Forskolin lowers intraocular pressure by reducing aqueous ... · Forskolin Lowers Intraocular...

Articles

Forskolin Lowers Intraocular Pressure byReducing Aqueous Inflow

Joseph Caprioli, Marvin Sears, Larry Bausher, Douglas Gregory, and Alden Mead

Forskolin is a diterpene derivative of the plant Coleus forskohlii that stimulates adenylate cyclaseactivity without interacting with cell surface receptors. Forskolin lowers the intraocular pressure ofrabbits, monkeys, and humans. In rabbits, net aqueous humor inflow decreases, outflow facility remainsunchanged, and ciliary blood flow increases. Tolerance to the intraocular pressure lowering effect didnot occur in rabbits after topical doses given every 6 hr for 15 days. In vitro forskolin activates adenylatecyclase of crude particulate homogenates prepared from cultured human ciliary epithelia or from dissectedciliary epithelial processes of rabbit or human eyes. This activation is not blocked by timolol. Thestimulation of adenylate cyclase by isoproterenol in vitro is potentiated in the presence of forskolin.Forskolin represents a potentially useful class of antiglaucoma agents differing in molecular mechanismof action from previously used drugs. Invest Ophthalmol Vis Sci 25:268-277, 1984

The stimulation of adenylate cyclase in the ciliaryepithelium by cholera toxin,1"3 by commercial prep-arations of gonadotropins,4 or by activation of betareceptors5 can lower intraocular pressure (IOP) by re-ducing net aqueous inflow.6 Thus, receptors that arecoupled functionally to ciliary epithelial adenylate cy-clase can be activated by beta adrenergic agonists, suchas isoproterenol, and certain gonadotropic hormones(especially FSH and hCG).4 Forskolin is a unique di-terpene derivative of the plant Coleus forskohlii thatacts independently of cell surface receptors to increaseintracellular levels of cyclic AMP (cAMP).10"15 It hasbeen used experimentally for its positive inotropic andblood pressure lowering effect. We have demonstratedrecently that topical forskolin significantly reduces IOPin rabbits, monkeys, and humans.16

The purpose of this follow-up work was to examinein further detail the effect of forskolin on the eye bystudying (1) the effect of intravitreal forskolin on rabbitIOP; (2) the detailed dose-response relationship oftopical forskolin in rabbits; (3) the potential cardio-vascular side-effects of topical forskolin; (4) the systemicand ocular effects of forskolin after intravenous ad-ministration; (5) the effects of topical forskolin on re-gional ocular blood flow; (6) the effect on aqueous flow

From the Department of Ophthalmology and Visual Science, YaleUniversity School of Medicine, New Haven, Connecticut.

Supported in part by the Connecticut Lions Eye Research Foun-dation, Inc, the New Haven Foundation, and Foresight, Inc.

Submitted for publication: July 6, 1983.Reprint requests: Marvin L. Sears, MD, Professor and Chairman,

Department of Ophthalmology and Visual Science, Yale UniversitySchool of Medicine, 310 Cedar Street, Box 3333, New Haven, CT06510.

by topical forskolin using a new method of intravitrealinjection of fluorescein-dextran17; and (7) the in vitroeffects of forskolin on human and rabbit ciliary ade-nylate cyclase activity.

Materials and Methods

Forskolin was obtained from Calbiochem-BehringCorporation (La Jolla, CA), 1-isoproterenol and flu-orescein-dextran from Sigma Chemical Company (St.Louis, MO), and timolol was donated by Merck, Sharp& Dohme (West Point, PA). Radioactive microsphereswere obtained from New England Nuclear Corporation(Boston, MA).

Animals

Male New Zealand white rabbits weighing 2.0-2.5kg were used. Monkeys were Macaca arctoides weigh-ing 10-15 kg.

Subjects

Human subjects were normal male and female vol-unteers with no history of eye disease and normal eyeexaminations. Informed consent was obtained fromeach subject.

IOP Measurements

Rabbit IOP was measured after a drop of topicalproparacaine using an applanation pneumotonometercalibrated for rabbit eyes. Measurements were madeimmediately before treatment and at 1,2, 3, 4, 6, 10,12, and 24 hr after treatment with forskolin.

In monkeys, anesthesia was obtained using ketamine

268

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933112/ on 04/06/2017

No. 3 FORSKOLIN LOWERS INTRAOCULAR PRESSURE / Coprioli er ol. 269

10 mg per kg intramuscularly in addition to a dropof proparacaine topically just before IOP measurement.Measurements were made using an applanation pneu-motonometer before drug application and at 1,2, 4,and 8 hr after treatment.

Measurements of IOP in humans were made aftertopical proparacaine with a Goldmann applanationtonometer immediately before instillation of forskolinand at 1,2, 3, and 5 hr after treatment.

Topical Forskolin

Forskolin is poorly soluble in water but is readilysoluble in ethanol. An initial control study was con-ducted to determine the effect of topical ethanol aloneon rabbit IOP. A 50 t̂l topical dose in concentrationseven as low as 10% ethanol irritated the eyes and causedsignificant reductions in IOP. To avoid the problemsof topical ethanol, forskolin suspensions were madeusing a balanced salt solution containing 0.5% hy-droxypropyl methyl cellulose as a suspending agent.Suspensions were prepared immediately before use bythoroughly mixing finely powdered forskolin with thevehicle in a mortar and pestle. Topical applications of50 jil were given onto the cornea of the right eye ofall animals or subjects, while the left eye received atopical application of a 1% talc suspension in the samevehicle. Talc was used in the control drop in order toapproximate the fine particulate nature of the forskolinsuspension. The drug was applied well after the effectsof topical proparacaine used for baseline IOP mea-surements had worn off.

