INTRODUCTION

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

PROSTANOIDS are important autacoids that exert diverse physiological and pathophysiological effects in various systems. These involve modulation of neuronal activ   ity, including neurotransmitter release, pyrexia and sleep induction, alterations in platelet aggregation, and regulation of ion and water transport in kidneys, as well as of gastrointestinal motility and secretion. The best-characterized actions of prostanoids, however, are relaxation and contraction of smooth muscle. Of specific interest is the role of prostaglandins in the developing subject, wherein they play a major role in various conditions. Notably, prostaglandins are implicated in implantation of the fertilized egg (211), parturition (162, 200), breathing (124), control of brain and ocular circulation (38, 85, 122), and the regulation of ductus arteriosus tone (191). Of relevance, increased neonatal levels of prostanoids (103, 129, 134) have been suggested to contribute in the genesis of intraventricular cerebral hemorrhage (133, 220), patent ductus arteriosus (191), and possibly retinopathy of prematurity (73, 146).

There are five physiologically major prostanoids that are well characterized, prostaglandin (PG) E₂, PGF₂, PGD₂, PGI₂, and thromboxane (Tx) A₂; other prostanoid-like compounds, namely isoprostanes, also exert biological effects through receptor sites that are not yet clearly identified (179-181), and these will not be covered in this review. PGE₂, PGF₂, PGD₂, PGI₂, and TxA₂ produce their effects by acting on distinct G protein-coupled receptors (GPCRs). In the last two decades the pharmacological profile of specific agents and the cloning of GPCRs to prostanoids, along with the disruption of the genes encoding them, have provided great insight on the biology of prostanoid receptors and functions. A number of comprehensive reviews on the subject have been published (30, 53, 76, 147, 173). Lately, several studies have reported on the ontogenic profile and developmental regulation of prostanoid receptors. This review focuses on the implications of prostanoid receptors on the developing subject with particular attention to vascular physiology.

	FORMATION OF PROSTANOIDS

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

Prostaglandins and TxA₂, collectively named prostanoids, are mainly derived from arachidonic acid. Cyclooxygenase (COX) is the enzyme responsible for the committed step in the conversion of arachidonic acid to the bioactive prostanoids. There are two separate genes encoding COX proteins, COX-1 and COX-2. Both isozymes of COX catalyze the same reactions and share similar enzyme kinetics (18, 116), and their crystal structures are essentially superimposable (195, 218). The metabolism of the COX product PGH₂ is the first committed step in the production of prostanoids. Thereafter, each primary bioactive prostanoid is synthesized by way of individual enzymes. The catalysis of PGH₂ to PGE₂ occurs mostly via PGE synthase (PGES); PGE₂ can also be formed nonenzymatically from endoperoxides (183). There exists possibly two cytosolic glutathione (GSH)-dependent human isoforms purified from cerebrum (158) and two microsomal forms, of which one is GSH dependent and the other possibly independent (226). The molecular identity of PGES has more recently been determined (102, 136, 206); it appears that a GSH-dependent microsomal form is mainly coupled to COX-2 and the cytosolic one to COX-1 (136, 206). There are currently three known PGD synthases (PGDS): 1) GSH-independent PGDS (GSH-I-PGDS) (also called brain-type PGDS) (190), 2) GSH-dependent PGDS (GSH-D-PGDS) (also called spleen-type PGDS) (41), and 3) GSH S-transferase (GST) (214); in addition, serum albumin catalyzes this conversion as well (45). PGF synthase (PGFS) is a dual functioning enzyme that requires NADPH, which catalyzes the reduction of PGH₂ to PGF₂ and of PGD₂ to 9,11-PGF₂ (a stereoisomer of PGF₂) (224) at two distinct sites. Concurrent but independent investigations into the mechanisms of PGI₂ (229) and TxA₂ (86, 215) synthesis led to the suggestion that the enzymes responsible for catalysis are analogous to P-450 monooxygenases. Subsequent purification and cDNA cloning of the PGI₂ (56, 79) and TxA₂ (159, 188) isomerases verified these predictions. Neither synthase has greater than 16% sequence identity to any other P-450 enzymes or to each other such that each constitutes its own subfamily, designated CYP8 and CYP5 for PGI₂ synthase and TxA₂ synthase, respectively. Of physiological relevance, because of inherent characteristics and reductive cofactor requirements (GSH and NADPH) for PGI₂, PGE₂, PGD₂, and PGF₂ synthases, TxA₂ synthesis is preferentially preserved over that of the other prostanoids during peroxidative processes (3, 6, 89, 182, 207, 223, 227).

High levels of prostaglandins, especially PGE₂, are detected in the blood and brain of the neonate (103, 129, 134). In retina, both increased COX-1 and COX-2 activities contribute to the augmented production of neonatal prostaglandins (81). In the brain, however, this mostly arises from increased expression and activity of the COX-2 pathway in brain vasculature, as opposed to adult brain, where prostanoid formation is catalyzed mainly by COX-1 (167). The rapid drop in prostaglandin levels in brain within the first 48-72 h after birth (103) is associated with a relative decrease in COX-2 expression, which seems to increase again thereafter (165). In the ductus arteriosus both COX-1 and COX-2 are involved in prostaglandin generation, but these enzymes play distinct roles (45, 49, 75, 203). COX-1 dominates throughout gestation in PGE₂ formation and regulation of ductal tone, whereas COX-2 is induced at term (49, 75), possibly by increasing oxygen tension perhaps via endothelin formation; in addition, because COX-2 is activatable by endotoxins, this enzyme may contribute more importantly to ductal tone in the prematurely born subject (49).

