0022-3666/93/2673-1126103.00/0
`Tits JOURNAL or PHARMACOLOGY AND ExPltlUMBNTAL THERAPEUTICS
`Copyrisbt C 1993 by The American Society for PbarmacoloSY and E"P"rimental Tberepeutica
`
`Vol. 267, No. 3
`Prinl«l in U.S.A.
`
`Binding and Functional Characterization of Alpha-2 Adrenergic
`Receptor Subtypes on Pig Vascular Endothelium1
`
`CHARLES S BOCKMAN, WILLIAM B. JEFFRIES and PETER W. ABEL
`Department of Pharmacology, Creighton University School of Medicine, Omaha, Nebraska
`Accepted for publication August 2, 1993
`
`subtypes identified in other tissues. Vascular endothelium con(cid:173)
`
` 
`tained 23% a/pha-2A and 77% a/pha-2C adrenergic receptors.
`

`In the presence of indomethacin, the rank order of potency for
`
`agonists that cause endothelium-dependent vascular relaxation
` 
`
`was p-iodoclonidine > clonidine > UK-14,304 > guanabenz >
`ÿ
`epinephrine> norepinephrine. Ks values for antagonist inhibition
`of epinephrine-induced, endothelium-dependent vascular relax(cid:173)
`ation correlated best with K, values for antagonist binding at the
`a/pha-2A adrenergic receptor subtype. These results suggest
`that the alpha-2A and alpha-2C adrenergic receptor subtypes
`are present on pig vascular endothelium and that the alpha-2A
`adrenergic receptor subtype mediates indomethacin-insensitive,
`endothelium-dependent relaxation of pig epicardial coronary ar(cid:173)
`teries.
`
`ÿ
`
` 
`

