Potent a
`2A-Adrenoceptor–Mediated Vasoconstriction by
`Brimonidine in Porcine Ciliary Arteries
`
`Anna Wikberg-Matsson1,2 and Ulf Simonsen3
`
`PURPOSE. An investigation into whether a
`2-adrenoceptor ago-
`nists induce contractions in the porcine ciliary arteries and to
`characterize the functional receptor subtype mediating these
`responses.
`METHODS. Isolated arteries from the intraocular part of the
`porcine ciliary artery were suspended in microvascular myo-
`graphs for isometric tension recording. The segments were
`contracted with the a
`2-adrenoceptor agonists brimonidine,
`apraclonidine, and oxymetazoline. To determine which sub-
`types of the a
`2-adrenoceptor mediate this contraction, antag-
`onists subselective for the different a
`2-adrenoceptors were
`added to the vessel bath before concentration–response curves
`for brimonidine were obtained. The following a
`2-adrenoceptor
`antagonists were applied: BRL44408 (a
`2A-selective), ARC239
`
`
`
`
`2C-selective), and prazosin (a(a2B- and a 2B- and a2C-selective).
`RESULTS. The a
`2-adrenoceptor agonists induced vasoconstric-
`tion in the porcine ciliary artery with the following potency
`order (EC50) expressed in nanomolar: brimonidine 2.11,
`oxymetazoline 5.26, and apraclonidine 13.0. As a reference,
`noradrenaline was tested, and its EC50 was determined to be
`247 nM in the ciliary artery. In the porcine ciliary arteries
`BRL44408, ARC239, and prazosin caused concentration-depen-
`dent and parallel rightward shifts of the concentration–re-
`sponse curves for brimonidine. Schild analyses for the antago-
`nists against brimonidine yielded regression lines with slopes
`of unity and functional antagonist potencies
`(pKB)
`for
`BRL44408 (7.8), ARC 239 (5.8) and for prazosin (6.0) suggest-
`ing the presence of functional a
`2A-adrenoceptors. Moreover,
`there was a good correlation of pKB with ligand-binding affinity
`
`(pKi) of the a2A-adrenoceptor in the porcine eye tissue.
`CONCLUSIONS. The a
`2-adrenoceptor agonists brimonidine, apra-
`clonidine, and oxymetazoline are potent vasoconstrictors in
`the porcine ciliary artery. In the present work, it was shown
`for the first time that the a
`2A-adrenoceptor subtype mediates
`this contraction.
`(Invest Ophthalmol Vis Sci. 2001;42:
`2049 –2055)
`
`T he a
`
`agonists brimonidine and apra-
`2-adrenoceptor
`clonidine are both powerful ocular hypotensive agents
`when applied topically. Brimonidine is established for treat-
`ment of primary open-angle glaucoma in patients with high
`intraocular pressure (IOP). Apraclonidine is mainly used to
`
`From the 1Department of Neuroscience, Ophthalmology, Univer-
`sity Hospital, Uppsala, Sweden; the 2Department of Neuroscience, Unit
`of Pharmacology, Uppsala University, Sweden; and the 3Department of
`Pharmacology, University of Aarhus, Denmark.
`Supported by Kronprinsessan Margaretas Arbetsna¨mnd fo¨r syn-
`skadade, Stockholm, Sweden (AW-M).
`Submitted for publication December 18, 2000; revised April 9,
`2001; accepted April 26, 2001.
`Commercial relationships policy: N.
`The publication costs of this article were defrayed in part by page
`charge payment. This article must therefore be marked “advertise-
`ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
`Corresponding author: Anna Wikberg-Matsson, Department of
`Neuroscience, Ophthalmology, University Hospital, SE-751 85 Uppsala,
`Sweden. anna.wikberg-matsson@ogon.uu.se
`
`Investigative Ophthalmology & Visual Science, August 2001, Vol. 42, No. 9
`Copyright © Association for Research in Vision and Ophthalmology
`
`prevent increased IOP after argon laser or YAG laser treatment
`in the anterior segment of the eye.1–3 Both apraclonidine and
`brimonidine lower the IOP by decreasing aqueous flow.4 – 6
`The mechanism of action of these drugs is thought to be by
`activation of the a
`2-adrenoceptors in the ciliary body, which
`decreases cyclic adenosine monophosphate (cAMP) levels,
`causing a decrease of the aqueous humor production.7 Al-
`though the main mechanism behind the IOP reduction of
`apraclonidine and brimonidine is a reduction of flow,5,8,9 it has
`been suggested that they may also have some effect on out-
`flow,10,11 but the detailed mechanisms have not yet been
`completely clarified.