	EFFECTS OF PROSTANOIDS ON CEREBRAL AND OCULAR VASCULAR TISSUE AND DUCTUS ARTERIOSUS

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

Prostanoids produce significant effects on circulation. They mediate vasomotor response to asphyxia, ischemia, hypercapnia, and hypotension, as well as both systemic and pulmonary hypertension (35-37, 66, 80, 91, 118, 120, 122, 141, 170, 228). The vasomotor effects of prostanoids vary not only by virtue of their nature but also as a function of the type of tissue and their development. PGD₂ and PGI₂ are for the most part vasorelaxants, PGF₂ and TxA₂ are constrictors, and the effects of PGE₂ depend on the type of tissue, which takes into account the receptor and receptor coupling; as will be discussed later, these seemingly opposing effects of PGE₂ are due to the existence of multiple receptor subtypes with different signal transduction pathways that are often coexpressed.

The developmental profile of prostaglandin-mediated actions has been thoroughly investigated, mostly in the ductus arteriosus, as well as in the vascular beds of the brain and eye. On retinal and brain pial microvasculature and main arteries, PGE₂ seems to be an important vasodilator in the neonate (88, 119) and, as a result, probably contributes to the lower cerebral and retinal vascular resistance characteristic of fetal life. Prostanoids have been implicated in cerebral and retinal blood flow autoregulation in the adult and newborn subject (34, 36, 38, 85, 169). PGE₂, along with PGI₂ and PGF₂, is released from brain and ocular vasculature in response to acute rises in systemic blood pressure (34-36). Because PGF₂ (and PGE₂ on parenchymal vessels) causes minimal constriction in newborns compared with adults (7, 88, 125), the balance of prostaglandin action is shifted onto PGI₂-elicited [and possibly PGD₂-elicited (5)] relaxation, which in turn curtails the upper limit of cerebral blood flow autoregulation. Accordingly, inhibition of COX and more specifically of COX-2 in the newborn unveils an otherwise appropriate autoregulatory response (34, 36, 81, 126).

PGI₂ is readily released in response to conditions that evoke adaptive responses, such as during acute hypercapnia, hypotension, and ischemia, and exerts a major role in increasing blood flow to the central nervous system (66, 118, 119, 121, 166, 197, 198). However, it is of interest that in response to prolonged hypercapnia, PGE₂ [and not PGI₂ or nitric oxide (NO); Refs. 32, 99, 121, 122, 231] may become the main regulator of vascular tone, but this effect is mediated indirectly through induction of endothelial NO synthase (eNOS; Ref. 141), consistent with a role for PGE₂ in governing the expression of this enzyme (62). Along these lines, PGE₂, by acting through its EP₃ receptor, exerts a dominant role in controlling the increased constitutive NOS expression in brain and microvasculature of the newly born subject (62, 63); in choroid, however, it is PGD₂, via the DP receptor, that seems to govern increased neonatal eNOS expression (61). Thus the high levels of prostaglandins during birth transition regulate constitutive NOS expression, which in turn contributes to the relatively limited cerebral and ocular autoregulatory ability of the neonate (61, 83). Contrary to PGE₂ and PGF₂, the effects of PGI₂ on brain vasculature do not differ as a function of age (88). This age-dependent action of PGI₂ varies, however, according to the tissue, such that in retina PGI₂ is a less effective relaxant in the newborn than in the adult (7), whereas in choroid the reverse is observed (5). Along the same lines, the relaxant effects of PGD₂ are greater in the immature than the mature subject (5).

TxA₂ is largely responsible for the delayed effects of asphyxia-ischemic insults, as observed in retinal, choroidal (3, 6, 37), pulmonary (207), and placental (223) circulatory beds. These compromising actions of TxA₂ are further facilitated during oxidant stresses, especially in the immature subject (3, 6, 97) relatively devoid of antioxidants (55, 208); correspondingly, during oxidative stresses, TxA₂ is generated more abundantly in the newborn than in the adult (6, 97). In this context TxA₂ has been found to mediate the effects of major stable products of peroxidation, the isoprostanes (97, 114). Furthermore, not only is TxA₂ production greater in the immature than in the mature subject, but also TxA₂-induced constriction is more pronounced in fetus and newborn than in adult (7, 97). Moreover, an effect previously unidentified for TxA₂ acting via TP was recently unveiled, specifically, neuromicrovascular degeneration (Fig. 1) resulting from endothelial cell death, which contributes to the vasoobliteration in models of retinopathy of prematurity (22).

View larger version (60K): [in this window] [in a new window]   Fig. 1.   Role of thromboxane A2 (TxA2) mimetic on retinal microvascular degeneration. TxA2 mimetic U-46619 (20 pmol; estimated ocular concentration 0.5 µM) was injected (0.4 µl; eye volume ~40 µl) preretinally in vitreous of rat pups on postnatal days 7, 9, and 11 in the absence (B) or presence of TP antagonist (C; CGS-22652; gift from Novartis); control animals (A) received same volume (0.5 µl) of saline. Note markedly reduced vascular density (ADPase staining) on postnatal day 12 of U-46619-exposed rats; this effect is prevented by CGS-22652 and corroborates role of TP in vasoobliteration in models of retinopathy of prematurity (22).

In ductus arteriosus PGE₂ is the major prostaglandin that affects tone (48). Although the ductus generates more PGI₂ than PGE₂, PGE₂ is markedly more potent in relaxing this vessel (46). The other prostanoids play small, if any, physiological roles (191, 192, 194). The effects of PGE₂ on ductal tone are also developmentally regulated with loss of responsiveness in the immediate neonate compared with the fetus (1, 47). Physiological mechanisms contributing to decreased actions of circulating PGE₂ include loss of the placenta, which is the major source of circulating PGE₂ in the fetus (210), an increase in pulmonary blood flow at birth, because the lungs are a major site of prostaglandin catabolism (213), and a decrease in relaxant PGE₂ receptors (23, 29). Increased oxygen tension and PGE₂ oppose each other's actions in controlling ductal tone (191); in addition, elevated oxygen tension also inhibits PGI₂ synthase, which decreases formation of ductal PGI₂, thus promoting contraction of the ductus arteriosus (192, 193).

	MOLECULAR CHARACTERISTICS OF PROSTANOID RECEPTORS

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

As discussed above, prostanoids exert their effects through their respective receptors. Changes in prostanoid actions as a function of development take into account receptor density and coupling. Before proceeding with a description of the ontogeny of prostaglandin receptors we will review the pharmacology, molecular biology, and signaling of these receptors.