 
`
`ÿ ÿ
`
`ÿ
`

ÿ

ÿ ÿÿ
`
`Three of the four alpha-2 ADR subtypes have been cloned from
`the human genome and the binding characteristics of the
`expressed receptors confirm the pharmacological classification
`of alpho-2A, alpho-2B and alpho-2C ADR subtypes (Bylund et
`al., 1992).
`Functional studies suggest the presence of alpho-2 ADRs on
`vascular endothelium. For example, in isolated coronary arter(cid:173)
`ies, norepinephrine and clonidine cause endothelium-depend(cid:173)
`ent relaxation that is inhibited by selective alpho-2 ADR an(cid:173)
`tagonists such as rauwolscine (Cocks and Angus, 1983; Van(cid:173)
`houtte and Miller, 1989). Similar results were found using
`several different isolated arteries (Angus et al., 1986) and veins
`(Miller and Vanhoutte, 1985). It has also been reported that
`endothelium-dependent, alpho-2 ADR-induced relaxation of
`pig coronary arteries is mediated by an endothelium-derived
`relaxing factor, nitric oxide (Richard et al., 1990). We recently
`reported that norepinephrine-induced release of an endothe(cid:173)
`lium-derived relaxing factor, nitric oxide, is enhanced in min(cid:173)
`eralocorticoid hypertension (Bockman et al., 1992). Taken to(cid:173)
`gether, these results suggest that endothelial alpho-2 ADRs
`may play an important role in regulating vascular tone in both
`normal and pathological conditions.
`The alpho-2 ADR subtypes present on vascular endothelium
`
`ABSTRACT
`Alpha-2 adrenergic receptor subtypes were characterized in
`membranes of pig vascular endothelium using (3H]rauwolscine.
`Alpha-2 adrenergic receptor subtypes that mediate endothelium(cid:173)
`dependent vascular relaxation were studied In vitro by using ring
`segments of pig epicardial coronary arteries. Specific (3H]rau(cid:173)
`wolscine binding in endothelial membranes was saturable and to
`a single class of high-affinity sites with a mean Ko of 0.217 ±
`0.05 nM and 8,,...,,_ of 156 ± 28 fmol/mg of protein. Nonlinear
`regression analysis indicated that competition binding curves for
`drugs that distinguish the a/pha-2A adrenergic receptor subtype
`from the alpha-2C adrenergic receptor subtype fit best to two(cid:173)
`site binding models. K, values for drugs in binding to endothelial
`alpha-2 adrenergic receptors correlated well with their K, values
`for alpha-2A (r = .98) and a/pha-2C (r = .97) adrenergic receptor
`
`It is now known that there are four alpha-2 ADR subtypes
`that can be distinguished from one another based on affinity
`values for subtype-selective alpha-2 ADR antagonists. For ex(cid:173)
`ample, the rank order of potency of subtype-selective drugs for
`inhibiting specific [3H)rauwolscine binding to three of the
`alpho-2 ADR subtypes is oxymetazoline > BAM-1303 (811((2-
`phenylimidazol-1-yl)methyl)-6-methylergoline) > spiroxatrine
`> ARC-239 l2-[2,4-(O-methoxyphenyl)-piperazin-1-yl)ethyl-
`4,4dimethyl-l,3-(2H4H)-isoquinolindione-HC11 for the alpho-
`2A ADR, spiroxatrine > ARC-239 > BAM-1303 > oxymetazo(cid:173)
`line for the alpho-2B ADR and BAM-1303 > spiroxatrine >
`ARC-239 > oxymetazoline for the alpho-2C ADR (Blaxall et
`al., 1991). In addition, it is now known that the alpho-2D ADR
`is a species homologue similar to the alpho-2A ADR (Bylund
`et al., 1992). Because no single drug differentiates the alpho-2A
`ADR subtype from the alpho-2B ADR subtype from the alpho-
`2C ADR subtype from the alpho-2D ADR subtype, it is neces(cid:173)
`sary to use a combination of drugs to obtain pharmacological
`profiles characteristic of an individual alpho-2 ADR subtype.
`
`Received for publication February 12, 1993.
`1 Thia work was supported by the American Heart Association- Nebraska
`Affiliate.
`
`ABBREVIATIONS: ADA, adrenergic receptor; Ks, functional equilibrium dissociation constant; K,. inhibition binding equilibrium dissociation constant;
`K,.,, high-affinity inhibition binding equilibrium dissociation constant; Ku., low-affinity inhibition binding equilibrium dissociation constant.
`
`1121
`
`Slayback Exhibit 1094, Page 1 of 8
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`1993
`
`Endothelial Alpha-2 Adrenoceptors
`
`1127
`
`ments. K, values were calculated from I Cao values by using the method
`of Cheng and Prusoff (1973). All values are given as means ± S.E. The
`F test was used to determine whether or not the binding data fit best
`to a one- or two-site binding model. A value of P < .05 was used to
`conclude that the two-site model fit the data best.
`Measurements of endothelium-dependent relaxation of cor(cid:173)
`onary arteries. The proximal third of the left circumflex coronary
`arteries was dissected from pig hearts obtained from a local slaughter(cid:173)
`house. The arteries were cleaned of fat and adhering tissue, cut into 3-
`mm long ring segments and mounted between two stainless steel pins
`passed through the lumen of the rings. Ring segments were placed in
`water-jacketed glass muscle chambers that contained Krebs' solution
`maintained at 37°C gassed with 95% OJ5% CO2, pH 7.4. One pin was
`attached to a Grass FT.03 force transducer for measurement of iso(cid:173)
`metric tension with a Grass polygraph (Quincy, MA). After a 1-hr
`equilibration period at 6 g of resting tension (determined to be optimum
`in preliminary length-tension experiments), ring segments were con(cid:173)
`tracted with a maximal contractile concentration (45 mM) of KCl and
`then washed for 30 min. Ring segments were then contracted to one(cid:173)
`
`half of maximal contraction with KCl and, when the response reached
` 
`a stable level of tone, cumulative concentration-response relaxation
`

`
`curves for agonists were generated. Relaxation-response curves were
`obtained in the presence of 10 µ,M indomethacin to inhibit production
`of cyclooxygenase products, 0.1 µ,M desipramine to inhibit neuronal
`uptake of catecholamines, 0.1 µ,M prazosin to block alpha-1 ADR(cid:173)
`mediated contraction and 10 µ,M nadolol to block beta ADR-mediated
`ÿ
`relaxation. To quantify potency values of agonists, the ECao for the
`relaxation responses was calculated from concentration-response
`curves by linear regression of all points between 20% and 80% of the
`maximal response. ECao values are expressed as means ± S.E.
`Functional determination of antagonist aff'mity values. The
`pA2 value for the selective alpha-2 ADR antagonist rauwolscine was
`calculated as described by Arunlakshana and Schild (1959). After the
`determination of control concentration-response curves for epineph(cid:173)
`rine, the tissues were washed and equilibrated for 1 hr with 10 nM
`rauwolscine. Then epinephrine concentration-response curves were
`repeated. The tissues were then thoroughly washed with Krebs' solution
`and equilibrated for 1 hr with 30 nM rauwolscine before repeating the
`epinephrine concentration-response curves. This same sequence was
`repeated again using 100 nM rauwolscine. Dose ratios of ECao values
`in the presence and absence of rauwolscine were calculated and Schild
`plots were constructed by plotting the log of the dose ratio minus 1
`versus the log of the concentration of rauwolscine. Linear regression of
`the plotted points was used to determine the x-intercept (pA2). The
`slopes of the Schild regressions were considered different from unity if
`the 95% confidence interval did not include the value of 1. The pA2
`values were converted to their antilogs and expressed as Ka values. The
`Ka values for the alpha adrenergic antagonists, ARC-239, prazosin,
`SKF-104856 and spiroxatrine, were calculated as described by Mackay
`(1978). After the determination of control concentration-response
`curves for epinephrine, the tissues were washed and equilibrated for 1
`hr with a single concentration of antagonist and epinephrine concen(cid:173)
`tration-response curves were repeated. Antagonist concentrations, pre(cid:173)
`dicted to cause 10-fold shifts in concentration-response curves, were
`chosen based on the data from preliminary experiments. The Ka values
`were determined from the following equation: log Ka = log [antagonist]
`-
`log (dose ratio - 1), where the dose ratio= ECao in the presence of
`antagonist divided by ECao in the absence of antagonist. Time control
`experiments were performed and showed that concentration-response
`curves for epinephrine did not change over the time required to conduct
`antagonist affinity determinations.
`
` 
`

