype was then determined to be pre-dominantly AT1. This was consistent with the results of Ramirez et al.4 Despite the limitations we believe that theseresults provide supportive evidence to our findings in theUSF study.

USF is known to be regulated by muscle tone and extra-cellular matrix change in the ciliary muscle. A possiblemechanism for the USF decrease by AII may be the musclecontraction. Lograno and Reibaldi23 studied the effect of AIIon isolated human ciliary muscle and found that the musclecontracts when AII was added at concentrations between 1and 20mM. The maximum response obtained by AII wasmore than 60% of the maximum muscle contraction inducedby high K+.

In summary, AII in the anterior chamber was indicated toattenuate USF; however this effect was considered too smallto influence IOP under physiological conditions.

Acknowledgements

The authors thank Makoto Mizuno* and Toshio Sada* fortheir helpful advice. The authors are also very grateful toYumiko Mori† for her excellent technical assistance in thestudy.

References

1. Bill A. The aqueous humour drainage mechanism in thecynomolgus monkey (Macaca irus) with evidence for un-conventional routes. Invest Ophthalmol Vis Sci. 1965;4:911–919.

Research Laboratories of *Pharmacology and Molecular Biology,and †Neuroscience and Immunology, Sankyo Co., Ltd.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

he U

nive

rsity

of

Man

ches

ter

on 1

0/26

/14

For

pers

onal

use

onl

y.

Uveoscleral outflow and angiotensin II 143

2. Bill A, Hellsing K. Production and drainage of aqueoushumour in the cynomolgus monkey (Macaca irus). InvestOphthalmol Vis Sci. 1965;4:920–926.

3. Sramek SJ, Wallow IH, Tewksbury DA, Brandt CR, PoulsenGL. An ocular renin-angiotensin system. Immunohisto-chemistry of angiotensinogen. Invest Ophthalmol Vis Sci.1992;33:1627–1632.

4. Ramirez M, Davidson EA, Luttenauer L, Elena PP, CuminF, Mathis GA, De Gasparo M. The renin-angiotensinsystem in the rabbit eye. J Ocul Pharmacol Ther. 1996;12:299–312.

5. Ikemoto F, Yamamoto K. Renin-angiotensin system in theaqueous humor of rabbits, dogs and monkeys. Exp Eye Res.1978;27:723–725.

6. Igic R, Kojovic V. Angiotensin I converting enzyme (kini-nase II) in ocular tissues. Exp Eye Res. 1980;30:299–303.

7. Vita JB, Anderson JA, Hulem CD, Leopold IH.Angiotensin-converting enzyme activity in ocular fluids.Invest Ophthalmol Vis Sci. 1981;20:255–257.

8. Laliberte MF, Laliberte F, Alhenc Gelas F, Chevillard C.Immunohistochemistry of angiotensin I-converting enzymein rat eye structures involved in aqueous humor regulation.Lab Invest. 1988;59:263–270.

9. Sramek SJ, Wallow IH, Day RP, Ehrlich EN. Ocular renin-angiotensin: Immunohistochemical evidence for the pres-ence of prorenin in eye tissue. Invest Ophthalmol Vis Sci.1988;29:1749–1752.

10. Strittmatter SM, Braas KM, Snyder SH. Localization ofangiotensin converting enzyme in the ciliary epithelium ofthe rat eye. Invest Ophthalmol Vis Sci. 1989;30:2209–2214.

11. Wallow IH, Sramek SJ, Bindley CD, Darjatmoko SR,Gange SJ. Ocular renin angiotensin: EM immunocyto-chemical localization of prorenin. Curr Eye Res. 1993;12:945–950.

12. Brandt CR, Pumfery AM, Micales B, Bindley CD, LyonsGE, Sramek SJ, Wallow IH. Renin mRNA is synthesizedlocally in rat ocular tissues. Curr Eye Res. 1994;13:755–763.

13. Mallorga P, Babilon RW, Sugrue MF. Angiotensin II recep-tors labelled with 125I-[Sar1, Ile8]-AII in albino rabbit oculartissues. Curr Eye Res. 1989;8:841–849.

14. Vogh BP, Godman DR. Effects of inhibition of angiotensinconverting enzyme and carbonic anhydrase on fluid pro-duction by ciliary process, choroid plexus, and pancreas. J Ocul Pharmacol. 1989;5:303–311.

15. Giardina WJ, Kleinert HD, Ebert DM, Wismer CT, ChekalMA, Stein HH. Intraocular pressure lowering effects of therenin inhibitor ABBOTT-64662 diacetate in animals. J OculPharmacol. 1990;6:75–83.

16. Costagliola C, Di Benedetto R, De Catoprio L, Verde R,Mastropasqua L. Effect of oral captopril (SQ 14225) onintraocular pressure in man. Eur J Ophthalmol. 1995;5:19–25.

17. Constad WH, Fiore P, Samson C, Cinotti AA. Use of anangiotensin converting enzyme inhibitor in ocular hyper-tension and primary open-angle glaucoma. Am J Ophthal-mol. 1988;105:674–677.

18. Costagliola C, Verolino M, Iaccarino G, Leonarda De RosaM, Ciancaglini M, Mastropasqua L. Losartan potassiumoral administration decreased intraocular pressure inhumans. A pilot study. Clin Drug Invest. 1999;17:329–332.

19. Goh Y, Araie M, Nakajima M, Azuma I, Hayaishi O. Effectof topical prostaglandin D2 on the aqueous humor dynam-ics in rabbits. Graefes Arch Clin Exp Ophthalmol.1989;227:476–481.

20. Bill A. The routes for bulk drainage of aqueous humour inrabbits with and without cyclodialysis. Doc Ophthalmol.1966;20:157–169.

21. Poyer JF, Gabelt B, Kaufman PL. The effect of topicalPGF2a on uveoscleral outflow and outflow facility in therabbit eye. Exp Eye Res. 1992;54:277–283.

22. Danser AH, Derkx FH, Admiraal PJ, Deinum J, de Jong PT,Schalekamp MA. Angiotensin levels in the eye. Invest Oph-thalmol Vis Sci. 1994;35:1008–1018.

23. Lograno MD, Reibaldi A. Receptor-responses in freshhuman ciliary muscle. Br J Pharmacol. 1986;87:379–385.

Cur

r E

ye R

es D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

he U

nive

rsity

of

Man

ches

ter

on 1

0/26

/14

For

pers

onal

use

onl

y.