The earliest prostanoid receptor classification system was based on the differential effects of PGE and PGF analogs on three isolated tissues (guinea pig uterus, human myometrium, and rabbit jejunum) (171). Subsequent reports supported directly (72) or indirectly (12) the existence of multiple receptor types, and Kennedy et al. (108) proposed a prostanoid receptor classification. Using a comparison of the rank orders of agonist potency in a range of smooth muscle preparations and prior evidence in the literature, these investigators hypothesized the existence of distinct receptors for each of the bioactive prostanoids. Under the proposed classification, the receptors for PGD₂, PGE₂, PGF₂, PGI₂, and TxA₂ are denoted DP, EP, FP, IP, and TP, respectively. Studies also identified PGE-sensitive tissues that exhibited different activities to specific agonists and antagonists, prompting a further division of the EP receptors into EP₁, EP₂, EP₃, and EP₄ (52, 53). In addition, a newly identified receptor that binds PGD₂, named chemoattractant receptor-homologous molecule on TH2 cells because of its properties, has been uncovered in T helper type 2 cells, eosinophils, and basophils (92); however, this seven-transmembrane PGD₂ receptor does not contain characteristic molecular signatures of prostanoid receptors but rather of chemokine receptors (139).

The cloning of the prostanoid receptors identified further heterogeneity, which arises as a result of alternative mRNA splicing. The molecular mechanism of alternative mRNA splicing involves the formation of multiple mRNA and protein products from a single gene. Specifically, splice variants have been identified for the EP₁, EP₃, FP, and TP receptor homologs of various species. There are currently nine known subtypes of the human EP₃ receptor: EP_3-1a, EP_3-1b, EP_3-II, EP_3-III, EP_3-IV, EP_3-V, EP_3-VI, EP_3-e, and EP_3-f (8, 111, 176). Mouse (100, 152, 199), rat (204), bovine (145), and rabbit (31) counterparts have also been identified for some of these EP₃ subtypes; subtypes of the EP₃ receptors are distinguished by variances in the tail of the carboxyl-terminal portion. Human subtypes also exist for the TP receptor (174), specifically TP and TP, which, as for EP₃ subtypes, vary by the tail of their carboxyl-terminal end. Two splice variants have been identified for each of the rat EP₁ (160) and ovine FP (172) receptors. These have been designated as EP₁ and EP_1-variant and as FP_A and FP_B, respectively; variants of EP₁, FP, and TP also differ between each other by their carboxyl-terminal portions. Human homologs to EP₁ and FP subtypes have not been identified.

The eight known types of prostanoid receptors are each encoded by an individual gene. Phylogenetic analyses indicate that receptors sharing a common signaling pathway have higher sequence homology than receptors sharing a common prostanoid as their preferential ligand (28, 177, 212). The effects of prostanoid receptors on smooth muscle reflect this relationship. Thus EP₂, EP₄, DP, and IP induce smooth muscle relaxation and are more closely related to each other than to the other prostanoid receptors. Similarly, EP₁, FP, and TP receptors cause smooth muscle contraction and form another group based on sequence homology. The EP₃ receptors also usually stimulate smooth muscle contraction and define a third group. The signal transduction pathways underlying these mechanisms of prostanoid action are shared within these groups and will be explained in a section below. On the basis of these phylogenetic analyses, it has been suggested that the COX pathway may have evolved from PGE₂ and an ancestral EP receptor (147). The evolution of the different EP receptor types from this ancestral prostanoid receptor would have linked PGE₂ to different signal transduction pathways. The receptors for the other prostanoids would have then evolved by gene duplication of these different EP receptor subtypes.

Alternative splicing of the exon encoding the seventh transmembrane domain occurs at a position approximately 9-12 amino acids into the carboxy terminus of the EP₃, FP, and TP receptors of various species. The rat EP₁ receptor is also subject to alternative splicing but instead diverges midway into the sixth transmembrane domain. The variant form (rEP_1-variant) contains none of the amino acids that are highly conserved within the seventh transmembrane domain of the other prostanoid receptors. Generally, prostanoid receptor isoforms exhibit similar ligand binding but differ in their signaling pathways, their sensitivity to agonist-induced desensitization, and their tendency toward constitutive activity, as will be discussed. Whereas there is homology between the EP₃ receptor isoforms of different species, the human and mouse TP receptor isoforms demonstrate no homology. This may be indicative of other TP isoforms (147). The receptors that are subject to alternative splicing (EP₁, EP₃, FP, and TP) are phylogenetically related, perhaps suggesting the evolutionary conservation of the sequence(s) involved in this process. The rEP₁, EP_3-II, FP_A, and TP splice variants are all generated by the failure to utilize a potential splice site (173). The splicing out of various introns and the use of downstream exons generate the other alternatively spliced forms of EP₃. The regulation of the process of alternative splicing with respect to the prostanoid receptors has yet to be studied.

Prostanoid receptors are rhodopsin type, containing seven transmembrane domains, an extracellular amino terminus, and an intracellular carboxy terminus; accordingly, three intracellular and three extracellular loops are found. The three-dimensional crystal structure of a membrane-contained mammalian GPCR, only accomplished specifically for rhodopsin so far, confirms this topology (163). The evidence suggesting that prostanoid receptors are rhodopsin-type receptors precedes their molecular cloning and was based on coupling of the TP receptor to G proteins (17) and its sensitivity to agonist-induced desensitization (138). The purification of the TP receptor from human platelets (216) allowed its partial protein sequence to be identified, leading to the isolation of a cDNA for TP (93). Homology screening based on this sequence has established recombinant clones from various species for the eight individual prostanoid receptors previously defined pharmacologically. There are 28 amino acid residues conserved within all prostanoid receptor sequences, and 8 of these are shared with other GPCRs. These residues are believed to be particularly important in receptor structure and/or function. For instance, an Asp residue in the second transmembrane domain of various GPCRs is involved in ligand binding and signal transduction (184), although the role of this residue has not been studied directly in any prostanoid receptors. Two conserved Cys residues (1 in each of the 1st and 2nd extracellular loops) are suggested to form a disulfide bond and contribute to the stabilization of GPCRs in the membrane (59). In the rabbit EP₃ receptor, an Ala residue substituted for the Cys residue in the second extracellular loop had no effect on binding (14). This is in contrast to studies of the human TP receptor, where substitution of Ser for the analogous Cys completely abolished ligand binding (39, 54). These studies also showed a similar effect by mutating the Cys of the first extracellular loop, which corroborates evidence demonstrating a loss of agonist binding to TP after chemical perturbation of these Cys residues (with dithiothreitol or sulfhydryl alkylation) (60).