 
`
`ÿ ÿ
`
` ÿ 
`
`ÿ
`

ÿ

ÿ ÿÿ
`
`Results
`Radioligand binding experiments.
`[3H]Rauwolscine
`binding in vascular endothelial membranes is shown in figure
`lA. Nonlinear regression analysis of individual saturation bind(cid:173)
`ing isotherms indicated that [3H)rauwolscine bound with high
`
`are unknown. Thus, we characterized the alpha-2 ADR sub(cid:173)
`types present on pig vascular endothelium by determining the
`affinities of several alpha-2 ADR subtype-selective drugs for
`inhibiting specific [3H]rauwolscine binding. We also deter(cid:173)
`mined the affinities of alpha-2 ADR subtype-selective drugs in
`inhibiting endothelium-dependent, alpha-2 ADR-mediated re(cid:173)
`laxation of pig coronary arteries. Furthermore, we correlated
`the binding affinities with the affinities obtained from func(cid:173)
`tional studies to determine which alpha-2 ADR subtype me(cid:173)
`diates endothelium-dependent vascular relaxation.
`
`Methods
`
`Drup. The drugs used were obtained from the following sources:
`oxymetazoline HCl, indomethacin, (-)-epinephrine bitartrate, (-)(cid:173)
`norepinephrine bitartrate, clonidine HCl, nadolol, desipramine HCl
`and guanabenz (Sigma, St. Louis, MO); rauwolscine HCl, spiroxatrine,
`p-iodoclonidine HCl, prazosin HCl and phentolamine mesylate (Re(cid:173)
`search Biochemicals, Natick, MA); UK 14,304 (5-bromo-6-(2-imidazo(cid:173)
`line-2-yl-amino]quinoxaline) (Pfizer Central Research, Sandwich, Eng(cid:173)
`land); ARC-239 (Thomae, Biberach, Germany); SKF 104856 (2-vinyl-
`7-chloro-3,4,5,6-tetrahydro-4-methyl-thieno[ 4,3,2ef] [ 3]benzazepine)
`(SmithKline Beecham Pharmaceuticals, King of Prussia, PA); BAM-
`1303 (kindly provided by Dr. David Bylund, Department of Pharma(cid:173)
`cology, University of Nebraska Medical Center, Omaha, NE); and (3H)
`rauwolscine (78 Ci/mmol; New England Nuclear, Boston, MA).
`Endothelial membrane preparation. Pig thoracic aortas were
`obtained from a local slaughterhouse, cleaned of fat and connective
`tissue and placed into ice-cold Krebs' solution (in millimolar composi(cid:173)
`tion: NaCl, 120; KCl, 5.5; CaCl2, 2.5; NaH1PO,, 1.4; MgCl2, 1.2; Na(cid:173)
`HCOa. 20; dextrose, 11.1; CaNa2-EDTA, 0.027) equilibrated with 95%
`0,/5% CO2 for transport to the laboratory. The aortas were cut longi(cid:173)
`tudinally and pinned down on Styrofoam backing. The luminal surface
`was washed with ice-cold isotonic phosphate-buffered saline containing
`137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO, and 1.5 mM KH2PO,,
`pH 7.6. Luminal scrapings of vascular endothelium were obtained by
`passing a number 10 scalpel blade, held at a 45° angle, over the luminal
`surface of the aorta. The scrapings were pooled in 10 ml of ice-cold
`phosphate-buffered saline for washing by centrifugation at 3000 x g at
`4°C for 15 min. The pellet was resuspended in 10 ml of 50 mM Tris
`HCl, pH 8.0, and homogenized with a Tekmar Tissuemizer (Cincinnati,
`OH) at a setting of 80 for 20 sec. The homogenate was centrifuged at
`35,000 X g at 4°C for 20 min. The pellet was stored at -80°C.
`Radiolitrand binding 88811y&. The pellet was resuspended and
`homogenized in 200 volumes of 25 mM glycylglycine buffer, pH 7.6.
`For saturation binding experiments, total [3H]rauwolscine binding was
`determined using duplicate tubes containing 970 µ,l of membrane sus(cid:173)
`pension, 10 µ,l of 5 mM HCl and 20 µ,l of [3H]rauwolscine, which ranged
`in concentration from 0.02 to 2 nM. To a parallel set of duplicate tubes,
`10 µ,l of 100 µ,M (-)-norepinephrine in 5 mM HCl were added to
`determine nonspecific binding. After a 45-min incubation in a shaking
`water bath at 25°C, membrane suspensions were filtered through GF/
`B glaas fiber filter strips (Whatman, Clifton, NJ) using a 48-sample
`cell harvester (Brandel, Gaithersburg, MD). Tubes and filters were
`washed four times with 5 ml of ice-cold 50 mM Tris HCl (pH 8.0) and
`radioactivity retained on the filters counted by liquid scintillation
`spectroscopy. Specific binding was calculated as the difference between
`total and nonspecific binding. For competition binding experiments,
`duplicate tubes containing 970 µ,l of membrane suspension, 20 µ,l of
`[ 8H]rauwolscine (0.1 nM final concentration) and 10 µ,l of increasing
`concentrations of various unlabeled drugs were incubated and processed
`as for saturation experiments. The protein concentration was deter(cid:173)
`mined by the method of Lowry et al. (1951) by using bovine serum
`albumin as the standard.
`The binding data were analyzed using a nonlinear least-squares
`curve-fitting program to determine Ko and B,_ values from saturation
`binding experiments and ICao values from competition binding experi-
`
`Slayback Exhibit 1094, Page 2 of 8
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`1128
`
`Bockman et al.
`
`A
`
`25
`
`• Specific: lllncllng
`o Total Binding
`
`& Nonap■clflc: lllncllng
`
`Vol. 267
`
`• AauNol■cln•
`& Splroxatrlne
`• 0xy11■tazollne
`■ AAC-2311
`
`-6
`-8
`- 0
`Log M [Antagonist]
`
`-4
`
`Fig. 2. Mean competition binding curves showing a/pha-2 adrenerglc
`antagonists inhibition of specific rHJrauwolsclne binding. For each con(cid:173)
`centration of antagonist, rHJrauwolsclne binding Is expressed as a
`
`percentage of the specific binding in the absence of any drug. One- and
` 
`two-site binding models were flt to Individual competition binding curves
`