`The posterior ciliary arteries of the eye are innervated by
`sympathetic nerve fibers and supply blood to the optic nerve,
`the choroid, the iris, and the ciliary body.12–15 It is generally
`
`
`2-adrenoceptors coexistaccepted that postjunctional a1- and a
`in the systemic vasculature.16,17 The a
`1-adrenoceptors cause
`vasoconstriction, whereas a
`2-adrenoceptors mediate vasodila-
`tion in large conductive arteries,18 but vasopressor a
`2-adreno-
`ceptors are well known in the resistance vasculature of both
`humans and animals.19 –21 In the eye, functional studies with
`different in vitro techniques have demonstrated a
`2-adrenocep-
`tors in the vascular bed. It has been shown by using the
`microsphere method that a
`2-adrenoceptor agonists decrease
`blood flow in the rabbit choroid and the ciliary body.22,23 The
`a
`2-adrenoceptors were verified in the bovine ciliary artery,
`where isolated segments of the intraocular part of the artery
`were studied on a small-vessel myograph.24
`On the basis of pharmacologic and molecular cloning evi-
`dence, a
`2-adrenoceptors have been divided into three sub-
`
`
`
`types: a2A, a2B, and a2C.25–27 All three subtypes have been
`cloned from the human, but only the a
`2A subtype has been
`cloned in the pig.28 We have investigated by pharmacologic
`characterization a
`2-adrenoceptor subtypes from the iris, the
`ciliary body, the retina, and the choroid of the porcine eye.29
`By using radioligand binding, we identified dense populations
`of the a
`2A-adrenoceptors in the choroid (900 femtomoles/mg)
`and the ciliary body (220 femtomoles/mg).
`Functional studies of the different a
`2-adrenoceptor sub-
`types have, to our knowledge, never been demonstrated in
`isolated segments of ocular arteries or veins. The posterior
`ciliary arteries supply blood to the choroid and the ciliary body
`of the eye. The purpose of the present study was to investigate
`whether some of the a
`2-adrenoceptor agonists, used in the
`clinical practice, induce vasoconstriction in the porcine ciliary
`arteries and to characterize the a
`2-adrenoceptor subtypes in-
`volved in these responses by use of selective antagonists: for
`
`
`
`2B- and athe a2A-adrenoceptor, BRL44408, and for the a 2C-
`adrenoceptors, ARC239 and prazosin.
`
`METHODS
`
`Preparation of Blood Vessels
`Fresh porcine eyes were obtained from the local abattoir. The ciliary
`arteries, located intraocularly before they penetrate the ciliary body,
`were identified, quickly dissected, and placed in Ca21-free physiolog-
`ical salt solution (PSS) at 4°C. Endothelium-intact ring segments (;2
`mm) were mounted on 40-mm wires in a small-vessel myograph for
`
`2049
`
`Downloaded from iovs.arvojournals.org on 07/29/2021
`
`Page 1 of 7
`
`SLAYBACK EXHIBIT 1017
`
`

`

`2050 Wikberg-Matsson and Simonsen
`
`IOVS, August 2001, Vol. 42, No. 9
`
`isometric recordings (J. P. Trading, Aarhus, Denmark).30 The experi-
`ments were performed in PSS of the following composition (in mM):
`119 NaCl, 4.7 KCl , 1.5 CaCl2, 1.17 MgSO4, 1.18 KH2PO4, 25 NaHCO3,
`0.027 EDTA, and 6 glucose. The preparations were allowed to equili-
`brate in oxygenated (95% O2 and 5% CO2) PSS at 37°C (pH 7.4), for
`approximately 30 minutes. The level of optimal passive tension for the
`porcine ciliary arteries was determined in preliminary experiments (n
`5 3) and defined as that tension at which the contraction to 125 mM
`KCl was maximal. The relationship between resting wall tension and
`the internal circumference was then determined, and the internal
`circumference, L60, corresponding to a transmural pressure of 60 mm
`Hg for a relaxed vessel, was calculated. A maximum level of 60 mm Hg
`was chosen, because higher pressure exposure destroyed the vessels.
`The vessels were set to the internal circumference L1, arrived at by L1
`5 0.9 3 L60. The effective internal lumen diameter was determined as
`5 L1/p.31 The final resting tension after normalization was found to
`l1
`be between 1.5 and 2.5 mN.
`The vessels were used either immediately or saved for the next day
`(and used within 36 hours). The preparations were kept in PSS in the
`refrigerator until used. There was no difference in response to a
`2-
`adrenoceptor agonists between these preparations and fresh prepara-
`tions that were used immediately after dissection.