There are several prostanoid receptor-specific motifs, including sequences in the second extracellular loop (G-R-Y-X-X-Q-X-P-G-T/S-W-C-F) as well as in the third and seventh transmembrane domain (M-X-F-F-G-L-X-X-L-L-X-X-X-A-M-A-X-E-R and L-X-A-X-R-X-A-S/T-X-N-Q-I-L-D-P-W-V-Y-I-L, respectively). These conserved regions are thought to play fundamental roles in the structure of the prostanoid binding domains. An Arg in the seventh transmembrane domain conserved between all prostanoid receptors was proposed to be the binding site for the carboxy moiety of the prostanoids (14, 98); the conserved motif in the second extracellular loop may also function in this regard (14). Residues are also conserved among prostanoid receptors for signal transduction. An Arg in the first intracellular loop is conserved between all prostanoid receptors. A mutation of this residue to Leu in the TP receptor was associated with a bleeding disorder (94).

As well as containing conserved residues, prostanoid receptors share several other characteristics with other members of the GPCR family. They enclose consensus sites for N-glycosylation of Asn residues (Asn-X-Ser/Thr) in their extracellular domains. The amount of glycosylation can be significant as demonstrated when purified TP receptors of molecular mass ~57 kDa were shifted to their predicted molecular mass (based on primary structure) of ~37 kDa on treatment with N-glycanase (131). The glycosylation sites are requisite for ligand binding at the human TP receptor (40), and mutation of the Asn residues or deletion of their carbohydrate moieties abolishes binding. In addition, it has recently been found that N-glycosylation of the EP₃ receptor is required for localization to the plasma membrane but not for appropriate folding of the protein (26).

	RECEPTOR SIGNALING

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

Early studies of the second messengers downstream of the prostanoids focused on cyclic nucleotides (53). For instance, PGE₂ and PGF₂ were reported to stimulate cAMP (33) and cGMP (64), respectively. Since then, other signal transduction pathways have been suggested by the observation of prostanoid-mediated activation of second messengers such as free Ca²⁺ and inositol phosphate (147). The molecular cloning of the prostanoid receptors has facilitated observation of their coupling to heterotrimeric G proteins. These heterotrimeric G proteins are composed of three structural subunits designated , , and , of which numerous subtypes exist for each (78). Functionally, G proteins comprise two subunits, since receptor activation provokes the dissociation of the G-subunit from a complex of the G-subunits. Both the G- and the G- subunits can act as effectors in signal transduction (44, 178).

Prostanoid receptors sharing a common signaling pathway have higher sequence homology than receptors sharing a common prostanoid as their preferential ligand (28, 177, 212). Thus three clusters of related receptors have been defined: 1) DP, IP, EP₂, and EP₄; 2) EP₁, FP, and TP; and 3) EP₃. Prostanoid receptors in group 1 are linked to heterotrimeric G proteins that are composed of a G-subunit that stimulates adenylate cyclase (designated G_s) to produce cAMP. An increase in intracellular cAMP concentration is observed after stimulation of the recombinant human DP (28), IP (27, 105, 142), EP₂ (177), and EP₄ (11, 20) receptors, in addition to their species homologs. The results obtained with recombinant receptors corroborated those obtained previously in isolated tissues. For instance, PGD-, PGE-, and PGI-responsive receptors cause the stimulation of cAMP production in platelets (230) and in vasculature (2, 5, 81). However, the recombinant human IP receptor can also mediate inositol phosphate production and increases in free Ca²⁺ levels by coupling with G_q (144). Likewise, EP₂, EP₄, and DP receptors in choroid do not couple to adenylate cyclase but rather to eNOS (2, 5); this may be evoked by G action on phosphatidylinositol 3-kinase (44), which in turn activates, sequentially, protein kinase B (PKB) (219) and eNOS (58, 71). Of interest, ontogenic differences in the expression of EP₂, EP₄, and DP in ocular vasculature do not explain the greater relaxation evoked by their stimulation in the young subject; rather, it is the greater expression of the eNOS per se that is responsible for their augmented vasomotor actions (2, 5, 84).

Prostanoid receptors in group 2 couple to increases in intracellular free Ca²⁺ through the activation by G_q of phospholipase C, with subsequent inositol phosphate liberation. This pathway has been demonstrated for FP using anti-G_q antibodies (101), which corroborates earlier results demonstrating inositol phosphate turnover in isolated luteal cells on PGF₂ administration (175). In the case of TP, G_q activation is the primary effector pathway (189) as shown during stimulation of native TP receptors in platelets (13). However, the previously described TP receptor splice variants TP and TP also can signal through G_i and G_s to inhibit and stimulate adenylate cyclase, respectively (95). EP₁ preferentially couples to G_q (53). An increase in inositol phosphate after its stimulation in brain and ocular vasculature is clearly observed (4, 128).