Topical suspension concentrations of 0.1%, 1.0%,and 4.0% were each tested in 20 rabbits, and the 1.0%topical suspension was used in 10 monkeys and 10human subjects; those data have been previously re-ported.16 Results of testing with a 0.5% topical sus-pension in rabbits are reported here.

Tolerance

The potential development of tolerance to the IOPlowering effect of forskolin after short term use wastested for by topical administration of 50 n\ of a 1.0%forskolin suspension every 6 hr to the right eye of 10rabbits, starting at 9 AM daily. The IOP of each rabbitwas measured at noon for 15 consecutive days.

Intravitreal Forskolin

A 10 n\ volume of forskolin in a 25% ethanol solutionwas injected into the midvitreous cavity of rabbit eyesusing a 25 n\ Hamilton syringe and a 30-gauge hublessneedle after topical proparacaine. Injections of 0.016Hg, 0.16 /ig and 1.6 ng of forskolin were given so thatfinal intraocular concentrations were approximately

10 8, 10 7, and 10 6 M, respectively, assuming a dis-persion volume of 4.0 ml. There was no noticeablereflux from the injection site and no conjunctival re-action was noted using this technique. All intravitrealinjections containing forskolin were given in the righteye while control injections of 10 ix\ of 25% ethanolwere given in the left eye.

Systemic Forskolin

Forskolin was administered intravenously to fiverabbits. Two of these rabbits were anesthetized withketamine (10 mg/kg) intramuscularly, and three wereanesthetized with urethane (5 ml/kg of a 25% solution)intravenously. All rabbits were heparinized with 500units of heparin sulfate per kg. In the two animalsanesthetized with ketamine, blood pressure was mon-itored with a 22-gauge silastic catheter placed in theleft auricular artery. The effect of ketamine was allowedto wear off before forskolin was administered throughthe left marginal ear vein. Urethanized rabbits had aleft femoral artery catheter (22 gauge) placed for bloodpressure monitoring and a left femoral vein catheter(22 gauge) placed for drug administration. These an-imals had the left anterior chamber cannulated witha 23-gauge needle fired by a needle gun.16 Simultaneoustracings of IOP and blood pressure were made usingcalibrated transducers and a Sanborn recorder.

Several dilutions of forskolin in absolute alcoholwere made; a volume for intravenous injection wasprepared by diluting a 10 n\ aliquot from an appropriatesolution with 0.5 ml of normal saline. This was de-livered as an intravenous bolus followed immediatelyby a 0.5-ml saline flush. Control injections of 10 /A ofethanol in 0.5 ml saline were given randomly. Forskolindoses ranging from 0.005 mg/kg to 0.50 mg/kg wereadministered.

Tonography

Tonography was performed on six eyes of three rab-bits using a 5.5-gm weight and a Mueller electronictonographic recorder. Animals were anesthetized withketamine 10 mg/kg intramuscularly 10 min beforetonography was performed. Tonographic tracings weremade immediately before, and 3 hr after, topical ap-plication of a 1.0% forskolin suspension bilaterally. Po

was verified using an applanation pneumotonometer.Standard tonographic tables were used to calculateoutflow facility.

Aqueous Flow Determination

Measurements of aqueous flow in rabbit eyes treatedwith a topical 1.0% forskolin were compared to controleyes. The method used was a modification of a fluo-


270 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / March 1984 Vol. 25

18 TI

16

14

IEEa.2 18

16

14

i0.16/iq

i i

• 6/ig

J L20 240 4 8 12 16

TIME (hours)Fig. 1. Intraocular pressure (mean ± SEM) versus time after in-

travitreal injection of 0.16 ng (n = 20) and 1.6 n% (n = 20) in rabbits.Open circles represent significant departures from baseline {P < 0.01).

rescein technique described by Maurice17 and involvesmeasurements of anterior chamber fluorescein con-centration after intravitreal injection of fluorescein-dextran. It is assumed that this large molecule exitsthe eye only via conventional outflow pathways andthat the anterior chamber fluorescein concentration isinversely proportional to the aqueous flow rate. In-travitreal injections of 10 /ul of a 10% solution of flu-orescein dextran (MW 40,000) were given bilaterallyin six rabbits. Two weeks later, four rabbits were treatedunilaterally with 50 n\ of a 1 % topical forskolin sus-pension, and two animals were treated with acetazol-amide (12.5 mg/kg) intravenously. Anterior chamberparacenteses, 100 (A, only, were done 3 hr later, andfluorescein concentrations were measured spectropho-tofluorimetrically. Relative aqueous flow rates weredetermined in the acetazolamide and forskolin-treatedeyes, relative to the untreated eyes, by calculating theratio of anterior chamber fluorescein concentration oftreated versus control eyes.

Blood Pressure after Topical Forskolin

Two rabbits were anesthetized with ketamine andfive with urethane as described above. A 22-gauge si-lastic catheter was introduced into the left auricularartery of the ketaminized animals and the left femoralartery of the urethanized animals. Blood pressure wasmonitored continuously via a transducer and a San-born recorder. After stabilization of blood pressure,50 n\ of 1.0% forskolin suspension was placed in theright eye of each animal. In the ketaminized rabbits,

IOP was measured every 30 min by applanation pneu-motonometry while the blood pressure was monitoredcontinuously for 6 hr. In the urethanized animals, theIOP in the treated eye was monitored continuously bycannulation as described above.

