`
`   
` 

 
`
`IPR Page 1/25
`
`Santen/Asahi Glass Exhibit 2007
`Micro Labs v. Santen Pharm. and Asahi Glass
`IPR2017-01434
`
`

`

`COLEMAN ET AL.
`
`.
`.
`.
`a. Selective agonists and antagonists .
`.
`b. Distribution and biological functions .
`2. Ligand-binding studies .................
`3. Second-messenger studies ..............
`4. Molecular biology .....................
`E. TP receptors ............................
`1. Functional studies ....................
`
`.
`.
`.
`a. Selective agonists and antagonists .
`b. Distribution .......................
`
`2. Ligand-binding studies .................
`3. Second-messenger studies ..............
`4. Molecular biology .....................
`III. Conclusions ................................
`IV. References .................................
`
`.......................................... 219
`.......................................... 220
`.......................................... 220
`.......................................... 221
`.......................................... 221
`.......................................... 221
`.......................................... 221
`.......................................... 221
`.......................................... 222
`.......................................... 222
`.......................................... 223
`.......................................... 223
`.......................................... 224
`.......................................... 224
`
`I. Introduction
`
`A. Historical Background
`
`The activity associated with the PGs* was first ob-
`served in 1930 by Kurzrok and Lieb in human seminal
`fluid. This observation was supported and extended by
`both Goldblatt (1933) and von Euler (1934). However, it
`was not for another 20 years that Bergstrom and Sjovall
`(1957) successfully purified the first PGs, PGE; and
`PGFM. During the next decade or so, it became clear
`that the biological activities of the PGs were extremely
`diverse and that the family included members other than
`the original two, these being named alphabetically from
`PGA2 t0 PGHz. Of these, PGAz, PGBz, and PGCz are
`prone to extraction artifacts (Schneider et al., 1966;
`Horton, 1979). PGG2 and PGHz are unstable intermedi-
`ates in the biosynthesis of this family of hormones (Ham-
`berg and Samuelsson, 1973). PGs can be biosynthesized
`from three related fatty acid precursors, 8,11,14-eicosa-
`trienoic acid (dihomo-‘y-linolenic acid), 5,8,11,14-eicosa—
`tetraenoic acid (arachidonic acid), and 5,8,11,14,17-ei-
`cosapentaenoic acid (timodonic acid), giving rise to 1-,
`2- and 3-series PGs, respectively (van Dorp et al., 1964);
`the numerals refer to the number of carbon-carbon dou-
`
`ble bonds present. In most animals, arachidonic acid is
`the most important precursor; therefore, the 2-series PGs
`are by far the most abundant.
`By the middle 1970s it was clear that PGs were capable
`of causing a diverse range of actions, but few efforts were
`made to investigate the receptors at which PGs acted
`Indeed, some doubted that they acted at receptors in the
`“classical” sense at all, behaving, rather, that by virtue
`of their lipid nature they dissolved in cell membranes
`and caused their biological actions by altering the phys-
`ical state of those membranes. However, despite this,
`interest was increasing in this new class of hormones,
`‘ Abbreviations: PG, prostaglandin; TX, thromboxane; PGI., pros-
`tacyclin; G., stimulatory G-protein; Gk inhibitory G—protein; 6., per-
`tussis toxin-insensitive G-protein; RCCT, rabbit cortical collecting
`tubule.
`
`and there was optimism about their potential as new
`drugs. This interest peaked with the discovery of the two
`unstable PG-like compounds, TXAz (Hamberg et al.,
`1975) and PGIz (Moncada et al., 1976). The collective
`term for this family of hormones is “the prostanoids.” At
`that time, the main problem with prostanoids as drugs
`was perceived to be one of stability, both chemical and
`metabolic, and there was an enormous amount of chem-
`ical effort directed toward developing more stable pros-
`tanoids. Despite successes in this regard, another prob-
`lem soon became apparent, and that was one of side
`effect liability. Indeed, the very range of the actions of
`this class of compounds, which on the one hand offered
`such opportunities for drug development, began con-
`versely to appear to be their limitation, because it ap-
`peared not to be possible to produce prostanoids as drugs
`without use-limiting side effects. It was this challenge
`that prompted a small number of groups of scientists to
`attempt to rationalise the “bewildering array” of actions
`of prostanoids by means of the identification and clas-
`sification of prostanoid receptors. Initially, in the 19703,
`most of the work directed toward the study of prostanoid
`receptors was designed to characterise specific binding
`sites for the radiolabeled natural ligands (Kuehl and
`Humes, 1972; Rao, 1973; Powell et al., 1974). Although
`this served to support the existence of specific mem-
`brane-binding sites, these sites may or may not have
`represented functional receptors.
`
`B. Studies of Receptor Identification and Chzssification
`1. Functional studies. The use of functional data to
`
`classify hormone receptors was pioneered by Ahlquist in
`1948, in an attempt to classify the receptors responsible
`for the biological actions of the catecholamines, adrena~
`line and noradrenaline. Despite the limited tools at his
`disposal, the outcome of these studies was the classifi-
`cation of adrenoceptors into a and fl subtypes, a classi-
`fication scheme that has stood to the present day. This
`work was subsequently extended by Lands and colleagues
`
`IPR Page 2/25
`
`IPR Page 2/25
`
`