`
`by using a nonHnear least-squares curv&-fitting program to obtain K,
` 
`values. Competition binding curves for rauwolsclne flt best to a one-site
`
`binding model; competition binding curves for splroxatrlne, oxymetazo(cid:173)
`ÿ
`line and ARC-239 flt best to a two-site binding model.
`
`branes was consistent with [3H]rauwolscine binding to alpha-2
`ADRs (table 1).
`We determined K1 values (table 1) for SKF-104856 and
`BAM-1303 in inhibiting specific [3H]rauwolscine binding in
`endothelial membranes to determine whether or not the alpha-
`2B ADR subtype was present. SKF-104856 differentiates the
`alpha-2B ADR subtype (K, < 10 nM) from the alpha-2A and
`ÿ ÿ
`alpha-2C ADR subtypes (KJB = 50-200 nM; Gleason and Bie(cid:173)
`hle, 1991; Simonneauxetal., 1991). However, SKF-104856does
`not differentiate the alpha-2A ADR subtype from the alpha-2C
`ADR subtype. The K, value for SKF-104856 in binding to
`alpha-2 ADRs on endothelial membranes was 54.5 nM, con(cid:173)
`sistent with binding to either the alpha-2A ADR, alpha-2C
`ADR or both receptor subtypes. Similar to SKF-104856, BAM-
`1303 also identifies the alpha-2B ADR subtype (K1 = 7-26 nM)
`from the alpha-2A and alpha-2C ADR subtypes (KJB = 0.4-2.5
`nM; Blaxall et al., 1991; Bylund et al., 1992). The K, value for
`BAM-1303 in binding to alpha-2 ADRs on endothelial mem(cid:173)
`branes was 0.394 nM. These results suggest the presence of the
`alpha-2A or alpha-2C ADR subtypes, or both, on endothelial
`membranes. K, values for SKF-104856 and BAM-1303 also
`suggest that the alpha-2B ADR subtype is not present on pig
`vascular endothelium.
`K, values for rauwolscine and BAM-1303 in inhibiting spe(cid:173)
`cific [8H]rauwolscine binding in endothelial membranes were
`obtained to determine whether or not the alpha-2D ADR sub(cid:173)
`type was present. Because rauwolscine and BAM-1303 have
`relatively low affinities (K, = 3.4 nM and 53 nM, respectively;
`Simonneaux et al., 1991) for the alpha-2D ADR subtype com(cid:173)
`pared with the alpha-2A, alpha-2B and alpha-2C ADR subtypes,
`both drugs can distinguish the alpha-2D ADR subtype from the
`other three alpha-2 ADR subtypes. Our high-affinity K 1 values
`for rauwolscine (0.261 nM) and BAM-1303 (0.394 nM) in
`inhibiting specific [3H]rauwolscine binding in endothelial mem(cid:173)
`branes indicate that the alpha-2D ADR subtype is not present
`on pig vascular endothelial membranes.
`We also determined K, values (table 1) from competition
`binding experiments (fig. 2) for ARC-239, spiroxatrine and
`
`ÿ
`