Human brachial artery systolic and diastolic pressurein the right arm were determined in the sitting positionusing a standard cuff. Pulse rate and blood pressureswere recorded at 1, 2, 3, and 5 hr after the topicalinstillation of 50 n\ of a 1.0% forskolin suspension.

Blood Flow Measurements

Regional ocular blood flow experiments were per-formed in 16 rabbits treated unilaterally with 50 /xl ofa 1.0% topical forskolin suspension. This was accom-plished using a modification of the radioactively la-belled microsphere technique described by Aim andBill19 as follows.

Animals were anesthetized with urethane (5 ml/kgof a 25% solution) given intravenously. With the animalin supine position, head straight, and eyes level, 22-gauge silastic catheters were placed in both femoralarteries and one femoral vein. Arterial blood pressurewas monitored via a calibrated transducer and a San-born recorder. A tracheostomy was performed and theanimal ventilated with a Harvard respirator (volume= 22-25 ml, rate = 30-50 per min). The chest wasopened by median sternotomy, and the left ventriclewas exposed by pericardotomy. Blood pO2, pCO2, andpH were measured using a blood gas analyzer and theventilation was regulated so that pH was always main-tained between 7.40-7.48. Left intraventricular injec-tions of 0.5 ml of a 1.0% microsphere suspension (spe-cific activity 10 mCi/g) were then performed.

Fifteen-micrometer microspheres of 141Ce, 57Cr, and103Ru were used. Four animals received sequential leftintraventricular injections of each of the above isotopesjust before drug treatment and at various times (0.5-3 hr) after drug treatment. The remaining animalsreceived injections of a single isotope only, from 0.5-8 hr after drug delivery.

Simultaneous with the injection of microspheres,collection of a reference blood sample from the femoralartery was begun at the rate of 2-3 ml per minute andcontinued for 1 min. Animals then were sacrificed byintracardiac injection of a saturated potassium chloridesolution. The eyes were enucleated and dissected im-mediately. The iris-ciliary body, medial rectus muscle,conjunctiva, choroid, and the anterior 2 mm of theoptic nerve were removed and placed in tared vialsfor weighing and counting. Samples were counted ona Beckman multichannel gamma spectrometer. Threeenergy windows were used to correspond to the energypeaks of each of the isotopes. Blood flows of all sampleswere calculated using the reference sample technique,19


No. 3 FORSKOUN LOWERS INTRAOCULAR PRESSURE / Coprioli er ol. 271

Fig. 2. Intraocular pressure(mean ± SEM) versus timeafter topical delivery of for-skolin suspension 0.1% (n= 20), 0.5% (n = 20), 1.0%(n = 20), and 4.0% (n = 20)in rabbits. Open circles rep-resent significant departuresfrom baseline (P < 0.01).

16

14

I

I 1 2

Q_H 18

16

14

0.1%

0.5%

I I I I I I I I I I

8 12 16

TIME (hours)

20 24

I 8 r

16

14

12

18

16

14

I i

1.0%

4.0%

i i i i i i i i i i i i i

8 12 16 20 24

TIME (hours)

and ratios of blood flow between treated and contra-lateral untreated eyes were determined.

Adenylate Cyclase Activity In Vitro

All procedures were conducted on ice except whereindicated otherwise. The dissection of ciliary processesfrom eyes has been described,20 except that the tissuewas dissected and the ciliary tips were collected underHank's balanced salt solution (BSS). Rabbit ciliary tis-sue from 20 eyes was pooled and homogenized in 2.0ml homogenizing medium (50 mM TRIS HC1, 1 raMDTT, pH 7.5) in a glass-teflon homogenizer using sevenstrokes of the pestle. The homogenate was centrifugedat 3000 X g at 0-5 °C for 5 min and the supernatantwas removed. The pellet was resuspended in 2 ml ho-mogenizing medium with seven strokes of the pestleas above. The resulting particulate suspension wasmixed immediately with 3.025 ml homogenizing me-dium, 25 /xl were withdrawn for protein determination,and the remaining 5.0 ml divided into 500-/*l aliquotsthat were frozen and stored under liquid nitrogen untiluse. Final preparation of an aliquot for assay involvedthawing, transferring the sample to a 1-ml glass-teflonhomogenizer, centrifuging as above, removing the su-pernatant, and resusperiding the pellet in assay mediumwith seven strokes as above.

Adenylate cyclase was assayed in duplicate at 30°Cas described3 in test tubes containing 100 mM TRISHC1, pH 7.5, 5 mM MgCl2, 50 mU/^1 creatine kinase,10 mM creatine phosphate, 2 mMDTT, 0.5 mg/mlbovine serum albumin, 1 mM cyclic AMP, 0.5 mMATP, 1 fid a-32P-ATP, 25 ng of particulate protein,and the presence or absence of sodium fluoride, for-skolin, or 1-isoproterenol. Recovery of product duringthe purification procedure was monitored with 3H-

cAMP. Human processes were dissected and frozenand stored under BSS in liquid nitrogen until theywere used. A crude particulate suspension was preparedas described above, except that 0.5 ml homogenizingmedium was used for homogenization and subsequentresuspension, and the resuspension was not dilutedfurther before final resuspension in assay medium.Cultured human ciliary epithelial cells were preparedand assayed as previously described.24

Statistics

Values are expressed as the mean ± S.E.M. (n).Statistical analyses were conducted using the Studentt test adjusted for paired data, and the Student's t testfor slopes of linear regression. P values of <0.05 wereconsidered significant.