`

`CLASSIFICATION OF PROSTANOID RECEPTORS
`
`207
`
`in 1967 who, using the same approach, demonstrated
`that, although the classification proposed by Ahlquist
`was essentially correct, it was an oversimplification and
`one of Ahlquist’s receptors, the B-adrenoceptor, could be
`further divided into two subtypes, termed fl, and 52. This
`approach to receptor classification, although now largely
`taken for granted, was revolutionary.
`The relatively large number of naturally occurring
`prostanoids, their high potencies, and the variety of the
`responses elicited by them in different cells throughout
`the mammalian body made this an ideal area in which
`to study receptor subtypes. This was first recognised by
`Pickles in 1967, when he demonstrated that a range of
`different prostanoids, both natural and synthetic, showed
`different patterns of activity on a variety of isolated
`smooth muscle preparations. Yet, Pickles did not extend
`this work, and during the next 15 years little further
`work was reported extending his original observations.
`The few studies that were published (Andersen and
`Ramwell, 1974; Andersen et al., 1980; Gardiner and Col-
`lier, 1980) demonstrated that not only were different
`rank orders of agonist activity observed with a relatively
`small range of both natural and synthetic prostanoids,
`over a wide range of isolated preparations, but certain
`consistent patterns emerged. However, this work was not
`developed to describe a comprehensive receptor classifi-
`cation. In 1982, Kennedy and his coworkers described a
`comprehensive, working classification of prostanoid
`receptors based on functional data with the natural ag-
`onists, some synthetic agonists, and a small number of
`antagonists (Kennedy et al., 1982; Coleman et al., 1984).
`Their classification of receptors into DP, EP, FP, IP,
`and TP recognised the fact that receptors exist that are
`specific for each of the five naturally occurring prosta-
`noids, PGs D2, E2, F2“, 12, and mi», respectively. It was
`clear that at each of these receptors one of the natural
`prostanoids was at least one order of magnitude more
`potent than any of the other four. Although in hindsight
`this observation may not seem remarkable, there are to
`this day no other examples of a family of hormones that
`demonstate such receptor selectivity; it is certainly not
`true of catecholamines, tachykinins, or leukotrienes. Al-
`though this broad classification into five classes of pros-
`tanoid receptors remains intact, evidence arose for a
`subdivision within the EP receptor family. There is now
`evidence for the existence of four subtypes of EP recep-
`tors, termed arbitrarily EPl, EPz, EP3, and EP4. The
`recent cloning and expression of receptors for the pros-
`tanoids has not only confirmed the existence of at least
`four of the five classes of prostanoid receptor, EP, FP,
`IP, and TP, but has also supported the subdivision of
`EP receptors into at least three subtypes, corresponding
`to EP,, ER,» (or EP,), and EPa. The current classification
`and nomenclature of prostanoid receptors is summarised
`in table 1.
`
`2. Radioligand-binding studies. During the 1970s, there
`
`were a large number of ligand-binding studies performed
`in a wide range of tissues using radiolabeled PGs (Rob-
`ertson, 1986). These studies made it clear that there are
`specific prostanoid-binding sites in the plasma mem-
`branes of such diverse tissues as liver, smooth muscle,
`fat cells, corpus luteum, leukocytes, platelets, and brain.
`In many of these, the ligand affinity (K4) is of the order
`of 1 to 10 nM, and the receptor density is in the range of
`1 pmol/mg protein. Furthermore, in many of the tissues
`exhibiting high affinity, and high density prostanoid—
`binding sites, it was known that prostanoids had biolog-
`ical activity, thus providing circumstantial support for
`these binding sites being functional receptors. Nonethe-
`less, these studies did not further our understanding of
`prostanoid receptor classification, because in most cases,
`radioligands were confined to [3H]PGs E1, E2, or F2." and
`either no competition studies were performed or compe-
`tition studies were undertaken with prostanoids that do
`not discriminate among receptor subtypes (Coleman et
`al., 1990). It was not until the 1980s that studies were
`performed using [3H]PGs D2 and 12, and the evidence for
`a wider range of different types of prostanoid ligand-
`binding site emerged.
`That some of these binding sites truly represented
`functional receptors was supported by the demonstration
`that they were capable of autoregulation, whereby bind-
`ing site numbers are modulated by exposure to ligand.
`Thus, exposure of the animal or tissue to high levels of
`unlabeled ligand resulted in a “down-regulation” or loss
`of binding sites (Robertson et al., 1980; Robertson and
`Little, 1983), and conversely, treatment with inhibitors
`of endogenous prostanoid synthesis led to a correspond-
`ing “up-regulation” of binding sites (Rice et al., 1981).
`In some of these studies, attempts were made to associate
`modulation of binding sites with alterations in function;
`for example, Richelsen and Beck-Nielsen (1984) dem-
`onstrated that down-regulation of PGEz-binding sites
`was accompanied by a reduction in PGEz-induced inhi-
`bition of lipolysis. However, it was not until more selec-
`tive, synthetic prostanoid agonists and antagonists be-
`came available, and distinct rank orders of agonist activ-
`ity in functional studies became apparent, that the
`association between binding sites and functional recep-
`tors became possible (see section I.B.3).
`3. Second-mssenger studies. Almost all of the studies
`of prostanoids and second messengers until the late 1980s
`were concerned with cych nucleotides, particularly
`cAMP. Butcher and colleagues were the first to demon-
`strate an association between PGs and CAMP (Butcher
`et al., 1967; Butcher and Baird, 1968), and although their
`observation made little initial impact, it became increas-
`ingly accepted that E-series PGs at least were capable of
`stimulating adenylyl cyclase to cause increases in intra-
`cellular cAMP (Kuehl et al., 1972, 1973). However, it
`became clear that prostanoid effects on adenylyl cyclase
`were not solely excitatory, and in 1972, a more complex
`
`IPR Page 3/25
`
`IPR Page 3/25
`
`