`
`Experimental Procedure
`
`After normalization, the contractile ability of the vessels was tested by
`stimulating the arterial rings with 125 mM K-PSS (equivalent to PSS but
`with NaCl replaced with KCl on an equimolar basis, giving a final
`concentration of 125 mM K1), which was continued until reproduc-
`ible responses were recorded. After washout, the preparation was
`exposed to noradrenaline (1026 M) and allowed to contract for ap-
`proximately 5 minutes. The PSS contained propranolol (1026 M) to
`inhibit possible activity from b-adrenergic receptors and indomethacin
`(2.8 3 1026 M) to prevent synthesis of endogenous prostaglandins.
`Responses to a
`2-Adrenoceptor Agonists. Cumula-
`tive concentration–response curves for brimonidine, apra-
`clonidine, and oxymetazoline were constructed, and the ago-
`nists were added to the baths cumulatively in half-log incre-
`ments. Two ring segments of the same vessels were used in
`parallel at the same time: A concentration–response curve for
`brimonidine was performed on one segment and a concentra-
`tion–response curve for brimonidine in the presence of antag-
`onist was performed on the other segment. Unfortunately, it
`was not possible to repeat the experiment in the same vessel,
`because the contractions induced by the a
`2-adrenoceptor ago-
`nists were reduced in a second exposure, excluding the repe-
`tition of the concentration–response curve for any of the a
`2-
`adrenoceptor agonists. In the experiments with oxymetazoline
`the tissues were incubated with 1 3 1026 M rauwolscine for 5
`2-adrenoceptor before adding 3 3
`minutes to protect the a
`1026 M phenoxybenzamine for 20 minutes to alkylate the
`a
`1-adrenoceptors, according to the method of MacLennan et
`al.32 After washing every 5 minutes for at least 60 minutes the
`contraction of the ciliary arteries were tested with phenyleph-
`rine (1 3 1026 M). If the vessel responded to phenylephrine,
`it was excluded from the experiment. Concentration–response
`curves were then determined with oxymetazoline.
`Responses to Brimonidine in the Presence of Antago-
`nists. The effect of different a
`2-adrenoceptor subselective an-
`tagonists on the concentration–response curves
`for bri-
`monidine was evaluated. In these experiments the following
`antagonists were used: BRL44408 which is selective for the
`a
`2A-adrenoceptor; ARC239, which shows low affinity for the
`a
`
`
`2A-adrenoceptor and high affinity for the a2B- and a2C-adre-
`noceptors in the pig29,33; and prazosin, which shows low
`affinity for the a
`
`2A-adrenoceptor and high affinity for the a2B-
`and a
`2C-adrenoceptors.26 The concentrations of antagonists
`were as follows: ARC239, 1026 to 1024 M; prazosin, 1026 to
`
`1024 M; and BRL44408, 1028 to 1026 M. As a control, another
`segment of the same vessel or a segment from a parallel vessel
`of the porcine ciliary artery from the same eye was used. The
`antagonist was added 30 minutes before the cumulative agonist
`concentration curves were determined.
`
`Drugs
`Indomethacin, (2)-noradrenaline bitartrate, papaverine hydrochloride,
`L-phenylephrine hydrochloride, and propranolol were purchased from
`Sigma Chemical Co. (St. Louis, MO). Oxymetazoline hydrochloride,
`p-aminoclonidine (apraclonidine), phenoxybenzamine hydrochloride,
`prazosin hydrochloride, rauwolscine hydrochloride, and [5-bromo-N-
`(dihydro-1H-imidazol-2-yl)]-6-quinoxalinamine (UK14304, brimonidine)
`were from Research Biochemicals International (Natick, MA). 2-[(4,5-
`Dihydro-1H-imidazol-2-yl) methyl]-2,3-dihydro-1-methyl-1H-isoindole
`(BRL44408) was from Tocris Cockson Ltd, Bristol, UK. 2-[2-(4-(2-
`Methoxyphenyl) piperazin-1-yl)ethyl]-4,4-dimethyl-1,3(2H,4H)-isoquin-
`olindione (ARC239) was a gift from Karl Thomae GmbH, (Biberach,
`Germany).