ÿ

ÿ ÿÿ
`
`ÿ
`
` 
`

 
`
`0
`
`0.5
`1.5
`1.0
`[FrH :11H-Rauwolsclne], nM
`
`2.0
`
`B
`
`0.
`
`25
`
`125
`100
`75
`50
`Bound (f1110l/11g protein)
`
`150
`
`Fig. 1. A) Illustrates mean saturation binding isotherms for ['H]rauwol(cid:173)
`sclne In pig vascular endothelial membranes. Concentrations of free ['HJ
`rauwolsclne, ranging from 0.02 to 2.0 nM, were incubated with endothe(cid:173)
`llal membranes (0.150 mg/ml of assay volume) in the presence (nonspe(cid:173)
`cific binding) and absence (total binding) of 100 µM (-)-norepinephrlne.
`Nonlinear regression analysis of six Individual specific binding Isotherms
`indicated that rHJrauwolsclne bound to a homogeneous population of
`binding sites with a mean Ko value of 0.217 ± 0.05 nM and S,_of 156
`± 28 fmol/mg of protein. Specific rHJrauwolsclne binding was 700/o to
`90% of total binding. B) Shows a mean Rosenthal plot derived from six
`individual experiments.
`
`affinity to a single class of binding sites. The mean KD value
`for [1H]rauwolscine in binding to alpha-2 ADRs on endothelial
`membranes was 0.217 ± 0.05 nM and the Bmu was 156 ± 28
`fmol/mg of protein. Specific [3H]rauwolscine binding was 90%
`of total binding at the KD concentration. Figure 1B illustrates
`a mean Rosenthal plot derived from individual saturation bind(cid:173)
`ing isotherms.
`The K1 for rauwolscine inhibition of specific [3H]rauwolscine
`binding in endothelial membranes was determined from com(cid:173)
`petition binding studies (fig. 2). Individual competition binding
`curves for rauwolscine fit best to a one-site binding model; this
`was consistent with the results from saturation binding exper(cid:173)
`iments. The mean K, value (0.261 ± 0.07 nM) for rauwolscine
`agreed well with the Ko value for [3H)rauwolscine from satu(cid:173)
`ration binding experiments. KD and K, values for rauwolscine
`were consistent with binding to alpha-2 ADRs. In addition, the
`rank order of potency (rauwolscine > ARC-239 > prazosin) for
`inhibiting specific [8H]rauwolscine binding in endothelial mem-
`
`Slayback Exhibit 1094, Page 3 of 8
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`1993
`
`Endothelial Alpha-2 Adrenoceptors
`
`1129
`
`TABLE 1
`AfflnltlN of alpha-2 ADR antagonists tor Inhibiting specific ['H]rauwolsclne binding (K,) In vascular endothelium and for Inhibiting
`endothelium-dependent vascular relaxation (Ka)
`The values are means ± S.E.; n • number of endothelial cell preparations Of arteries from individual animals. Binding data were analyzed using a nonlinear least-squares
`curve-fitting computer program. Affinity values from binding data that fit best to a two-site binding model are expressed as K., (high-affinity binding site) and Ka. (low(cid:173)
`afflnlty binding site). %High • percent of ~ l t y binding sites; %Low • percent of low-affinity binding sites.
`
`Rauwolsclne
`SKF-104856
`BAM-1303
`ARC-239
`Prazosln
`Splroxatrine
`Oxymetazoline
`• Not determined.
`
`""
`0.261 ± 0.07 (n = 5)
`54.5 ± 10.0 (n = 2)
`0.394 ± 0.14 (n = 2)
`
`""'
`
`""'
`
`26.4 ± 8.4 (n = 6)
`51.4 ± 5.3 (n = 2)
`0.595 ± 0.08 (n = 3)
`0.258 ± 0.11 (n = 5)
`
`72
`86
`n
`26
`
`544 ± 180 (n = 6)
`1620 ± 390 (n = 2)
`17.8 ± 4.0 (n = 3)
`27.0 ± 6.2 (n = 5)
`
`28
`14
`23
`74
`
`""'
`1.71 ± 0.26 (n = 4)
`42.7 ± 4.6 (n = 3)
`NO"
`1290 ± 140 (n = 6)
`1480 ± 170 (n = 4)
`24.6 ± 3.8 (n = 6)
`ND"
`
`
` 
`