`

`208
`
`COLEMAN ET AL.
`
`TABLE 1
`Classification and nomenclature of prostanoid receptors, with selective agonists and antagonists, and system of response tmnsduction‘
`
`Receptor/mbtype
`DP
`
`Selective agonists
`BW 245C, ZK110841,
`RS 93520
`
`EP
`
`EP,
`
`EP,
`
`EP.
`
`EP.
`FP
`
`H’
`
`TP
`
`Iloprost,1: 17-phenyl PGEg,
`sulprostonel
`Butaprost, AH13205, miso-
`promll
`Enprostil, GR63799, sulpro-
`stone,‘ misoprostol,”
`M&B 28767fi
`None
`Fluproetenol, cloproetenol,
`prostalene
`Cicaprost, iloprost, octimi-
`bate
`
`1144069, U46619, SQ 26655,
`EP 011, LED?
`
`Selective antagonists
`BW A868C, AH68091
`
`AH6809,"§ SC-19220
`
`None
`
`None
`
`AH22921,:[:1: AH2384811:
`None
`
`None
`
`AH23848,§§ GR32191,
`EP 092, SQ 29548,
`ICI 192605, L-655240,
`BAY u 3405, 8-145, BM
`13505
`
`W”
`tcAMP via G.
`
`TIntracellular
`Ca"
`tcAMP via G.
`
`chMP via G.
`TPI turnover via
`G.
`TcAMP via G,?
`1‘PI turnover via
`G,I
`TcAMP via G.
`
`TPI turnover via
`G.I
`
`‘ For more detailed information, see relevant section in text.
`T Also EP. receptor-blocking drug.
`1: Also IP receptor agonist
`§ Also DP receptor-blocking drug.
`I Also EP. receptor agonist.
`1 Also EP, receptor agonist.
`“ Also EP. receptor agonist.
`it Also TP receptor agonist.
`it Also TP receptor-blocking drug.
`5! Also EP. receptor-blocking drug.
`
`relationship between PGEs and adenylyl cyclase was
`reported in platelets (Shio and Ramwell, 1972). PGEI
`caused an elevation of platelet cAMP, but PGE2 caused
`a reduction. Interestingly, this distinction was reflected
`in the effects of PGE1 and PGE; on platelet aggregation.
`PGEI inhibited aggregation, whereas PGE2 potentiated
`the effect of aggregatory agents such as adenosine di-
`phosphate (Shio and Ramwell, 1971). This parallel be-
`tween cAMP and function not only provided evidence
`that the effect on cyclic nucleotide levels had functional
`relevance but also suggested that these might be receptor
`subtypes. A further distinction was observed at this time
`between the effects of E— and F-series PGs. Whereas
`PGE; and PGE; were seen to exert marked effects on
`levels of cAMP, both stimulatory and inhibitory, PGF2.,
`despite its marked functional activity in many different
`cell types, was virtually devoid of effect on CAMP (Kuehl
`and Humes, 1972; Smith et al., 1992). In fact, it became
`accepted that the actions of PCB, were mediated
`through elevation of cych guanosine 3’,5’-monophos-
`phate (Dunham et al., 1974; Kadowitz et al., 1975),
`although this idea has now lost support.
`As with studies of radioligand binding, studies of sec-
`ond-messenger systems in the 1970s were limited, there
`being few studies in which ranges of receptor-selective
`
`agonists were compared for both function and modula-
`tion of cyclic nucleotide levels. Where comparisons were
`reported, as with the binding studies, they involved com-
`parisons of the then available PGs, E, F, A, and B (Kuehl
`and Humes, 1972), and these give little insight into the
`receptor subtypes involved (Coleman et al., 1990). It was
`not until the 19803, when more selective agonists became
`available, that studies of intracellular levels of cAMP
`provided real evidence for the existence of prostanoid
`receptor subtypes (see section I.B.4).
`4. Molecqu biology. Development of highly potent TP
`receptor antagonists and introduction of their high-affin-
`ity radiolabeled derivatives in binding experiments in the
`1980s enabled solubilization and purification of the TP
`receptor. Using one of these compounds, S-145 (table 2),
`and its 3H-labeled derivative, Ushikubi et al.
`(1989)
`purified the human TP receptor from human platelets to
`apparent homogeneity, and based on the partial amino
`acid sequence of the purified protein, its cDNA was
`isolated in 1991 (Hirata et al., 1991). Subsequently, the
`cDNAs for numerous types and subtypes of prostanoid
`receptors have been cloned by homology screening, and
`the structures of the receptors that they encode have
`been elucidated. These receptors include the mouse TP
`receptor, the human EPl receptor, the mouse EPI, EP2,
`
`IPR Page 4/25
`
`IPR Page 4/25
`
`