Results

Intravitreal Forskolin

Intravitreal doses of 0.16 fig and 1.6 ng in rabbitsproduced significant decreases in IOP compared withbaseline (Fig. 1). Intravitreal injections of 0.016 ^gcaused no significant change in IOP, and no significantchange in IOP occurred in any of the control eyes.

Topical Forskolin

These data have been reported previously in part.16

Significant decreases in IOP occurred compared withbaseline after topical applications of 0.1%, 0.5%, 1.0%and 4.0% forskolin suspension in rabbits (Fig. 2).Twenty animals were used in each group. The largesteffect was seen with the 1.0% dose, with a fall in IOPfrom 17.4 ± 0.4 mmHg (20) to 13.1 ± 0.4 mmHg


272 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / March 1984 Vol. 25

-06 ' -0.4 -0.2 0

LOG FORSKOLIN DOSE

Fig. 3. Linear regression of the duration of statistically significantlowering of intraocular pressure as a function of the logarithm oftopical forskolin concentration (% suspension) in rabbits (r = 0.97,P = 0.002). X intercept indicates least effective dose.37

(20) at 3 hr, with a statistically significant decrease inIOP lasting from 2 to 16 hr after drug delivery.16 Theduration of significant effect as a function of dose isdisplayed in Figure 3. No significant IOP change wasdetected in the contralateral control eyes except in the1.0% group, which showed a small decrease at 3 hr.A transient superficial hyperemia developed in alltreated eyes, lasting for 30-45 min after instillation.No change in pupillary size occurred, and neither flare

18 T-

IEEa.O

16

14

16

14 1-

12

10

MONKEY

HUMAN

TIME (hours)

Fig. 4. Intraocular pressure (mean ± SEM) versus time after ad-ministration of 1.0% forskolin topically in monkeys (n = 10) andhumans (n = 10). Open circles represent significant departures frombaseline (P < 0.01).

18 r

17 o-

16enE

rf 15O

g|4

13 -

12 -

'''' I I

C 2J I

6 8 10

DAY OF TREATMENT

12 14

Fig. 5. Daily noon intraocular pressure (mean ± SEM) in rabbits(n = 10) versus day of treatment during a 15-day period of topicaladministration of a 1.0% forskolin suspension every 6 hr. Doses weregiven at 9 AM, 3 PM, 9 PM, and 3 AM daily. Tolerance to the drugdid not develop within this period. This slope of the linear regressionis not significantly different from 0 {P = 0.23).

nor cell was detected by biomicroscopy throughoutthe experiments.

In monkeys, IOP decreased in the treated eye from17.7 ± 0.6 mmHg (10) to 14.3 ± 0.8 mmHg (20) at2 hr (Fig. 4).16 No significant change in the contralateralcontrol eyes were found. Neither ocular irritation noraqueous flare nor cell was detected and no significantchange in pupillary size occurred during the experi-ment.

The human data show significant decreases in IOPat 1, 2, 3, and 5 hr after topical administration of 1.0%forskolin suspension (Fig. 4). Mean IOP fell from 14.7±0.8 mmHg (10) to 10.7 ± 1.1 mmHg (10) 2 hr afterdrug delivery.16 At the 2-hr interval, and only at thistime, a significant decrease was also found in the con-tralateral control eye. Mean IOP fell in the con-trol eye from 14.5 ±0 .8 mmHg (10) to 12.4 ±0 .8mmHg (10).

Tolerance

Topical administration of a 1.0% forskolin concen-tration to rabbits every 6 hr on 15 consecutive daysrevealed no significant decrease in the drug's ocularhypotensive effect (Fig. 5). The slope of the linearregression line through the daily mean noon IOP valuesin the treated eyes was not significantly different from0 (P = 0.23).

Tonography

Outflow facility in rabbits measured 3 hr after topicaladministration of a 1.0% forskolin suspension in six


No. 3 FORSKOUN LOWERS INTRAOCULAR PRESSURE / Coprioli er ol. 273

Table 1. Relative aqueous flow determinationsin rabbits

60 r

ControlAcetazolamideForskolin

Eyes(n)

444

AqueousJluorescein

concentration(ng/til)

3.16 ±0.425.69 ± 0.395.88 ± 0.44

Significance(P)*

0.0120.011

Relativedecrease

in aqueousflowf

44%46%

LJ

a.COCOLJ

cra.o

)LO

O

LJ

SO

40

30

20

* Results of t tests comparing aqueous fluorescein in the treated versuscontrol groups.

t Compared with control group.

eyes showed neither suggestion nor statistically sig-nificant change in outflow facility compared to themean baseline value of 0.18 ± 0.04 mmHg (P = 0.41).