`
`
` ÿ 
`
`ÿ
`
` 
`

 
`
`0 LIC-14.304 •endo
`• Epinephrine ••nclo
`
`k
`
`-9
`
`-8
`
`-5
`
`-7
`-6
`Log M [Agonist]
`ÿ ÿ
`Fig. 3. Mean concentration-response curves for a/pha-2 adrenerglc ag(cid:173)
`onists In causing relaxation of pig left circumflex coronary arteries with
`(+endo) and without (-endo) endothelium in the presence of nadolol (10
`µM), prazosin (0.1 µM), desipramine (0.1 µM) and indomethacln (10 µM).
`Relaxation responses are expressed as the percent relaxation to the
`base-line level of the tone present before KCI addition. ECeo values for
`agonists in nanomolar concentration are: epinephrine, 240 ± 62; UK
`14,304, 147 ± 50; p-lodoclonldine, 64.6 ± 22; guanabenz, 170 ± 60;
`clonldine, 144 ± 40; and norepinephrine, 1100 ± 280 (data not shown).
`Each curve represents mean responses of three to five arteries taken
`from Individual animals.
`
`ÿ
`

ÿ

ÿ ÿÿ
`
`rauwolscine (fig. 4A). Rauwolscine-induced shifts in the epi(cid:173)
`nephrine concentration-response curves are plotted according
`to the method of Arunlakshana and Schild (1959, fig. 4B).
`Slopes of individual Schild regressions from four separate ex(cid:173)
`periments were not different from unity; therefore, the slopes
`of regression lines were constrained to one for determining pA2
`values (slope of Schild regreBBions = -1.20 ± 0.03; pA2 value
`for rauwolscine = 1.71 ± 0.26 nM). Table l lists affinity values
`for rauwolscine, ARC-239, prazosin, SKF-104856 and spiroxa(cid:173)
`trine in inhibiting endothelium-dependent, alpha-2 ADR-me(cid:173)
`diated vascular relaxation.
`Correlation of binding with function. Table l lists an(cid:173)
`tagonist affinity values for alpha-2A and alpha-2C ADR binding
`sites (KJS) on vascular endothelium and for inhibiting endothe(cid:173)
`lium-dependent, alpha-2 ADR-mediated vascular relaxation
`(Kss). Ks values were similar to K1 values at the alpha-2A ADR
`subtype. By contrast, Ks values for subtype-selective drugs
`were 30- to 50-fold different compared with K1 values at the
`alpha-2C ADR subtype. Figure 5 illustrates this comparison
`between Ks and K1 values. The Ks versus K1 comparison at the
`alpha-2A ADR subtype correlated well (r = .98). In most cases,
`
`oxymetazoline in inhibiting specific (3H]rauwolscine binding
`to determine whether or not the alpha-2A ADR, alpha-2C ADR
`or both receptor subtypes were present on endothelial mem(cid:173)
`branes. ARC-239 distinguishes the alpha-2A ADR subtype (K1
`= 171 nM) from the alpha-2B and alpha-2C ADR subtypes (KJS
`= 2-13 nM; Bylund et al., 1992). In our experiments, ARC-239
`bound to a heterogeneous population of alpha-2 ADRs on
`endothelial membranes with a KJH value of 26.4 nM and a KIL
`value of 544 nM, consistent with binding to alpha-2A and
`alpha-2C ADR subtypes. We also obtained similar results using
`prazosin (table 1). Like ARC-239, spiroxatrine also distin(cid:173)
`guishes the alpha-2A ADR subtype (K1 = 8-13 nM) from the
`alpha-2B and alpha-2C ADR subtypes (KJS = 0.3-0.6 nM;
`Blaxall et al., 1991; Bylund et al., 1992). In our study, spiroxa(cid:173)
`trine also bound to a heterogeneous population of alpha-2 ADRs
`on endothelial membranes with a K1H value of 0.595 nM and a
`K1L value of 17.8 nM, consistent with binding to alpha-2A and
`alpha-2C ADR subtypes. Oxymetazoline also differentiates the
`alpha-2A ADR subtype (K1 = 0.8-1.5 nM) from the alpha-2B
`and alpha-2C ADR subtypes (KJS = 31-300 nM; Blaxall et al.,
`1991; Bylund et al., 1992). In endothelial membranes, oxyme(cid:173)
`tazoline bound to two pharmacologically distinct alpha-2 ADR
`sites with K1H and K1L values of 0.258 nM and 27.0 nM,
`respectively. Using these antagonists, we found from 14% to
`28% alpha-2A ADRs and from 72% to 86% alpha-2C ADRs
`(table 1). Taken together, these results suggest that pig vascular
`endothelium contains both alpha-2A and alpha-2C ADR sub(cid:173)
`types.
`Functional measurements of antagonist affinity. Mean
`concentration-response curves for alpha-2 ADR agonists in
`causing relaxation of pig left circumflex coronary arteries are
`shown in figure 3. The rank order of potency was p-iodocloni(cid:173)
`dine > clonidine > UK 14,304 > guanabenz > epinephrine >
`norepinephrine, which suggests alpha-2 ADRs mediate relaxa(cid:173)
`tion of pig coronary arteries. Compared with epinephrine, UK
`14,304 and norepinephrine were full agonists, whereas p-iodo(cid:173)
`clonidine, clonidine and guanabenz were partial agonists. In
`arteries without endothelium, epinephrine did not cause relax(cid:173)
`ation, which suggests beta adrenergic receptors were effectively
`blocked in our experiments. These results suggest alpha-2
`ADRs located on the endothelium mediate relaxation of pig
`coronary arteries.
`To determine the affinity of rauwolscine for inhibiting en(cid:173)
`dothelium-dependent, alpha-2 ADR-mediated relaxation in
`coronary arteries, we generated epinephrine concentration-re(cid:173)
`sponse curves in the presence of increasing concentrations of
`
`Slayback Exhibit 1094, Page 4 of 8
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`1130
`
`Bockman et al.
`
`A
`
`Taken together, these data suggest that the alpha-2A ADR
`subtype mediates endothelium-dependent relaxation of pig cor(cid:173)
`onary arteries .
`
`Vol. 267
`
`Discussion
`
`• Contral
`• • 10 nM Aauw
`■ +301'11Ra.
`• + 100 I'll R-
`
`The existence of a heterogeneous population of alpha-2 ADRs
`is now well established. Radioligand binding and molecular
`cloning studies both indicate the presence of four subtypes of
`alpha-2 ADRs (Bylund, 1992; Harrison et al., 1991). The phar(cid:173)
`macological characterization of these receptor subtypes is based
`on radioligand binding studies in tissues and cell lines contain(cid:173)
`ing only one subtype. Using the human colonic adenocarcinoma
`cell line (HT29 cell; alpha-2A), neonatal rat lung (alpha-2B),
`the opossum kidney cell line (OK cell; alpha-2C) (Blaxall et al.,
`1991) and the bovine pineal gland (alpha-2D) (Simonneaux et
`al., 1991), Bylund and coworkers identified drugs with different
`binding affinities for the four alpha-2 ADR subtypes. Alpha-2
`
` 
`ADR subtypes can be distinguished from one another based on
`