`

`CLASSIFICATION OF PROSTANOID RECEPTORS
`
`209
`
`TABLE 2
`Glossary of the chemical names ofprostanoid agonists and antagonists quoted as code numbers in this review
`
` Code Chemical name
`
`
`AH6809
`AH13205
`
`Al-l19437
`
`AH22921
`
`AH23848
`
`AY23626
`BAY In 3405
`
`BW 245C
`BW A8680
`
`EP 011
`EP 045
`
`EP 092
`
`EP171
`
`FCE 22176
`GR32191
`
`GR63799
`
`I-BOP
`
`ICI 192605
`K 10136
`L-655240
`
`M&B 28767
`N-0164
`
`ONO-3708
`
`RS 93520
`
`8-145
`
`SC-19220
`
`SQ 26655
`
`SQ 29548
`
`STA.
`U44069
`U46619
`ZK110841
`
`6~isopropoxy-9-oxoxanthene-2-caboxylic acid
`trans-2-[4-(l-hydroxyhexyl)phenyl]-5-oxocyclo pentaneheptanoic
`acid
`1a(Z),2B,5a(:t)-methyl,7-2-(4-morpholinyl)-3-oxo-
`5(phenylmethoxy)cyclopentyl-5-heptenoate
`[1a(Z),26,5a]-(:I:)-7-[5-[[(1,1'-biphenyl)-4-yl]methoxy]-2-(4-mor-
`pholinyl)-3-oxocyclopentyl]-5-heptcnoic acid
`[1u(Z),2fl,5a]-(:t)-7-[5-[[(1,1’-biphenyl)-4-yl]methoxy]-2-(4-mor-
`pholinyl)-3-oxocyclopentyl]-4-heptenoic acid
`ll—deoxy-prostaglandin E.
`3(R)-3-(4-fluorophenylsulphonamido)-l,2,3,4-tetrahydro-9-carba-
`zole propanoic acid
`5—(6wboxyhexyl)-1-(3~cyclohexyl-3-hydroxypropyl) hydantoin
`3-benzyl-5-(6—carboxyhexyl)-l-(2-cyclohexyl-2-hydroxyethylam-
`ino)-hydantoin
`17,18,19,20-tetranor-16-p-fluorophenoxy-9,11-etheno PGH,
`(1)-5—endo-(6’carboxyhex-2'Z-enyl)-6-em[N-(phenylcarba-
`moyl)hydrazono-methyl]-bicyclo[2.2.1] heptane
`9a,lla-ethano-m~heptanor-13-methyl-13-phenyl-thio-carbamoyl-
`hydrazine-prosta-5Z—enoic acid
`10a-homo-158-bydroxy-9a,11a-epoxy-16p-fluomphenoxy-w-tetra-
`nor-5Z,13E-dienoic acid
`(+)-13,14-didehydro-20—methyl-carboprostacyclin
`[1R-[1a(Z)2B,3a,5a]]'(+)-7-[5-[[(1,1'-biphenyl)-4-yl]methoxy]-3-
`hydroxy-2-(1-piperidinyl)cyclopentyl]—4-heptenoic acid
`[lR-[1a(Z),26(R‘),3a]-(—)-4-benwylamino)phenyl-7-[3-hydroxy-
`3-phenoxypropoxy)-5-oxocyclopentyl]-4—heptenoate
`[lS-(la213(52),3a(1E,38‘),4-a)]-7-[3-(3-hydroxy-4-(4’-iodophen-
`cry)-1-butenyl)—7ooxabicyclo-[2.2.1lheptan-Z-yll-S-heptenoic
`acid
`4(Z)-6-(2-o—chlorophenyl-1,3-dioxan-cis-5-yl) hexenoic acid
`13,14-didehydro-20-methyl PGF,‘
`3-[1-(4-chlorobenzyl)-5-fluoro—3-methyl-indol-2-yl]2,2—dimethyl-
`propanoic acid)
`15S-hydroxy-9-oxo-16.phenoxy-w-tetranorprost-13E-enoic acid
`Sodium p-benzyl-4- [1 -oxo- 2- (4-chlorobcnzyl) -3 -phenylpropyl]
`phenyl phosphonate
`(9,11)-(11,12)-dideoxa—9a,l1a-dimethylmethano—11,l2-methano-
`13,14-dihydro-l3-aza-14-oxo-15-cyclopentyl-16,17,18,19,20-pen-
`tanor-l5-epi-TXA2
`Z4-I(C3'S.1R,2R,3S,6R)-2C3’-cyclohexyl-3’-hydroxyprop-l-ynyl)-
`3-hydroxybicyclo[4.2.0]oct-7-ylidenel butyric acid
`(18,2R,3R,4R)-(5Z)-7-(3-[phenylaulphonyl-amino—bicyclo[2.2.1]
`hept-2-yl)hept-5—anoic acid
`1-acetyl-2-(8-chloro-10,11—dihydodibenz[b,f][1,4loxazepine-10-car-
`bony!) hydrazine
`(IS-(1a,2b(5Z),3a(1E,38‘),4a))-7—(3-(3-hydroxy-l-octeny1)-7-oxabi-
`cyclo(2.2.1)hept-2-yl)-5-heptenoic acid
`[15-[1a,23(5Z),3B,4a]-7-[3-[2-(phenylamino)-carbonyl]hydrazino]
`methyl]-7-oxobicyclo-[2,2,l] -hept-2-yl]-5-heptcnoic acid
`9a,11a-thia-11a-carba-prosta-5Z,13E-dienoic acid
`9a,11a-epoxymethano— 15S-hydroxy-prosta—52,13E-dienoic acid
`11a,9avepoxymethano-15S-hydroxy-prosta-5Z,13E-dienoic acid
`9Mxy-96-chloro-16,17,18,19,20-pentanor—15-cyclohexyl-PGFg.
`
`and EPa receptors, the rat EP; receptor, the mouse and
`bovine FP receptors, and the mouse IP receptor. The
`deduced amino acid sequences of the recombinant mouse
`receptors are shown in figure 1. Hydrophobicity and
`homology analysis of these sequences has revealed that
`all of them have seven hydrophobic segments character-
`istic of transmembrane domains, indicating that they are
`
`G-protein-coupled, rhodopsin-type meptors. The over-
`all homology among the receptors is not high, and the
`amino acid identity is scattered over the entire sequences,
`showing that they are derived from different genes. In-
`deed, Taketo et al. (1994) have identified the genetic loci
`of mouse EPz, EPa, and TP receptors on chromosomes
`15, 3, and 10, respectively. On the other hand, as shown
`
`IPR Page 5/25
`
`IPR Page 5/25
`
`