The EP₃ subtypes constitute group 3 of the prostanoid receptor family and employ as their primary effector pathway the inhibition of adenylate cyclase through the G_i-family (149). However, the molecular cloning of the bovine EP₃ receptor splice variants demonstrates the array of second messengers to which these receptors are coupled. Four subtypes of bovine EP₃ have been cloned (designated A, B, C, and D), and all show identical agonist binding properties (145). However, EP_3A acts through G_i to inhibit adenylate cyclase, EP_3B and EP_3C signal through G_s to activate adenylate cyclase, and EP_3D is coupled to G_i, G_s, and G_q, resulting in the inhibition and activation of adenylate cyclase as well as the activation of phospholipase C. On the other hand, nuclear EP₃ receptors seem to be G protein dependent but not coupled to adenylate cyclase or phospholipase C activation (25). A novel type of G protein regulation has also been reported for the EP_3B and EP_3C receptors. In addition to their stimulatory effects on G_s, they are thought to negatively regulate G protein activity by specifically inhibiting the GTPase activity of G_o, a member of the G_i-family (150). Along the same lines, EP_3D-induced ductus arteriosus relaxation is pertussis toxin, NO, and endothelium insensitive but dependent on ATP-sensitive potassium channel activation (29); the mechanisms remain to be elucidated, although direct receptor-channel interaction is a possibility (130). The EP₃ receptor subtypes may also differ in their levels of constitutive activity, the agonist-independent activity of the receptor. This evidence comes from studies of the mouse isoforms of the EP₃ receptor (designated , , and ). EP₃ is not constitutively active, whereas EP₃ is, with respect to the G_i-mediated inhibition of adenylate cyclase (148); the demonstration of levels of activity similar to EP₃ on treatment of EP₃-transfected cells with pertussis toxin (which inactivates G_i) confirms that the EP₃-G_i is constitutively active.

	PROSTANOID RECEPTOR DISTRIBUTION AND DEVELOPMENTAL CHANGES

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

Among the PGE₂ receptors, EP₃ and EP₄ are the most widely distributed and are detected in nearly every tissue (53, 147). EP₁ and EP₂, on the other hand, have a limited distribution. EP₁ is present mostly on vascular and nonvascular smooth muscle (38, 53), and EP₂ is present mostly in lung, placenta, endometrium, renal tubules, brain synaptosomes, and heart (127, 177), as well as on vascular and nonvascular smooth muscle (7, 117). FP receptors are found in corpus luteum, iris sphincter muscle, trabecular meshwork, and vascular smooth muscle (53, 147). IP and TP receptors are primarily localized in platelets and vascular smooth muscle; IP is also present in dorsal root ganglia. DP is the least abundant prostanoid receptor and is primarily found in retina; DP is also present in leptomeninges and choroid plexus (53, 147) and is weakly expressed in gut and lungs.

Studies on plasma membrane of brain microvessels revealed a two- to threefold lower density of EP and FP receptors in newborn compared with adult (128). In both porcine newborn and adult brain intraparenchymal microvessels, ~80% of PGE₂ receptors are of the EP₁ subtype and the remainder are EP₃ (128); EP₂ and EP₄ are undetectable. Accordingly, stimulation of EP₁, EP₃, and FP evokes less constriction of intraparenchymal brain vessels in the newborn than in the adult (125). Similar functional observations have been made on retinal and choroidal vessels (2, 5, 7). However, these ocular vascular beds exhibit a different profile of EP₂ and EP₄ expression in the developing subject, such that in retinal microvessels EP₄ is not found, whereas EP₂ density is greater in adult than in newborn, and in choroid all EP receptors are expressed more in the mature subject with the exception of EP₂, which is equivalently expressed in newborn and adult. The density of EP₂ and EP₄, however, does not readily translate functionally in choroid, where EP₂- and EP₄-induced relaxation is significantly greater in younger subjects (2). The same applies to DP, which exhibits a similar density in newborn and adult brain, retinal, and choroidal vessels but markedly greater relaxation in the immature subject (Ref. 2; personal observations). The developmental profile of IP and TP receptor expression on vascular beds has not been determined; functional assays using specific agonists have been described above. A summary of prostanoid receptor differences in newborn and adult brain and ocular vascular beds is presented in Table 1.

Although PGE₂ plays a cardinal role in fetal ductal tone, the types of EP receptors implicated have only recently been identified. EP₂, EP_3D- (the only EP₃ detected in ductus; Ref. 29), and EP₄ receptors have been detected in fetal porcine and ovine ductus arteriosus (0.75 gestation) (23, 29); EP₁ was not detected. EP₄ has been suggested to confer the principal relaxant function induced by PGE₂ (192). These ex vivo findings were substantiated in vivo in fetal sheep (~0.75 gestation) by using a selective EP₄ antagonist (THG-213); THG-213 caused a significant decrease in ductal diameter without changes in pulmonary blood flow and systemic pressures (unreported observations) (Fig. 2), consistent with a major role for EP₄ in fetal ductal tone (192).

View larger version (54K): [in this window] [in a new window]   Fig. 2.   Effect of EP4 antagonist (THG-213) on ductus arteriosus diameter of fetal sheep. Fetal sheep (~0.75 gestation) with intact placental circulation were injected with THG-213 (1 mg/kg, near maximum efficacious dose) via the umbilical vein. Ductus arteriosus diameter was measured echocardiographically before (A) and after drug injection (B). C: significant time-dependent decrease in diameter after administration of the selective EP4 antagonist THG-213 (gift from Theratechnologies); its effect approaches that of the nonselective cyclooxygenase (COX) inhibitor indomethacin (1 mg/kg). Arrowheads inserted on echocardiographic photos (A and B) point to ductus arteriosus.

The mechanisms responsible for the perinatal-related decrease in PGE₂ responsiveness of ductus arteriosus have recently been investigated. It has been observed that PGE₂ receptor density decreased by threefold in the immediate newborn compared with the fetus (0.75 gestation). As indicated above, EP₂, EP_3D-, and EP₄ receptors are present in fetus ductus in equivalent densities, whereas only EP₂ is found in the immediate postnatal ductus (23, 29). Surprisingly, but of relevance, EP₃ stimulation in ductus arteriosus was associated with relaxation in both fetal rabbits and lambs (29, 194). Thus a loss of EP_3D- and EP₄ receptors in the immediate newborn along with preservation of EP₂ density and function was consistent with a decreased response of the newborn ductus to PGE₂; hence EP₂ appears to mediate the vasorelaxant effects of PGE₂ in full-term neonatal ductus arteriosus.