`
`Analysis of Data
`The concentration–response curves were fitted to the Hill equation
`and calculated by nonlinear regression on a computer (Prism; Graph-
`Pad Software, San Diego, CA). Sensitivities to drugs were calculated on
`the basis of data from individual vessels and are expressed as EC50—
`that is, the agonist concentration needed to produce 50% of the
`maximal response. If antagonists produced parallel displacements of
`agonist concentration–response curves, Schild analysis was con-
`structed by use of least-squares linear regression of log (CR 2 1) against
`log antagonist concentration, where CR is the concentration ratio of
`the agonist in the absence and presence of antagonist.34 Concentration
`ratios were calculated at the EC50 level. Provided that the regression of
`the Schild plot is linear and that the slope is not significantly different
`from unity, pA2, which is the intercept along the abscissa scale of the
`Schild plot,
`is equal to the negative logarithm of the equilibrium
`5 2logKB
`5 pKB. The
`dissociation constant for the antagonist: pA2
`slope was also constrained to unity and more precise value of pKB was
`calculated. The solver function of the statistical analysis software (Ex-
`cel; Microsoft, Redmond, WA) was used to fit the model of linear
`regression.
`
`RESULTS
`
`The experiments were performed on ciliary arteries with an
`internal diameter of 246 6 27 mM (measured from 10 vessels).
`A concentration of 125 mM K-PSS induced a contraction of
`2.4 6 0.02 Nm (n 5 220) in these arteries.
`
`Agonist Studies
`As shown in Table 1 the rank order of potency of the agonists
`based on the EC50 values was brimonidine . oxymetazoline .
`apraclonidine @ noradrenaline. All agonists produced mono-
`phasic concentration–response curves in the porcine ciliary
`artery (Figs. 1, 2). In the experiments with oxymetazoline the
`tissues were first treated with 3 3 1026 M phenoxybenzamine
`in the presence of 1 3 1026 M rauwolscine to inactivate the
`a
`1-adrenoceptors (see the Methods section). The same method
`was also used in some pilot experiments with the porcine
`ciliary arteries to evaluate the effect of brimonidine in the
`presence of phenoxybenzamine. The mean of the calculated
`EC50 (n 5 4) from the concentration–response curves for
`brimonidine with phenoxybenzamine were 6.3 6 1.0 3 1029
`M and the maximal response (T) was 2.9 6 0.6 Nm. The
`maximal response of phenylephrine (1 3 1026 M) before
`adding phenoxybenzamine was 2.4 6 0.8 Nm, and after phe-
`noxybenzamine, phenylephrine (1 3 1026 M) merely raised
`tension (0.16 6 0.07 Nm). The control experiments were
`
`Downloaded from iovs.arvojournals.org on 07/29/2021
`
`Page 2 of 7
`
`SLAYBACK EXHIBIT 1017
`
`

`

`IOVS, August 2001, Vol. 42, No. 9
`
`Mediation of Vasoconstriction by a
`2A-Adrenoceptors
`
`2051
`
`TABLE 1. Potencies of Various Agonists in the Intraocular Part of the
`Porcine Ciliary Artery in the Presence of Propranolol (1 3 1026 M)
`and Indomethacin (2.8 3 1026 M)
`
`Agonist
`
`Noradrenaline
`Brimonidine
`Oxymetazoline
`Apraclonidine
`
`n
`
`6
`26
`12
`12
`
`EC50
`(31029 M)
`
`247 6 104*
`2.11 6 0.21
`5.26 6 1.11
`13.0 6 6.67
`
`Slope
`
`T (Nm21)
`
`1.0 6 0.13
`1.28 6 0.08
`1.67 6 0.25
`1.55 6 0.17
`
`3.3 6 0.8
`3.1 6 0.4
`2.2 6 0.4
`3.0 6 0.4
`
`Values are mean 6 SEM. n, number of vessel segments. The slope
`is determined from the Hill equation. EC50 is the concentration of
`agonist required to produce half-maximal contraction in the response
`obtained at the highest concentration of agonist applied. T (tension)
`describes the maximal response obtained with the respective sub-
`stance.
`* Denotes a significant-difference parameter of P , 0.001 versus
`brimonidine, paraminoclonidine, and oxymetazoline (Bonferroni test).
`In the experiments with oxymetazoline the vessels were incubated
`with 3 3 1026 M phenoxybenzamine.
`
`performed in exactly the same way but without incubating the
`vessel segments with phenoxybenzamine. The mean EC50 val-
`ues of the control sepcimens (n 5 4) were 6.4 6 2.1 3 1029
`M, and the maximal response was 2.5 6 0.5 Nm. The conclu-
`sion drawn from these results was that incubation with phe-
`noxybenzamine changed neither the EC50 nor the effect of
`response in the porcine ciliary arteries. Therefore, experiments
`with brimonidine were performed without prior incubation
`of phenoxybenzamine. All agonists induced comparable max-
`imal responses in the arterial segments of the ciliary artery
`(Table 1).