`

`210
`
`FP
`TP
`:91
`393
`EPZ
`IP
`
`E?
`T?
`an
`393
`£92
`IP
`
`COLEMAN ET AL.
`
`ll
`I
`72
`MSMNSSKQPVSPAAGLIAN'rTcorl-mmvrrsxImveILsxsuIAInmamRr-noxsxausmso-
`68
`MWPNGTSLGACFRPVNIT--LQBRRAIASPIFAASPCAIBLGSNLIALSVIAGA-----mmpassnmce-
`usrcemtsmnm'rcmPRLPn'rsvaPrGDNG'rsPALPIrsmmnvsuvmmovacmmmurmrvas— 81
`msmnsassmams-srrooccsvsvarsrrmvwammsasr---Ran£sxam<sn-—LCIG 7o
`MAEVGGTIPRSNRELQRCVLLTTTIMSIPGVNASFSSTPERLNSPVTIPAVME‘IE‘GVVGNLVAIWL------cxsmoxsrrrrnvco-
`85
`M masocnpcppsvrpcsnsmesawomsecwsrrrvoosvesarsrwrvaevvcucmen.------GARRRSH—PSAFAVLVTG-
`85
`
`
`
`IV—
`__ III _____—
`~Lv1'rorrcHLINGG Isvrwasomrarnossrncsrrcrsuvrscmptmsmrmmv'mn E‘HSTKITSKH-VKMILSGVCMF
`—LVLmrLGLLerAIVASQHAALLmrDPscmr FMGVAMVPFGLCPWIRFVGITRPFSRPTATSR—R-AWATVGLVWVA
`-1.mmmosv1PGALVLaLr'rAGaA-PAGGA-u-CHmocmvmmsmscmvnacvevrorLIsmvsvaa—mmm
`mmnvoou‘rsrwnmsoasweomrssamrrch-rvmssuvasmvzaamrmmrnsrmKr-R-Nrrmcms
`-I.AV'I'DLLGTI.LVSPVTIATYMKG-QWPGDQA---LCDYSTFILLFEBLSGLSIICAMSIERYLAINHAYFYSHYVDK-RLAGLTLE‘AIYAS
`-mvmnmrcnsravwsr-—G—LAHGGm---z.cn'rrormmmsrnnrmvaacmssrrmnomsP-Pcnarursrrnr
`
`162
`157
`166
`160
`171
`176
`
`235
`220
`250
`235
`255
`246
`
`285
`284
`336
`300
`339
`302
`
`366
`341
`403
`365
`513
`416
`
`v _———_——
`_
`AVWAVLPImHRDYQIoASRmnNTBHIE —————— DWE-DRE'YLLE’FSFLGLLALGVS rscmvrevrmav-Krasq-----------
`FP
`scammmwmsvormswcnm---------QRGDWFGLImLmSASVGLSLLm’I‘VSV-ATL-------------------
`TP
`BPl mvnmvsvmsmrpomnsmpac------GHR-QALLAGLIAGMLAALLAALVCNTLSGLALLRA-awasassssrmncp
`BP3
`WWVGRYSVQIWNCHSTGPAGNETDPARBPGSVAFASWDGLLALVVT racsLA'rImvsa-----------------
`EPZ mmcteaseaormmcnon'rnm ------armsrmmrssFLILATVLarVLVCGALLRMHRQmRRTSLGTaQHHA
`IP
`CCLE‘CSLPLDSLGEHQQYCPGSWCFI-RMR--SAP------GGCAFSLAYASLMALLVTSIFE‘CNGSVTLSLYPMYRQQRRHHGSFVP----
`
`VI
`------------------------------------QHRQGRSHHLmr rmncvscvcwsrrtv——mmmmGN-NSPVT-—c-ET
`----------CRVYHT ———————————————— R--EATQRPRDCBVMVQLVGIMVVATVCWMPLLVFIMQTLLQTPPVHSE'SGQLLRATE
`DDRRRIGSRGPRLASASSASSITSATATLRSSRGGGSARRVHAHDVEMVGQLVGIMVVSCICWSPLLV--LVVLAIGGWN-SNSLQ--R-PL
`----------camavsos ---------------- SAQme-rrzraromcmsvcwsrurmmrmousveocx’roucxsxa
`MmVASVACRGHAGASPALQRLSDF--RRRR—--SFRRIAGAEIQMVILLIATSLVVLICSIPLWRVE‘INQLYQPNVVK-——DISRNP-
`-------------------------------------TSRAREDEVYHLILLAIMVIMAVCSLPIMIRGFTQAIA-PDSRE---MG-----
`
`VII__
`---nmmrmrmmwrmmmmnsacccvmISLHstLssrmsmvamsaspmxasooassamr.
`-HounmvarmxmmrnmsvmmPRESSQLQAVSLRRPPAQAMLSGP
`--—rL-nvamsmrwrwvrrummmamnavsmcprsmmxsawassSLRSSRHSGFSHL
`CNS E'anmsnnrmmltmrcoIaDHrNYAsss'rSLPCPGSSALMWSDQLER
`---Dmrarasvsrrmmumxrvnsxamx1 KCLE'CRIGGSGRDSSAQHCSESRRTSSAMSGHSRSFLARELKEISSTSQTL ( 90)
`---DLLAFRE‘NAENPIIDPWVE‘ILE‘RKAVE‘QRLKE‘WLCCLCARSVHGDLQAPLSRPAS (60)
`
`E‘P
`TP
`EPl
`5P3
`592
`IP