Prostanoids function in an autocrine, intracrine, and/or paracrine manner. The wide tissue distribution of prostanoids and their receptors underscores the various effects mediated through their interactions. Several lines of evidence suggest that prostanoids may act not only on cell-surface receptors, but also intracellularly at receptors other than peroxysome proliferator-activated receptors (90, 110, 168), which mediate actions of metabolites of PGD₂ (mostly PGJ₂ and 15-deoxy-^12,14-PGJ₂) but not of PGE₂ and PGF₂ (110, 232). The enzymes responsible for prostaglandin synthesis, cytosolic phospholipase A₂ (PLA₂, which releases arachidonic acid) and COX-1 and -2, have been found to be located at the nuclear membrane, and these can also translocate to this site in response to stimuli (135, 185, 196). A membrane-associated transporter that facilitates influx of prostanoids has been identified (186). A number of other observations pointed toward possible intracellular sites of action for major prostaglandins. For instance, a robust increase in eNOS expression was observed after stimulation of EP₃ despite a marked paucity of plasma membranal EP₃ receptors in newborn microvessels (62). Similarly, the total absence of EP₂ receptors in plasma membrane preparations of the newborn brain was inconsistent with the neuroprotective effects of EP₂ stimulation in the neonate (140). Thus it was proposed that prostanoids may exert effects via intracellular receptors in the vicinity of their formation. This inference has recently been tested and corroborated. DP and all known subtypes of EP receptors were detected in nuclear membrane fractions using binding studies, where their relative distribution differed from that in plasma membrane. In addition, the presence of EP₁, EP₃, and EP₄ on perinuclear membranes was identified on native endothelial cells by immunocytochemical techniques employing confocal as well as electron microscopy (24, 25). Furthermore, cells transfected with the receptor containing exogenous immunogenic epitopes revealed robust perinuclear localization of the receptor, an example of which appears in Fig. 3 (COS cells, which normally do not express prostanoid receptors, transfected with EP₃ tagged amino terminally with FLAG). The same intracellular localization was observed in HEK-293 cells (which normally do not express prostanoid receptors) transfected with EP receptors fused to green fluorescent protein (24). The functionality of these receptors was confirmed by stimulating isolated nuclei with EP agonists, where a dose-dependent increase in nuclear calcium entry was revealed in addition to an induction of gene transcription for c-fos (24) and inducible NOS (iNOS) (25). Furthermore, inhibition of the prostaglandin transporter in whole endothelial cells stimulated with PGE₂ prevented eNOS transcription (62). Although the factors and molecular domains involved in cellular localization as well as the signaling mechanisms remain to be identified, these studies revealed for the first time the presence of functional GPCRs at the nuclear membrane and provided a mechanism by which PGE₂ could affect expression of constitutive NOS in the developing subject. A model depicting this regulation of NOS by PGE₂ acting via nuclear EP₃ receptors is presented in Fig. 4. These findings set forth new perspectives in the biology of prostanoid receptors.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Cellular immunolocalization of the EP3 receptor. COS cells, which normally do not express prostanoid receptors, were transfected with a plasmid (pCMV-TAG2 vector) containing the cDNA for human EP3 fused downstream to the FLAG epitope. Untransfected cells (A) do not react to FLAG antibody. Transfected cells immunoreact to the FLAG antibody, revealing localization to the plasma membrane as well as the nucleus (B); the latter localization is concentrated in the perinuclear region, as noted by the perinuclear halo (C), consistent with cellular EP3 immunoreactivity by confocal and electron microscopy (24, 25).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Proposed mechanism for developmental and hypercapnia-induced regulation of endothelial nitric oxide synthase (eNOS) by high perinatal levels of prostaglandin (PG) E2 acting via perinuclear EP3 receptors in neural vasculature. PGs, which are in high level in the perinate and during hypercapnia, can be produced at the plasma membrane or nuclear envelope catalyzed by phospholipase A2 (PLA2) and COX pathways. PGs, specifically PGE2, can act at the vicinity of their formation on EP3 on the nuclear envelope; PGs generated at the cell surface can also be incorporated via PG transporter (PGT) to act intracellularly on perinuclear EP3. Stimulation of nuclear membrane EP3 can induce transcription of eNOS gene. In the perinate the production of PGs, especially PGE2, is augmented; this induces increased eNOS expression. The increased eNOS expression leads to augmented NO generation, which causes relaxation of the neural vasculature; this may lead to excess brain blood flow and/or oxygen delivery to the retina, predisposing, respectively, to intraventricular cerebral hemorrhage (IVH) and retinopathy of prematurity (ROP) in the developing subject. AA, arachidonic acid. CBF, cerebral blood flow.

	PROSTANOID RECEPTOR REGULATION: PHYSIOLOGICAL IMPLICATIONS IN THE DEVELOPING SUBJECT

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

Prostanoid receptor expression regulation is not yet clearly elucidated. Different factors act on the cis-acting regulatory elements of their respective genes (21, 69, 106, 111, 156, 157). The 5'-flanking region and first intron of many of these receptor genes have basal promoter motifs (such as a TATA box), as well as several responsive motifs, including those for proinflammatory agents (such as nuclear factor-B). Although many different motifs have been identified, less information is available regarding the actual regulation of receptor gene expression. The studies to date suggest that species or cell-type differences may play a role. For instance, the TP receptor contains a phorbol ester response element, and TP receptor expression can be stimulated in human erythroleukemia (HEL) cells (143). However, despite the presence of response elements for glucocorticoids and interleukin (IL)-6 in the TP gene, these factors are insufficient to induce TP expression in HEL cells (109). They can induce TP expression in rat cultured vascular smooth muscle cells (202). In vivo regulation of prostanoid receptor expression has not yet been studied directly, however.