`
`Antagonist Studies
`To identify the subtypes of a
`2-adrenoceptors mediating the
`vasoconstrictive response of the porcine ciliary artery, concen-
`tration–response curves for brimonidine were performed in
`the presence of a
`2-adrenoceptor subtype–selective antago-
`nists. Increasing concentrations of the a
`2A-selective antagonist
`BRL44408 (1028, 33 1028, 1027, 33 1027, and 1026 M),
`caused parallel rightward shifts for the brimonidine concentra-
`tion–response curves without affecting the maximum re-
`sponse, indicating a competitive antagonism. Analysis of the
`data by Schild regression gave a slope (0.90 6 0.10) that was
`not significantly different from unity and pA2 of 7.85 6 0.12
`(Figs. 3A, 3D). The antagonist ARC239 (concentrations: 1026,
`3 3 1026, 1025, 33 1025, and 1024 M), produced rightward
`shifts of the concentration–response curves for brimonidine,
`with no depression of the maximum response. Analysis of the
`data by Schild regression gave a slope that was not significantly
`different from unity (0.98 6 0.10) and pA2 of 5.86 6 0.12 (Figs.
`3B, 3D). The antagonist prazosin (concentrations: 1026, 33
`1026, and 1025) produced rightward shifts of the concentra-
`tion–response curves for brimonidine, without reducing the
`maximum responses over the ranges used for analysis. Analysis
`of the data by Schild regression gave a slope that was not
`different from unity (1.0 6 0.10) and pA2 of 6.02 6 0.06 (Figs.
`3C, 3D). The Schild plots for all antagonists were constrained
`to unity giving pKB of ARC239 (5.84 6 0.12), BRL44408
`(7.77 6 0.12), and prazosin (6.02 6 0.07). The values for pA2
`and pKB obtained from the analysis of the concentration–
`response curves are given in Table 2.
`
`DISCUSSION
`
`The first interesting finding of the present study is the clear
`demonstration of the a
`2-adrenoceptor agonists’ being potent
`
`vasoconstrictors in the intraocular part of the porcine ciliary
`artery. It has been shown that apraclonidine and brimonidine,
`both of which are used clinically, have a 10- to 100-fold higher
`potency than noradrenaline. The agonist oxymetazoline was
`included in the study, because it has previously been shown to
`be a
`2A-adrenoceptor selective in porcine tissues.33 Because
`oxymetazoline has affinity for a
`
`1- as well as a2-adrenoceptors
`in the pig,35 the a
`1-adrenoceptors were inactivated by treat-
`ment of the vessels with phenoxybenzamine (3 3 1026 M) in
`the presence of rauwolscine (1 3 1026 M) according to Ma-
`cLennan et al.32 The choice of brimonidine as agonist in the
`antagonist experiments was based on the higher selectivity for
`a
`
`2-adrenoceptor versus a1-adrenoceptor (790-fold), compared
`with apraclonidine (100-fold) in binding assays.36,37
`The presence of a
`2-adrenoceptors in the ocular vessels of
`both the animal and the human is not a clear-cut fact, however.
`In vitro studies by Nyborg and Nielsen,24 who used the myo-
`graph technique on isolated vessels, have verified a
`2-adreno-
`ceptors in the intraocular part of the bovine ciliary artery.
`However, Yu et al.38 found only a
`1-adrenoceptors in the ex-
`traocular part of the human ciliary arteries. Retinal vessels have
`a smaller diameter (30 –35 mm) than the ciliary artery and
`consequently are technically more difficult to study in vitro.
`Therefore, most studies are performed with noninvasive tech-
`niques. However, Spada et al.39 studied the effect of a
`2-adre-
`noceptor agonists on human retinal vessels transplanted into
`hamster cheek pouch membrane, and vasoconstriction was
`observed with apraclonidine but not with brimonidine, which
`is more a
`
`2-adrenoceptor selective, suggesting that the a2-adre-
`noceptors do not act as vasoconstrictors there.
`Several studies in human ocular vessels have been per-
`formed with noninvasive techniques such as color laser Dopp-
`ler. Topically applied brimonidine has not shown any effect on
`the flow velocity in various ocular or retrobulbar vessels mea-
`sured with this technique.40,41 Still, because it is not known to
`what extent the topically applied drug reaches therapeutic
`concentrations in the retina or the retrobulbar tissues, these
`results do not exclude the presence of a
`2-adrenoceptors in
`those vessels.