Aqueous Flow Determination in Rabbits

The anterior chamber fluorescein-dextran concen-trations in the control, acetazolamide-treated, and for-skolin-treated eyes are referred to in Table 1. Comparedwith the nontreated eyes, there was a significant de-crease of mean aqueous flow of approximately 46%in the forskolin-treated eyes and approximately 44%in the acetazolamide-treated eyes. The decrease inaqueous flow of 46% in the forskolin treated animalsis similar to the decrease in outflow pressure in a grouptreated similarly and reported earlier using anotherfluorometric method for flow.16 These data supportthe conclusion that the decreased IOP occurring aftertopical forskolin is caused by a decrease in aqueousinflow without a significant change in outflow facility.*

Effect of Systemic Forskolin on Systolic BloodPressure and IOP in Rabbits

Intravenous doses as small as 0.0005 mg/kg causeda detectable decrease in systolic blood pressure in rab-bits. A 0.5 mg/kg dose given to three rabbits causeda profound and prolonged drop in mean blood pres-sure, with recovery to baseline after approximately 45min. A dose-response curve relating drop in meanblood pressure to intravenous forskolin dose was con-structed (Fig. 6). With intravenous doses less than 0.5mg/kg, only small transient volumetric changes in IOPwere recorded in the cannulated eyes. Doses of 0.5mg/kg in three rabbits caused a steady state IOP de-crease of 3 to 5 mmHg with the onset at approximately45 min and recovery to baseline occurring at approx-imately 200 min.

* A recent study done in the eyes of cynomologous monkeys aftersingle dose topical 1% suspension of forskolin indicated that aqueousflow measured by noninvasive fluorometry decreased by 35% whiletoriographic outflow did not change (personal communication:P. Lee, S. Podos, C. Severin, T. Mittag).

io -»

-3.0 -2.0 -1.0

LOG FORSKOLIN DOSE

Fig. 6. Linear regression of the maximum decrease in the meanblood pressure as a function of the logarithm of the intravenousforskolin dose (mg/kg) in five rabbits (r = 0.93, P < 0.001).

Effects of Topical Forskolin on Blood Pressurein Rabbits and Humans

Applications of topical 1.0% forskolin suspensioncaused no change in systolic or mean blood pressurein seven rabbits. Mean baseline blood pressure was 75± 4 mmHg (7), and measurements at 0.5, 1.0, 2.0 and3.0 hr after topical application of drug yielded meansystolic pressures of 77 ± 4 mmHg (7), 78 ± 3 mmHg(7), 8 ± 4 mmHg (7), 76 ± 4 mmHg (7), and 78 ± 3mmHg (3), respectively. Furthermore, continuouslyrecorded tracings showed no change in blood pressureor pulse rate that could be related to the administrationof the drug.

BLOOD FLOW

'•°H\-

IOP

0.7 T

0 1 2 3 4 5 6 7 8

TIME (hours)

Fig. 7. Mean, ratio of iris-ciliary body blood flow in the treatedeye versus the untreated contralateral control eye along with themean ratio of IOP in the treated versus untreated eye in the samerabbits with time. All animals received 50 /xl of a 1.0% topical forskolinsuspension in the treated eye at time 0, while the other eye received50 n\ of suspension vehicle only.


274 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / Morch 1984 Vol. 25

IOOO

800

600

400

200

7 6 5 4

-LOG (Forskolin), M

Fig. 8. Dose response curve for stimulation of rabbit ciliary ad-enylate cyclase by forskolin. The presence of 1 nM timolol had noeffect on cyclase activation by forskolin and yielded a dose responsecurve identical to the one shown above.

In humans, mean systolic and diastolic blood pres-sures before treatment with 1.0% topical forskolin were116 ± 5 and 63 ± 3 mmHg (10), respectively; nosignificant changes occurred at 1, 2, 3, or 5 hr. Pulserate was 66 ± 4 min"1 and remained constant duringthe experiment.

Blood Flow Measurements

The ratio of iris-ciliary body blood flow in the treatedversus untreated contralateral eyes is plotted as a func-tion of time after forskolin administration in Figure7, along with IOP of the same eyes. The mean bloodflow in the iris-ciliary body of all untreated eyes was170 ± 21 (16) gm/min/100 gm fresh tissue. Blood flowin the conjunctiva, rectus muscle, choroid, and optic

Table 2. Stimulation of human ciliary adenylatecyclase in vitro

Cultured cells*Ciliary

processes*

100 nM 1-isoproterenol4 mM NaF61 nM forskolin

2.3 ± 0.3 (2)22.9 ± 2.2 (2)57.5 ± 7.0 (2)f

5.3 ± 0.7 (2)9.2 ± 2.2 (2)

29.7 ± 2.8 (2)f

* Numbers represent ratios of activity in the presence of agent to basalactivity.

t Activation by forskolin was significantly greater than isoproterenol orNaF.

nerve in the treated eyes was not found to be signif-icantly different from the untreated eyes.

The peak iris-ciliary blood flow occurred at 1 hr, ata time when IOP was not significantly different frombaseline, while the maximum IOP decrease occurredat 4 hr, at a time when iris-ciliary blood flow hadreturned to a value not significantly different from themean baseline value.