`
`FP
`T?
`391
`593
`£92
`IP
`
`FIG. 1. Amino acid sequence alignment of the mouse prostanoid receptors. The amino acid wquences of the mouse PGF receptor (FP), TXA,
`receptor (TP), EP; receptor (ER), EP. receptor (Eh), EP, receptor (EPz), and PGI: receptor (IP) are aligned to obtain the optimum homology.
`The approximate positions of the putative transmembrane regions are indicated by horizontal lines above the sequences. Amino acids conserved
`are shown by bold letters.
`
`in the figure, these receptors show strong conservation
`of sequence in several regions, indicating that they prob-
`ably evolved from a common ancestor. The most con-
`served region is the seventh transmembrane domain,
`where the consensus sequence of L-X-A/Y-X-R-X-A-
`S/T-X-N-Q(P)—I—L-D-P-W-V(I)-Y(F)-I(L)-L-L/F-R is
`shared. Another region of significant homology is the
`second extracellular loop between the fourth and fifth
`transmembrane regions. Because these sequences are
`shared by the prostanoid receptors from various species,
`but not by other rhodopsin-type receptors, Narumiya et
`al. (1993) suggested that they are related to structural
`requirements of prostanoid recognition. Particular atten-
`tion has been paid to the conserved arginine in the
`seventh transmembrane domain, which is located at a
`position analogous to lysine“ of the bovine rhodopsin
`molecule that makes a Shiff base with its ligand, a11-
`trans-retinal. In the rhodopsin-type receptor,
`ligand
`binding and recognition are suggested to occur in the
`outer half of the seven transmembrane segments (Sa-
`
`varese and Fraser, 1992). Because the carboxyl group is
`essential for biological activity of most prostanoids, it
`was proposed that the arginine serves as the binding site
`for the a-carboxyl group of prostanoid molecules. The
`other transmembrane domains of the prostanoid recep-
`tors are more hydrophobic than those of the monoamine
`receptors, which may facilitate binding to the cyclopen-
`tane ring and aliphatic side chains of prostanoid mole-
`cules.
`
`All of the recombinant receptors have been expressed
`in cultured cells and their ligand-binding properties and
`signal transduction pathways studied. The results ob-
`tained with each receptor type are described in detail in
`subsequent sections. These studies may give us more
`accurate information than those obtained by pharmaco-
`logical and biochemical studies in native tissues, because
`the expressed receptor system permits the study of ho-
`mogeneous populations of receptors without the compli-
`cation of the presence of other receptor types. Of course,
`there are also limitations to this approach. The fact that
`
`IPR Page 6/25
`
`IPR Page 6/25
`
`