`The variability in myogenic tone of the vessel segments
`could be another explanation for the variable response evoked
`by a
`2-adrenoceptors in vitro in eye arteries. Dunn et al.42
`suggest that the a
`2-adrenoceptors are only activated in the
`presence of a small increase in tone of the vessels. In the work
`by Nyborg and Nielsen24 the vessels were precontracted with
`30 mM K1 or prostaglandin F2a to raise the myogenic tone
`before adding the a
`2-adrenoceptor agonist, which was not the
`case in the study by Yu et al.38 In the porcine ciliary arteries,
`raising basal tone by increasing the extracellular K1 concen-
`tration did not change the contractile responses induced by
`brimonidine, compared with responses obtained in physiolog-
`ical buffer (Wikberg-Matsson, unpublished observation, 2001).
`Therefore, we did not precontract the vessels before the a
`2-
`adrenoceptor agonist was added.
`The regional differences of a
`2-adrenoceptor distribution
`throughout the different parts of the vascular bed is a well
`known fact.43 One possible reason that a
`2-adrenoceptor ago-
`nists induce a powerful vasoconstriction in the isolated ciliary
`arteries and not in the retinal vessels is that the ciliary arteries
`contain higher amounts of a
`2-adrenoceptors than do other
`smaller vessels of the eye. Earlier studies support this theory:
`Resistance arteries (to which the ciliary arteries belong) show
`well-pronounced contractions in response to a
`2-adrenoceptor
`stimulation, and there is also a correlation between the diam-
`eter of the vessel and the response by a
`2-adrenoceptor ago-
`nists, both in humans and animals.20,21,44
`In summary, the information on a
`2-adrenoceptors in the
`ocular vessels in mammals is scattered and inconsistent for
`
`Downloaded from iovs.arvojournals.org on 07/29/2021
`
`Page 3 of 7
`
`SLAYBACK EXHIBIT 1017
`
`

`

`2052 Wikberg-Matsson and Simonsen
`
`IOVS, August 2001, Vol. 42, No. 9
`
`FIGURE 1. Original trace recordings
`(A) bri-
`showing the effect of
`(B) apraclonidine,
`(C)
`monidine,
`oxymetazoline, and (D) noradrena-
`line (Na) at
`increasing concentra-
`tions (in half-log units indicated by
`arrows) of the agonists in the por-
`cine ciliary artery. The contractility
`of the vessels are first tested with 125
`mM K-PSS and 1 3 1026 M noradren-
`aline. The experiments were per-
`formed in the presence of 1026 M
`propranolol and 2.8 3 1026 M indo-
`methacin. In the experiment with
`oxymetazoline (C), 3 3 1026 M phe-
`noxybenzamine (b) was added 5
`minutes after 1 3 1026 M rauwol-
`scine (a), and the contractility of the
`vessel was also tested with 1 3 1026
`M phenylephrine (Phe) before and
`after the treatment with phenoxy-
`benzamine. w, wash.
`
`several reasons. There are methodological aspects (i.e., it is
`difficult to compare the in vitro techniques with the noninva-
`sive techniques), the distribution of the a
`2-adrenoceptors in
`the vascular beds vary, and the distribution of a
`2-adrenocep-
`tors also may differ to some extent between species.
`
`The second important finding of the present study was that
`the a
`2A-adrenoceptors mediate vasoconstriction in the exam-
`ined intraocular part of the porcine ciliary artery. This conclu-
`sion is based on the fact that the selective a
`2A-adrenoceptor
`antagonist BRL44408 produced, at low concentration, a dextral
`
`Downloaded from iovs.arvojournals.org on 07/29/2021
`
`Page 4 of 7
`
`SLAYBACK EXHIBIT 1017
`
`

`

`IOVS, August 2001, Vol. 42, No. 9
`
`Mediation of Vasoconstriction by a
`2A-Adrenoceptors
`
`2053
`
`has been shown by other authors to have low affinity for the
`a
`
`
`2A-adrenoceptor and high for a2B- and a2C-adrenocep-
`tors.26,45 The Kd values (in this study recalculated to pKi)
`obtained from the ligand binding studies give an accurate guide
`to the affinities of the antagonists at the respective subtypes of
`receptor. Given in Table 2 are the pKB values obtained from
`concentration–response curves with subselective antagonists
`in the present study and pKi values of the affinities obtained
`from binding experiments in the earlier studies. The good
`
`correlation of pKi and the pKB implies that the a2A-adrenocep-
`tor is responsible for the contraction induced by a
`2-agonists.