Adenylate Cyclase Activity In Vitro

The dose-response curve for activation of adenylatecyclase from rabbit ciliary processes is shown in Figure8. The highest forskolin concentrations stimulated ad-enylate cyclase significantly more than maximal stim-ulation with either 4 mM NaF or 100 /tM 1-isopro-terenol. The dose response curve of forskolin activationof cyclase activity in the presence of 10 ^M timololwas identical with the curve in the absence of timolol(data not shown). A 25.1-fold activation of rabbit ciliaryprocess adenylate cyclase by forskolin (6.1 nM) oc-curred in the presence of 4 mM NaF, which was greaterthan the sum of the responses produced by either ac-tivator alone (forskolin, 9.3-fold; NaF, 6.3-fold). Ac-tivation of human ciliary process adenylate cyclase isshown in Table 2. The cyclase activation in eitherfreshly dissected human processes or human culturedcells was significantly greater than activation by either1-isoproterenol or NaF.

Stimulation of rabbit ciliary process adenylate cy-clase by isoproterenol in combination with forskolinis synergistic. Activation by 100 /xM isoproterenol inthe presence of 0.2, 2.0, or 20 /xM forskolin was sig-nificantly greater than the sum of the responses pro-duced by either activator alone.

Discussion

Forskolin is a diterpene derivative of the plant Coleusforskohlii that acts on the adenylate cyclase catalyticsubunit to increase intracellular cAMP, may not re-quire the guanine nucleotide regulatory subunit, anddoes not require a cell membrane bound protein re-ceptor. Forskolin produces cellular responses depen-dent on cAMP as a second messenger, and has theability to potentiate greatly the hormonal activationof adenylate cyclase.10"15 We have reported recentlypreliminary findings demonstrating dramatic IOP re-duction after topical administration in rabbits, mon-keys, and humans.16

Intravitreal injection or topical administration offorskolin in suspension significantly reduces IOP inrabbits by reducing aqueous inflow without any in-crease in outflow facility. Topical administration offorskolin in monkeys and in normd human volunteersalso caused significant decreases in IOP.16 During and



after topical ocular delivery of forskolin in rabbits andhumans, blood pressure and pulse were measured. Nochanges occurred. Intravenous administration of for-skolin in rabbits produced a steady state decrease inintraocular pressure only after doses large enough toapproximate the intraocular concentrations probablyachieved after topical administration.

Iris-ciliary body blood flow increased by approxi-mately 2.5-fold 1 hr after topical forskolin, while cho-roidal flow remained unchanged. This increase prob-ably was caused by a direct vasodilatory effect in theciliary body. The increase in blood flow and decreasein net aqueous flow mimic the cholera toxin effect.2'3

Forskolin activated adenylate cyclase in rabbit andfreshly dissected human ciliary processes, as well as inhuman cultured ciliary epithelial cells. Timolol didnot inhibit this activation, substantiating the findingthat forskolin does not work by interaction with thebeta-adrenergic receptor. The adenylate cyclase recep-tor complex consists of a membrane-bound receptorprotein, a guanine nucleotide regulatory subunit, anda catalytic subunit. These components of the "secondmessenger" system are present in the ciliary epitheliumin rabbits20"22 and humans.5'23 Increased ciliary ade-nylate cyclase activity after beta receptor stimulationin both species has been demonstrated,5'21"23 in accordwith beta2 reception.23 Furthermore, activation of ad-enylate cyclase induces protein phosphorylation incultured epithelium, indicating the presence of acAMP-dependent protein kinase system in this tissue.24

These findings suggest that a complete, functionallycoupled, "second messenger" system is present in the

Table 3. Effect of forskolin on isoproterenolstimulation of rabbit adenylate cyclase

Forskolinconcentration

Cyclase activityforskolin alone(pmol/min/mg

protein)

Isoproterenolstimulation}

(pmol/min/mgprotein)

00.22.0

20.0

49.7124292616

± 2.± 12± 18± 51

3(3)*(3)(3)(3)

13.744.793.7102

± 1.5(3)± 6.7 (3)t± 27.4 (3)t±23 O)%

* Values represent mean ± SEM (n).t Isoproterenol stimulation is denned as adenylate cyclase activity in the

presence of 100 /iM isoproterenol minus the activity in the absence of iso-proterenol for each forskolin concentration. This represents the componentof only isoproterenol stimulation in the presence of forskolin.

t Isoproterenol stimulation in the presence of forskolin is greater than thestimulation in the absence of forskolin (P < 0.05).

ciliary epithelium. In the present work we found thatisoproterenol, in combination with forskolin, showeda synergistic effect with cyclase activation in the pres-ence of both drugs being significantly greater thanthe sum of the activation using either agent alone(Table 3).

A reduction in net aqueous humor inflow mediatedby increased cAMP levels in the ciliary epitheliumrepresents a unifying hypothesis concerning the reg-ulation of IOP4 (Fig. 9). Stimulation of the adenylatecyclase complex via beta receptor agonism in the ciliaryepithelium can lower IOP.6"9 Topical application ofvarious beta-adrenergic agonists produce changes inaqueous flow (though relatively small), the directionof which seems to vary with the method of measure-

Fig. 9. The proposed re-lationship between exoge-nous and endogenous fac-tors on the regulation of in-traocular pressure via mod-ulation of aqueous inflow bycyclic AMP. CLINICAL SYNDROME:

PREGNANCYMYOTONIC DYSTROPHYCHORIOCARCINOMAEXOGENOUS GONADOTROPINS

| GONADOTROPINS )}

— JAQUEOUSFLOW

IOP


276 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / Morch 1984 Vol. 25

ment used.25 Fluorometric determinations of aqueousflow have been done after topical fluorescein,26 ion-tophoresis,27 oral or intravenous fluorescein,28 posteriorchamber fluorescein,29 or after intravitreal injection offluorescein labelled dextran.17 The latter technique al-lows determinations over a longer period of time, candisentangle multiple phases (eg, an initial increase fol-lowed by a decrease), and can demonstrate relativelysmall changes in flow induced by catecholamines thatotherwise may go undetected. Maurice has demon-strated, for example, that topical epinephrine causesfirst a slight transient increase in flow, followed by aprolonged decrease.17