`

`CLASSIFICATION OF PROSTANOID RECEPTORS
`
`211
`
`only the receptor cDNAs of a limited number of species
`have been cloned means that we do not know whether
`
`discrepancies between the properties of the recombinant
`receptors and the pharmacological analyses are attrib-
`utable to species differences or to receptor subtypes (see
`discussion by Hall et al., 1993). Another limitation is
`that the recombinant receptors have been expressed and
`analysed only in Chinese hamster ovary cells or simian
`kidney COS cells, and this results in coupling of a recep-
`tor from one species with a G-protein from a different
`species; moreover, the pool of G proteins in CHO cells
`and COS cells may differ from that of the tissue in which
`a receptor is normally expressed. With prostanoid recep-
`tors, effects on the ligand-binding properties may be
`minimal, because, unlike the monoamine receptors,
`guanosine triphosphate analogues appear to exert little
`effect on prostanoid binding. Nonetheless, the ligand-
`binding profiles of the recombinant receptors (detailed
`in section II.A.2) are generally in good agreement with
`those characterized earlier in pharmacological and bio-
`chemical experiments.
`
`II. Types, Subtypes. and Isoforms of Prostanoid
`Receptors
`
`A. DP Receptors
`1. Functional studies. a. SELECTIVE AGONISTS AND
`
`ANTAGONISTS. Although PGDZ and various close ana-
`logues do behave as DP agonists, none is particularly
`selective (Giles and Leff, 1988). Indeed, PGD2 itself
`possesses relatively potent FF and even TP receptor
`agonist activity. However, there are a number of potent
`and highly selective DP receptor agonists. One of these
`is 9-deoxy-A9-PGD2 (PGJz, Bundy et al., 1983), but the
`first to be identified, and the most widely used was BW
`2450, a hydantoin prostanoid analogue (Caldwell et al.,
`1979). This compound is interesting in that it is at least
`one order of magnitude more potent than the natural
`ligand, PGDZ, as a DP receptor agonist but, on the other
`hand, appears to be several orders of magnitude less
`potent at other prostanoid receptors and lacks the rela-
`tively high FF and TP receptor agonist activity associ-
`ated with PGD2 itself. Since the discovery of BW 2450,
`another selective DP receptor agonist, ZK110841, was
`reported by Thierauch et al. (1988), although there is
`much less
`information concerning this compound.
`ZK110841 is interesting in that it is not an analogue of
`PCB; but of PGE2. One additional compound of struc-
`tural significance is RS 93520, which is superficially a
`P012 analogue, and has some weak IP agonist activity
`but is much more potent as a DP receptor agonist (Al-
`varez et al., 1991). Quantitative data for some selective
`DP receptor agonists and antagonists are summarised in
`table 3.
`
`The study of DP receptors has been facilitated by the
`availability of antagonists. The first compound shown to
`possess DP receptor-blocking activity was the phloretin
`
`derivative, N-0164, which weakly, but selectively, antag-
`onised inhibitory activity of PGDz on human platelets
`(MacIntyre and Gordon, 1977). Subsequently, an EP1
`receptor-blocking drug, AH6809, was shown to exhibit
`DP receptor-blocking activity but was, again, rather
`weak, with a pAg of about 6.0 (Keery and Lumley, 1988).
`The most significant development was that of BW A8680
`(Giles et al., 1989), an analogue of the agonist, BW 2450.
`In fact, the Wellcome group synthesised a wide range of
`analogues of a related series of high-efficacy agonists to
`antagonists.
`b. DISTRIBUTION AND BIOLOGICAL FUNCTION. DP
`
`receptors are perhaps the least ubiquitous of the prosta-
`noid receptors. Only in American Type Culture Collec-
`tion CCL 44 cells, a cell line derived from bovine embry-
`onic trachea (Ito et al., 1990), have DP receptors been
`shown to exist as a homogeneous receptor population; in
`all other tissues in which they have been identified, they
`exist only in association with one or more other prosta-
`noid receptor types. Therefore, it is difficult to study
`them in isolation. Fortunately, the available potent and
`selective DP receptor agonists and antagonists have
`proved valuable in the study of this receptor type.
`DP receptors are distributed largely in blood platelets,
`vascular smooth muscle, and nervous tissue, including
`the central nervous system. There are also examples of
`DP receptors in gastrointestinal, uterine, and airway
`smooth muscle in some animal species (Coleman et al.,
`1990). Responses mediated by DP receptors are predom-
`inantly inhibitory in nature, e.g., inhibition of platelet
`aggregation and relaxation of smooth muscle and possi-
`bly inhibition of autonomic neurotransmitter release.
`However, DP receptors are associated with excitatory
`events in some afferent sensory nerves, where they can
`induce pain or, probably more correctly, hyperalgesia
`(Ferreira, 1983; Horiguchi et al., 1986). The distribution
`of DP receptors is highly species specific, e.g., human
`platelets appear to have a particularly rich population of
`inhibitory DP receptors (MacIntyre and Armstrong,
`1987), whereas the platelets of most laboratory species
`appear to contain few if any DP receptors, and as far as
`uterine smooth muscle is concerned, DP receptors appear
`to be confined to the human (Sanger et al., 1982).
`2. Ligand-binding studies. Few ligand-binding studies
`have been reported for DP receptors. PGDg-specific bind-
`ing sites have been identified in human platelet mem-
`branes, at which PGs of the E, F, and I series have
`substantially lower binding affinities but at which the
`DP receptor agonist, BW 2450, has high affinity (Cooper
`and Ahern, 1979; Town et al., 1983). Binding sites for
`[3H]PGD2 have also been identified in rat brain synaptic
`membranes (Shimizu et al., 1982).
`3. Second-messenger studies. The evidence relating to
`DP receptors and second-messenger coupling is largely
`indirect. Simon et a1. (1980) demonstrated that PGs D2,
`E2, and I; are approximately equipotent in stimulating
`
`IPR Page 7/25
`
`IPR Page 7/25
`
`