`these differential affinity values for subtype-selective alpha-2
`ADR antagonists. Using several subtype-selective alpha-2 ADR
`drugs, we characterized the alpha-2 ADR subtypes present on
`ÿ 
`vascular endothelium. We also studied the functional role of
`alpha-2 ADR subtypes in mediating endothelium-dependent
`ÿ
` 
`vascular relaxation.
`Using drugs that distinguish the alpha-2B and alpha-2D ADR
`
`
`subtypes from the alpha-2A and alpha-2C ADR subtypes, we
`determined that the alpha-2B and alpha-2D ADR subtypes are
`

`not present on pig vascular endothelial membranes. For ex(cid:173)
`  
`ample, in vascular endothelial membranes, K, values for BAM-
`1303 and SKF-104856 were 10- to 50-fold different from their
`ÿ ÿ
`KJS at the alpha-2B ADR subtype (Blaxall et al., 1991; Bylund
`et al., 1992; Gleason and Hieble, 1991). In vascular endothelial
`membranes, K, values for BAM-1303 and rauwolscine were
`different by 135-fold and 13-fold, respectively, from their KJS
`at the alpha-2D ADR subtype (Simonneaux et al., 1991). How(cid:173)
`ever, the K,values for rauwolscine, BAM-1303 and SKF-104856
`in endothelial membranes are consistent with their KJS at the
`alpha-2A and alpha-2C ADR subtypes (Blaxall et al., 1991;
`Bylund et al., 1992; Gleason and Hieble, 1991). The competition
`binding curves for rauwolscine, BAM-1303 and SKF-104856 in
`endothelial membranes fit best to one-site binding models,
`which was consistent with their lack of selectivity between the
`alpha-2A and alpha-2C ADR subtypes.
`We identified a heterogeneous population of alpha-2 ADR
`
`ÿ
`