Mechanisms that regulate translation also partake in governing receptor expression, but these remain poorly defined. Despite the high homology between the prostanoid receptors of various species, there are differences in the translation initiation sites of some receptor types that affect the amino-terminal extracellular domain of the receptor (147). For instance, the human, bovine, and rabbit EP₃ receptor is 20 amino acid residues longer than the rat and mouse homologs. The human IP receptor is 30 amino acids shorter than the mouse and rat homologs. In contrast, the human DP receptor is only one residue longer than its mouse and rat counterparts.

Desensitization of the prostanoid receptors is also not fully understood. Most of the prostanoid receptors exhibit agonist-induced (homologous) downregulation of their receptors, but the time dependence varies according to the receptor type. Many of the receptors are known to be susceptible to the second messenger kinases, but the role for GPCR kinases (GRKs) is not clear. GPCRs are often posttranslationally modified at their cytoplasmic domains. Phosphorylation of Ser and Thr residues occurs as part of the process of receptor desensitization, which is the waning of receptor response after persistent stimulation (9, 67, 123). Heterologous desensitization involves feedback inhibition by the second messenger kinases that these receptors activate (i.e., PKA and PKC). Stimulation of these pathways can cause phosphorylation and subsequent desensitization of any GPCR containing the appropriate PKA or PKC consensus site(s) (Ser/Thr-X-Lys/Arg and Arg-X-X-Ser, respectively). Homologous desensitization is an alternate pathway that involves receptor phosphorylation by a specific GRK and its subsequent binding by arrestin, a protein that sterically restricts signaling to the G protein. Desensitization of all prostanoid receptor types except DP has been demonstrated. Many reports identify the importance of heterologous desensitization in the regulation of prostanoid receptor activity, and a role for homologous desensitization has only recently been suggested.

Both short-term (5 min) and long-term (24 h) desensitization of the mouse EP₁ receptor involve PKC, and the desensitization is observed as a suppression of the agonist-mediated dose response and a reduction in EP₁ mRNA levels, respectively (104). The mouse EP₂ receptor undergoes long-term agonist-induced desensitization in the form of receptor downregulation after a 12-h exposure to PGE₂ but is insensitive to short-term desensitization (30 min) (155). Short-term (1 h) desensitization of human EP₂ was observed on PKC activation (233). Splice variants of a given species homolog of EP₃ differ in their sensitivity to desensitization. The mouse EP₃ exhibits sequestration after short-term (30 min) PGE₂ exposure and downregulation on long-term (24 h) PGE₂ exposure, while mouse EP₃ does not undergo agonist-induced desensitization (152, 153). Similarly, the human EP_3-II demonstrates slow persistent desensitization in contrast to the rapid transient desensitization exhibited by the EP_3-III and EP_3-IV receptors (10). Desensitization of EP₄ is perhaps the most intensely studied to date of the EP receptors. The mouse EP₄ receptor is sensitive to both short-term (30 min) and long-term (12 h) desensitization (155). The rapid agonist-induced desensitization of human EP₄ is independent of second messenger kinases and is instead regulated by GRKs (19, 153). Even though the desensitization of the IP receptor has been studied for the native receptor in a single cell type (neuroblastoma-glioma cell hybrid, NG108-15), the results are contradictory. Long-term (17 h) desensitization has been observed as receptor sequestration and downregulation with no change in agonist affinity (113), where the kinases involved appear to depend on the agonist used (112). In contrast, another group has identified that long-term desensitization is specifically accompanied by a concurrent downregulation of the G_s-subunit (9). PKC is observed to mediate short-term (45 min) desensitization of the native FP receptor in bovine iris sphincter (201). Short-term desensitization in vivo of the FP receptor also occurs in the ovine corpus luteum, where its temporal nature has been suggested to influence the oxytocin-mediated pulsatile release of PGF₂ (115). The TP receptor has been shown to undergo desensitization, specifically as short-term (10 min) agonist-dependent phosphorylation, which could be blocked by antagonist (161). The desensitization of the phospholipase C effector pathway downstream of TP has also been shown in human platelets (65). A recent report demonstrates that the TP splice variants exhibit different sensitivities to desensitization; TP is susceptible to agonist-induced desensitization in a GRK-dependent manner, while TP does not desensitize (164). In contrast to the other prostanoid receptors, there are no direct reports of desensitization of the DP receptor.

Because prostanoids, especially prostaglandins and to a much lesser extent TxA₂, are increased during the birth transition period (6, 103, 125, 134), and because the density of many prostanoid receptors is decreased during this time (see Table 1), homologous developmental regulation of the prostanoid receptors has been invoked. A reduction in prostanoid levels in newborn brain, retina, and choroid to those of the adult, by titrating the dose of the COX inhibitor ibuprofen (24-48 h) (81), led to a concomitant increase in [³H]PGE₂ and [³H]PGF₂ binding in vascular tissue (2, 81, 125, 126). This increase in prostaglandin receptors could be specifically prevented by PGE₂ or PGF₂ agonists, suggesting a cause-effect relationship between the binding and the prostaglandin levels in the immediate newborn period (125, 129); receptor-coupled mechanisms seemed unaffected. However, this homologous upregulation was not uniformly observed for all prostanoid receptors. The constriction-evoking EP₁, EP₃, and FP exhibited a homologous regulation by the high PGE₂ and PGF₂ tissue levels during the birth transition period, and densities of these receptors increased to adult values when prostaglandin levels were reduced to those of the adult. In contrast, the densities of the relaxation-inducing EP₂, EP₄, and DP receptors in brain and ocular vasculatures were unaffected by prostaglandin synthase inhibition (2, 5, 81); likewise, IP and TP receptor-associated functions were also unaltered (5). The absent modulation in TP-induced constriction may be related to the minor changes in TxA₂ levels in the neonate, but this does not apply to EP₂, EP₄, IP, and DP receptors where PGE₂, PGI₂, and PGD₂ significantly change (5). Hence, changes in prostanoid receptor expression during the birth transition period can only in part be attributed to a homologous regulation by their ligands. The findings also disclose a possibly physiologically relevant pattern whereby high perinatal levels of prostaglandins lead to a downregulation of receptors associated with vasoconstriction without affecting those associated with relaxation, favoring altogether a vasorelaxation. An insufficient ability to limit blood flow and oxygen delivery to brain and retina in a premature subject may participate in the development of intraventricular cerebral hemorrhage and retinopathy of prematurity (85). However, in the term neonate this excess vasorelaxation may be regarded as beneficial, especially at the end of parturition when frequent episodes of hypoxia occur secondarily to intensifying uterine contractions (43, 57). A graphic scheme depicting the modulation of contractile prostaglandin receptors by endogenous prostanoids in the immediate neonate is presented in Fig. 5.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Homologous regulation of PG receptors by high levels of PGs in the newborn (NB) immediately after birth: possible contribution in the regulation of neurovascular tone. In the immediate neonate the production of PGs, especially PGE2 but also to some extent PGI2, PGD2, and PGF2, is high compared with the adult. As a result, the contraction-inducing EP1, EP3, and FP in brain and ocular microvasculature are downregulated in the young newborn. In contrast, no reduction in relaxation-inducing EP2, EP4, DP, and IP receptor expression and/or functions is detected. This may contribute to a net increased relaxation of the neural vasculature of the newborn compared with that of the adult. Augmented vasorelaxation may counter an appropriate increased perfusion pressure- or hyperoxia-induced autoregulatory response in the immature subject; the resulting increased brain blood flow and oxygen delivery to the retina may predispose, respectively, to IVH and ROP of the developing subject.