Work in our laboratory demonstrated that choleratoxin, a potent irreversible stimulator of adenylate cy-clase, stimulates adenylate cyclase in rabbit and humanciliary epithelium and lowers IOP by reducing aqueousflow in rabbits when administered in minute (10~10

M) doses by either intravitreal injection or close arterialinfusion.1"3 The gonadotropins—hCG and FSH—which are glycoprotein stimulators of cAMP produc-tion also decrease IOP and aqueous flow in either nor-mal or oophorectomized rabbits by either intravitrealor by systemic intramuscular injection.4 Thus, beta-adrenergic agonists, certain glycoprotein hormones,cholera toxin, fluorides, and forskolin all stimulate ad-enylate cyclase, albeit by different molecular pathways.Catecholamines29"31 and cholera toxin2'3 (D. Maurice,personal communication) have been shown previouslyto reduce IOP in steady state by lowering aqueousinflow. Certain clinical syndromes such as myotonicdystrophy, choriocarcinoma, and pregnancy are as-sociated with high serum gonadotropin levels and lowIOP.4 We now have confirmed a dramatic reductionin aqueous flow and IOP with forskolin.16 Exogenouslyapplied substances and endogenous circulating hor-mones may regulate aqueous inflow (and thus IOP)via a final common pathway, the adenylate cyclasereceptor complex in the ciliary epithelia.

The mechanism by which elevated intracellularcAMP levels lead to decreased aqueous production isnot known. One hypothesis is introduced here. Theeffect of cholera toxin in the gut,32 cochlea,33 and cho-roid plexus34 has been investigated. In each case, stim-ulation of adenylate cyclase in epithelial cells producesbulk flow of water and electrolytes from the basal sideof the cell out through the cellular apex in accord withthe polarity of the cells. In the eye, however, the apicesof the pigmented and nonpigmented ciliary epitheliaare pointed toward each other as a result of the in-vagi nation of the optic vesicle during embryologic de-velopment, as reviewed by Hogan et al.35 The non-pigmented epithelium thus has its base toward theaqueous compartment and its apex toward the blood.In addition, histochemical localization of adenylate

cyclase in the ciliary body apparently is restrictedlargely to the nonpigmented epithelium with little orno activity in the pigmented epithelium.36 Stimulationof the enzyme in the nonpigmented layer may promotereabsorption of fluid from the posterior chamber andsecretion into the ciliary stroma, thereby decreasing"net" aqueous inflow. Vascular effects on secretionsecondary to increased ciliary body blood flow cannotyet be excluded.

Forskolin and its analogues may represent a newclass of antiglaucoma drugs differing in its molecularmechanism from any previously used drug. Its effecton IOP should be additive with other glaucoma drugs,owing to its unique mode of action, and may evenpotentiate the effect of certain drugs, eg, epinephrine,which function (partially) through beta-receptor agon-ism. Tachyphylaxis or tolerance may not occur becauseforskolin's action does not involve the cell surface re-ceptor. Slight modification of the molecule may in-crease its water solubility and cornea! penetration, re-ducing the dose delivered to the surface of the eye andenhancing its effectivity. More work in this area iswarranted to determine forskolin's potential thera-peutic usefulness.

Key words: forskolin, adenylate cyclase, glaucoma, aqueoushumor formation, intraocular pressure, cyclic AMP, bloodflow, ciliary epithelium

References

1. Sears ML: The aqueous. In Adler's Physiology of the Eye, 7thedition, Moses RA, editor. St. Louis, CV Mosby, 1980, pp. 204-226.

2. Sears ML, Gregory D, Bausher L, Mishima H, and StjernschantzJ: Receptor for aqueous humor formation. In New Directionsin Ophthalmic Research, Sears ML, editor. New Haven, YaleUniversity Press, 1981, pp. 163-183.

3. Gregory DS, Sears ML, Bausher L, Mishima H, and Mead A:Intraocular pressure and aqueous flow are decreased by choleratoxin. Invest Ophthalmol Vis Sci 20:371, 1981.

4. Sears ML and Mead A: A major pathway for the regulation ofintraocular pressure. Intl Ophthalmol Clin 6:201, 1983.

5. Gregory DS, Bausher LP, Bromberg BB, and Sears ML: Thebeta-adrenergic receptor and adenyl cyclase of rabbit ciliary pro-cesses. In New Directions in Ophthalmic Research. Sears ML,editor. New Haven, Yale University Press, 1981, pp. 127-145.

6. Eakins K: The effect of intravitreous injections of norepinephrine,epinephrine, and isoproterenol on the intraocular pressure andaqueous humor dynamics of rabbit eyes. J Pharmacol Exp Therap140:79, 1963.

7. Ross EA and Drance SM: Effects of topically applied isoproterenolon aqueous dynamics in man. Arch Ophthalmol 83:39, 1970.

8. Gaasterland D, Kupfer C, Ross K, and Gabelnick HL: Studiesof aqueous humor dynamics in man. III. Measurements in youngnormal subjects using norepinephrine and isoproterenol. InvestOphthalmol 12:267, 1973.