`

`212
`
`COLEMAN ET AL.
`
`TABLE 3
`Potencies of some DP receptor agonists and antagonists‘
`
`Agonists
`
`mggf£gfi?m-
`0.03—0.7
`
`BW 2450
`
`ZK110841
`
`0.2-1.0
`
`RS93520
`
`1.0
`
`References
`Town et al., 1983;
`Narumiya and
`Toda, 1985
`Thierauch et al.,
`1988; Ito et al.,
`1990
`Alvarez et al.,
`1991
`
`Antagouists
`
`BW A86850
`
`pA.
`9.3
`
`References
`Giles et al., 1989
`
`AH6809
`
`6.0-6.6
`
`Keery and Lumley, 1988;
`Ito et al., 1990
`
`‘ Data obtained on human platelets, rabbit transverse stomach strip, rat peritoneal mast cells, and bovine embryonic trachea cells.
`
`adenylate cyclase activity in human colonic mucosa.
`Because PGDz is only a very weak agonist at EP and IP
`receptors, this argues that, among others, DP receptors
`must be present in this preparation, coupling positively
`to adenylate cyclase, presumably via G.. However, the
`demonstration by Ito et a1. (1990) that activation of DP
`receptors in American Type Culture Collection CCL 44
`cells (see above) results in an increase in levels of intra-
`cellular cAMP supports this association. Furthermore,
`several selective agonists exist for the DP receptor (Giles
`et al., 1989), and both of these and PGDz itself have been
`shown to bind to a specific DP receptor in platelets to
`cause an increase in cAMP formation (Gorman et al.,
`1977b; Schafer et al., 1979; Siegl et al., 1979a; Whittle et
`al., 1978), again suggesting that DP receptors can couple
`to G. to stimulate adenylate cyclase (Halushka et al.,
`1989).
`
`B. EP Receptors
`1. Functional studies. a. SELECTIVE AGONISTS AND
`
`ANTAGONISTS. i. EPl receptors. Although sulprostone
`was first identified as a potent EP; receptor agonist, it
`is more potent at EP3 receptors (Bunce et al., 1990). In
`fact, to date, there is no reported example of a highly
`selective EP, receptor agonist. Two compounds that