`Our conclusion that the a
`2A-adrenoceptor mediates con-
`traction in the porcine ciliary artery is not a great surprise in
`the view of the earlier study of porcine tissue homogenates of
`the iris, choroid, and ciliary body which also demonstrated
`only the a
`2A-adrenoceptor. From the pig retina, however, both
`the a
`
`2A- and a2C-adrenoceptors were detected in lower densi-
`ties (20 and 3.6 femtomoles/mg, respectively).29 Radioligand
`binding has also been used to characterize a
`2-adrenoceptor
`subtypes in other species: in the ciliary body of the rabbit46
`and in the ciliary body, iris, choroid, and retina of the cow.47,48
`Comparable to the binding studies in animals, a study of human
`ocular tissue homogenates showed only the a
`2A-adrenoceptor
`in the human iris, ciliary body, and choroid.49 According to the
`results in both animals and humans, binding studies indicate
`that the predominant subtype is the a
`2A subtype in the richly
`vascularized tissues of the eye, such as the choroid and the
`ciliary body. This suggests that the human ciliary vessels also
`contain a
`2A-adrenoceptors.
`In contrast to these binding data, immunofluorescence la-
`beling of the human ciliary body indicates the presence of a
`2B-
`and a
`
`2C-adrenoceptor subtypes, but not the a2A subtype.50
`Similarly, studies using polymerase chain reaction (PCR) sug-
`gest the presence of the a
`
`2B and a2C subtypes, but not the a
`2A
`subtype in a transformed cell line of human nonpigmented
`epithelium.50
`Which of the three a
`2-adrenoceptor subtypes is coupled to
`the inhibition of aqueous humor production in the human eye
`is not known, however. Furthermore, it remains to be evalu-
`ated whether the a
`2A-adrenoceptors are involved in the vaso-
`
`FIGURE 2. Cumulative concentration–response curves for the vaso-
`constrictive effects of brimonidine (M), apraclonidine ((cid:141)), oxymeta-
`zoline (E), and noradrenaline (f) in the presence of 1 3 1026 M
`propranolol and 2.8 3 1026 M indomethacin. Values are means 6 SEM
`(n 5 6 –23 preparations). The concentration–response curves for
`oxymetazoline were obtained after incubation with phenoxyben-
`zamine (3 3 1026 M).
`
`displacement of the concentration–response curve for bri-
`monidine. Significant parallel displacement occurred at high
`concentrations of ARC239 and prazosin, both of which are
`antagonists with low affinity for a
`2A-adrenoceptor and high
`affinity for the a
`
`2B- and a2C-adrenoceptor.
`In two previous studies the a
`2-adrenoceptor subtypes were
`characterized pharmacologically in major organs including eye
`tissues of the pig.29,33 By using radioligand binding, we iden-
`tified dense populations of a
`2A-adrenoceptors in the choroid
`(900 femtomoles/mg) and ciliary body (270 femtomoles/mg),
`but less density in the iris (87 femtomoles/mg). Furthermore,
`drug affinities for selective antagonists were evaluated. Among
`these, BRL44408 was shown to be selective for the pig a
`2A-
`adrenoceptor, whereas ARC239 showed low affinity for the
`a
`
`
`2A-adrenoceptor and high for the a2B- and a2C-adrenoceptors
`in the pig. Prazosin which was included in the present study
`
`FIGURE 3.
`Cumulative concentra-
`tion–response curves
`for brimon-
`idine (M) and presence of
`(A)
`BRL44408 in concentrations of 1028
`M (f), 3 3 1028 M (E), 1027 M (F),
`3 3 1027 M ((cid:130)), and 1026 M ((cid:140)); (B)
`ARC239 at 3 3 1026 M (f), 1025 M
`(E), 3 3 1025 M (F), and 1024 M
`((cid:130)); (C) prazosin, at 1026 M (f), 3 3
`1026 M (E), and 1025 M (F); and (D)
`the Schild regression for the effect
`of BRL44408 (M), ARC239 (f), and
`prazosin ((cid:131)) on concentration–re-
`sponse curves of brimonidine in the
`porcine ciliary artery. The experi-
`ments were performed in the pres-
`ence of 1026 M propranolol and
`2.8 3 1026 M indomethacin. The re-
`sults are means 6 SEM of four to
`seven experiments.