9. Takase M: Effects of topical isoproterenol and propranolol onflow-rate of aqueous in rhesus monkey. Acta Soc OphthalmolJpn 80:379, 1976.



10. Seamon KB, Padgett W, and Daly JW: Forskolin: unique di-terpene activator of adenylate cyclase in membranes and inintact cells. Proc Natl Acad Sci USA 78:3363, 1981.

11. Metzger H and Lindner E: The positive inotropic acting forskolin,a potent adenylate cyclase activator. Arzneim-Forsch/Drug Re-search 31:1248, 1981.

12. Seamon KB and Daly JW: Forskolin: a unique diterpene activatorof cyclic AMP generating systems. J Cyclic Nucl Res 7:201,1981.

13. Seamon KB and Daly JW: Activation of adenylate cyclase bythe diterpene forskolin does not require the guanine nucleotideregulatory protein. J Biol Chem 256:9799, 1981.

14. Cuthbert AW and Spayne HA: Stimulation of sodium of chloridetransport in epithelia by forskolin. Br J Pharmacol 76:33, 1982.

15. Insel PA, Stengel D, Ferry N, and Hanoune J: Regulation ofadenylate cyclase of human platelet membranes by forskolin. JBiol Chem 257:7485, 1982.

16. Caprioli J and Sears M: Forskolin lowers intraocular pressurein rabbits, monkeys, and man. The Lancet 1:958, 1983.

17. Maurice D: A simple method for measuring aqueous flow inthe rabbit. ARVO Abstracts. Invest Ophthalmol Vis Sci24(Suppl):5, 1983.

18. Sears ML: Miosis and intraocular pressure changes during ma-nometry. Arch Ophthalmol 63:707, 1960.

19. Aim A and Bill A: The oxygen supply to the retina. II. Effectsof high intraocular pressure and of increased arterial carbondioxide tension on uveal and retinal blood flow in cats. ActaPhysiol Scand 84:306, 1972.

20. Bromberg BB, Gregory DS, and Sears ML: Beta adrenergic re-ceptors in ciliary processes of the rabbit. Invest Ophthalmol VisSci 19:203, 1980.

21. Neufeld AH and Sears ML: Cyclic AMP in ocular tissues of therabbit, monkey and human. Invest Ophthalmol 14:688, 1974.

22. Neufeld AH and Page ED: In vitro determination of the abilityof drugs to bind to adrenergic receptors. Invest Ophthalmol VisSci 16:1118, 1977.

23. Nathanson JA: Human ciliary process adrenergic receptor:pharmacologic characterization. Invest Ophthalmol Vis Sci21:798, 1981.

24. Coca-Prados M, Kondo K, and Sears ML: Protein phosphor-ylation in cultured human ciliary epithelia in response to ac-tivators of adenylate cyclase, cyclic AMP and analogs. In Glau-coma Update II, International Glaucoma Symposium, Carmel,California, September 22-27, 1982. Krieglstein GK and Ley-dhecker HW, editors. Heidelberg, Springer-Verlag, 1983, pp.1-6.

25. Sears ML: Adrenergic agonists. In Handbook of ExperimentalPharmacology, Vol. 69 of Handbuch der experimentellen Phar-makologie. Sears ML, editor. Heidelberg, Springer-Verlag, 1983,pp. 193-248.

26. Yablonski ME and Gray JR: Use of the fluorotron master tomeasure aqueous flow. ARVO Abstracts. Invest Ophthalmol VisSci 24(Suppl):88, 1983.

27. Jones RF and Maurice DM: New methods of measuring therate of aqueous flow in man with fluorescein. Exp Eye Res 5:208,1966.

28. Araie M and Takase M: Effects of various drugs on aqueoushumor dynamics in man. Jpn J Ophthalmol 25:91, 1981.

29. Miichi H and Nagataki S: Effects of cholinergic drugs and ad-renergic drugs on aqueous humor formation in the rabbit eye.Jpn J Ophthalmol 26:425, 1982.

30. Goldmann H: L'origine de l'hypertension oculaire dans le glau-come primitif. Ann Ocul (Paris) 184:1086, 1951.

31. Nagataki S: Effects of adrenergic drugs on aqueous humor dy-namics in man. Acta Soc Ophthalmol Jpn 81:1795, 1977.

32. Field M: Intestinal secretion: effect of cyclic AMP and its rolein cholera. N Engl J Med 284:1137, 1971.

33. Feldman AM and Brusilow SW: Effects of cholera toxin oncochlear endolymph production: model for endolymphatic hy-drops. Proc Natl Acad Sci USA 73:1761, 1976.

34. Epstein MH, Feldman AM, and Brusilow SW: Cerebrospinalfluid production: stimulation by cholera toxin. Science 196:1012,1977.

35. Hogan MJ, Alvarado JA, and Weddell JE: Ciliary body andposterior chamber. In Histology of the Human Eye, Philadelphia,WBSaunders, 1971, p. 282.

36. Tsukahara S and Maezawa N: Cytochemical localization of ad-enyl cyclase in the rabbit ciliary body. Exp Eye Res 26:99, 1978.


Forskolin lowers intraocular pressure by reducing aqueous ... · Forskolin Lowers Intraocular...

Documents

Transcript of Forskolin lowers intraocular pressure by reducing aqueous ... · Forskolin Lowers Intraocular...