ÿ

ÿ ÿÿ
`
`-8
`
`-7
`
`-4
`-5
`-6
`Log M [Epinephrine]
`
`-3
`
`B
`
`2
`
`-9
`
`-7
`
`-8
`Log M [Rauwolsclne]
`Fig. 4. A) Illustrates mean concentration-response curves for epineph(cid:173)
`rine-induced relaxation of pig left circumflex coronary arteries with en(cid:173)
`dothelium in the absenoe (Control) and presence (+Rauw) of rauwolsclne.
`Relaxation response curves are plotted as the percent relaxation to the
`base-line level of tone present before KCI addition. B) Shows a mean
`Schild plot of four individual experiments. Rauwolsclne-induced shifts in
`the epinephrine relaxation-response curves are expressed as the log of
`the dose ratio - 1.
`
`the 95% confidence intervals of the plotted points contained
`the line of identity, which illustrates the similarity between Ks
`and K1 values at the alpha-2A ADR subtype. Conversely, the
`regression line correlating antagonist KJS for the alpha-2C ADR
`subtype with Kss was significantly different from the line of
`identity and the comparison did not correlate as well (r = .80).
`
`,.._ -10
`C
`.2
`-:;;
`:
`
`-9
`
`II C
`lo:~
`
`8
`
`-
`
`& ...
`Q. C .. < Q.
`NII "y -7
`
`r= 0.98
`
`, , , ,
`
`,'
`,
`,,'
`
`, , ,
`,
`, ,
`
`r= 0.80
`
`, , , ,
`
`,
`, ,
`,
`, , ,
`, ,
`, ,
`, , ,
`,
`
`, , ,
`, , ,
`, , ,
`, ,
`, , ,
`,
`-10 -5
`
`-6
`
`-7
`-8
`a2c pK,
`(Binding in Endothe~um)
`
`-9
`
`-10
`
`-5
`
`-6
`
`-7
`-8
`a2 ,_ pK1
`(Binding in Endothelium)
`
`-9
`
`Fig. 5. Correlation plots of affinity values for ARC-239,
`prazosln, spiroxatrlne, SKF-104856 and rauwolsclne
`for inhibiting a/pha-2 ADR-mediated, endotheliunHie(cid:173)
`pendent relaxation (pKa, y-axis) and for inhibiting spe(cid:173)
`cific (3H)rauwolsclne binding in endothelial membranes
`(pK,, x-axis). Affinity values are expressed as the mean
`± 95% confidence interval and are taken from table 1.
`The dashed line represents the line of Identity and the
`solid line, the regression line of the plotted points. The
`correlation coefficient (r) is listed.
`
`E
`::,
`~
`~ -6
`"O
`C
`~
`
`Slayback Exhibit 1094, Page 5 of 8
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`1993
`
`Endothelial Alpha-2 Adrenoceptora
`
`1131
`
`suggest both alpha-2A and alpha-2C ADR subtypes are present
`on pig vascular endothelium.
`Results from the present study show that alpha-2 ADRs on
`endothelial cells mediate relaxation of pig coronary arteries,
`which was consistent with the findings of others (Cocks and
`Angus, 1983; Vanhoutte and Miller, 1989). However, the alpha-
`2 ADR subtype that mediates endothelium-dependent vascular
`relaxation is unknown. To study the alpha-2 ADR subtype that
`mediates endothelium-dependent vascular relaxation, we deter(cid:173)
`mined the affmities or Ks of drugs for inhibiting endothelium(cid:173)
`dependent, alpha-2 ADR-mediated relaxation and then com(cid:173)
`pared them to their respective binding affinities (Kre) for the
`alpha-2A and alpha-2C ADR subtypes present on vascular
`endothelial membranes. The Ks versus Kr comparison corre(cid:173)
`lated better (r = .98) for the alpha-2A ADR subtype than for
`the alpha-2C ADR subtype (r = .80). In addition, the absolute
`Ks values obtained from functional studies were similar to the
`absolute Kr values obtained from binding studies at the alpha-
`
` 
`2A ADR subtype and the 95% confidence intervals of most of
`

`the plotted points in this correlation contained the line of
`
`identity. By contrast, the regression line correlating antagonist
`Kr& for the alpha-2C ADR subtype with Ks& was significantly
`different from the line of identity, which illustrated the 30- to
`50-fold difference between Ks values and Kr values for subtype(cid:173)
`ÿ
`selective drugs at the alpha-2C ADR subtype. Consistent with
`classic receptor theory (Ruffolo, 1982) and the results from
`other studies (Abel and Minneman, 1986; May et al., 1985), Ks
`and Kr values will approximate one another if the radiolabeled
`receptor is mediating the