`
`Downloaded from iovs.arvojournals.org on 07/29/2021
`
`Page 5 of 7
`
`SLAYBACK EXHIBIT 1017
`
`

`

`2054 Wikberg-Matsson and Simonsen
`
`IOVS, August 2001, Vol. 42, No. 9
`
`TABLE 2. Effect of a
`2-Adrenoceptor–Selective Antagonists on Brimonidine-Induced Contraction of
`Porcine Ciliary Artery and Drug Affinities in Binding Experiments
`
`Antagonist
`
`BRL44408
`ARC239
`
`Prazosin
`
`n
`
`7
`7
`
`4
`
`pA2
`
`pKB
`
`Slope
`
`pKi-a
`2A
`Choroid*
`
`pKi-a
`2B
`Kidney*
`
`pKi-a
`2C
`Retina*
`
`7.85 6 0.12
`5.86 6 0.12
`
`7.77 6 0.12
`5.84 6 0.12
`
`0.90 6 0.10
`0.98 6 0.10
`
`8.13
`6.03
`
`6.31
`7.41
`
`6.46
`6.92
`
`6.02 6 0.06
`
`6.02 6 0.07
`
`1.01 6 0.10
`
`5.9
`
`7.5
`
`7.2
`
`Human Tissues or Cell Lines†
`
`The values are expressed as means 6 SEM. n, number of vessel segments. pA2 is the intercept at the
`abscissae of the regression line in Schild plots. pKB is the negative logarithm of the equilibrium dissoci-
`ation constant of the antagonist–receptor complex, calculated from a plot constrained to a slope of unity.
`* pKi values calculated from Kd values obtained in binding studies in porcine tissues.29,33
`† Mean affinities expressed as pKi values from different human tissues.26
`
`constriction of human ciliary arteries. Assuming that different
`a
`2-adrenoceptor subtypes are coupled to different physiologi-
`cal responses in the human eye, this fact would be of impor-
`tance in designing novel, more selective drugs with greater
`effectiveness and fewer adverse effects.
`In summary, we have shown in the present study that the
`a
`2-adrenoceptor agonists induce a powerful contraction of the
`anterior intraocular part of the posterior porcine ciliary artery
`and that the vasoconstrictive effect is mediated by the a
`2A-
`adrenoceptor.
`
`Acknowledgments
`
`The authors thank Johan Stjernschantz at the Department of Neuro-
`science, Unit of Pharmacology, Uppsala University, Uppsala, Sweden,
`for all facilities placed at their disposal and for valuable discussions,
`and Albert Alm for helpful discussions.
`
`References
`
`1. Robin AL, Pollack IP, House B, Enger C. Effects of ALO2145 on
`intraocular pressure following argon laser trabeculoplasty. Arch
`Ophthalmol. 1987;105:646 – 650.
`2. Robin AL, Pollack IP, deFaller JM. Effects of topical ALO 2145
`(p-aminoclonidine hydrochloride) on the acute intraocular pres-
`sure rise after argon laser iridotomy. Arch Ophthalmol. 1987;105:
`1208 –1211.
`3. Brown RH, Stewart RH, Lynch MG, et al. ALO 2145 reduces the
`intraocular pressure elevation after anterior segment laser surgery.
`Ophthalmology. 1988;95:378 –384.
`4. Burke JA, Potter DE. Ocular effects of a relatively selective a
`2
`agonist (UK14, 304-18) in cats, rabbits and monkeys. Curr Eye Res.
`1986;5:665– 676.
`5. Gharagozloo NZ, Relf SJ, Brubaker RF. Aqueous flow is reduced by
`the alpha-adrenergic agonist, apraclonidine hydrochloride (ALO
`2145). Ophthalmology. 1988;95:1217–1220.
`6. Serle JB, Steidl S, Wang RF, Mittag TW, Podos SM. Selective a
`2-
`adrenergic agonists B-HT 920 and UK14304-18: effects on aqueous
`humor dynamics in monkeys. Arch Ophthalmol. 1991;109:1158 –
`1162.
`7. Mittag TW, Tormay A. Drug responses of adenylate cyclase in iris
`ciliary body determined by adenine labeling. Invest Ophthalmol
`Vis Sci. 1985;26:396 –399.
`8. Maus TL, Nau C, Brubaker RF. Comparison of the early effects of
`brimonidine and apraclonidine as topical ocular hypotensive
`agents. Arch Ophthalmol. 1999;117:586 –591.
`9. Schadlu R, Maus TL, Nau CB, Brubaker RF. Comparison of the
`efficacy of apraclonidine and brimonidine as aqueous suppressants
`in humans. Arch Ophthalmol. 1998;116:1441–1444.
`10. Toris CB, Tafoya ME, Camras CB, Yablonski ME. Effects of apra-
`clonidine on aqueous humor dynamics in human eyes. Ophthal-
`mology. 1995;102:456 – 461.
`
`11. Toris CB, Gleason ML, Camras CB, Yablonski ME. Effects of bri-
`monidine on aqueous humor dynamics in human eyes. Arch Oph-
`thalmol. 1995;113:1514 –1517.
`12. Chandra SR, Friedman E. Choroidal blood flow. II: the effects of
`autonomic agents. Arch Ophthalmol. 1972;87:67– 69.
`13. Alm A, Bill A. The effect of stimulation of the cervical sympathetic
`chain on retinal oxyg