	PROSTANOID RECEPTOR GENE DISRUPTION

TOP ABSTRACT INTRODUCTION FORMATION OF PROSTANOIDS EFFECTS OF PROSTANOIDS ON... MOLECULAR CHARACTERISTICS OF... RECEPTOR SIGNALING PROSTANOID RECEPTOR... PROSTANOID RECEPTOR REGULATION:... PROSTANOID RECEPTOR GENE... CONCLUSION REFERENCES

Targeted gene ablation or disruption has provided important information on the role of prostanoid receptors, especially because of the lack of suitable antagonists for this receptor family. The disruption of all of the prostanoid receptor genes has now been reported.

There are three individual reports of the genetic disruption of the EP₁ receptor gene (EP₁^/) (16, 217, 225). A gender-specific effect on blood pressure homeostasis was observed. Surprisingly and as yet not understood, male but not female EP₁^/ mice exhibited less PGE₂-induced hypotension relative to wild-type mice (16), suggesting that in males most of the vasodepressor action of PGE₂ is mediated by EP₁. A decrease in the number of preneoplastic lesions in EP₁^/ mice relative to wild-type mice was found using a mouse model of colon cancer (225).

Of all of the prostanoid receptors, the ablation of the EP₂ receptor gene (EP₂^/) has been studied most intensely (15, 96, 107, 211). Contradictory effects of the EP₂^/ phenotype are apparent for blood pressure homeostasis. Both hypertension (107) and hypotension (211) are reported for EP₂^/ mice relative to wild type; the hypertension is consistent with the vasorelaxant effects of EP₂ stimulation, while the hypotension possibly relates to an effect of EP₂ on activation of the renin-angiotensin system (211). The hypotensive responses of PGE₂ are markedly attenuated in female but not male EP₂^/ mice (15), suggesting a major role for EP₂ in vasodepressor responses of PGE₂ in females. Salt-sensitive hypertension is observed in EP₂^/ mice, suggesting a role for EP₂ in the regulation of sodium in the kidney (107, 211). The EP₂ receptor plays an important role in reproduction since female EP₂^/ mice exhibit a reduced litter size, a decrease in ovulation number, and a reduced fertilization rate (96, 107, 211). An EP₂-dependent role was demonstrated in the expansion of the follicular granulosa cells surrounding the oocyte, which is a process known as cumulus expansion. These results suggest that the incomplete cumulus expansion observed in the absence of EP₂ contributes to the reduced ovulation and the failure of fertilization (96).

The knockout of the EP₃ receptor gene (EP₃^/) has also been reported (68, 205, 217). EP₃ is clearly involved in PGE₂-induced pyrexia because EP₃^/ mice fail to mount a febrile response to exogenous (i.e., lipopolysaccharide) and endogenous (i.e., IL-1) pyrogens (217). PGE₂ also functions through the EP₃ receptor to concentrate urine; however, these effects are deemed unessential for the normal regulation of urinary osmolality (68). The EP₃^/ mouse was also reported to be unable to secrete duodenal bicarbonate on luminal perfusion with PGE₂ relative to wild type (205). Because these EP₃^/ mice subsequently exhibit susceptibility to acid-induced injury, this receptor may function in maintaining mucosal integrity. EP₃ is also implicated in vasopressor responses, but this role is also gender specific; male but not female EP₃^/ mice exhibited more PGE₂-induced hypotension relative to wild-type mice (16), suggesting a pressor response attributed to EP₃ in males.

Individual reports of the effects of ablation of the EP₄ receptor gene (EP₄^/) concur that it has various functions in the vascular system (15, 16, 154, 187). The principal observation in EP₄^/ mice is their inability to close the ductus arteriosus immediately after birth; this causes pulmonary edema and death within 72 h of birth of nearly 100% of animals (154, 187). Interestingly, the ductus is unresponsive to COX inhibition, suggesting that other mechanisms are responsible for its patency; a role for NO in ductal patency is possible given its contribution in this regard (50, 70). When this mouse strain was bred on a mixed genetic background, ~20% of EP₄^/ survived (16); in female EP₄^/ mice hypotensive response to PGE₂ was attenuated, suggesting as for EP₂ a major role for EP₄ in vasodepressor responses of PGE₂ in females (15). The gender-related, pressor-mediated effects of EP receptors reflect sexual dimorphism of blood pressure regulation, but the mechanisms are not clear.

FP-knockout (FP^/) mice have been generated and studied for their reproductive function (200). Female FP^/ mice are fertile, carry their litters to term, but fail to undergo parturition. In mice this is largely because progesterone levels remain elevated because PGF₂ is responsible for the luteolysis that subsequently reduces progesterone, which otherwise maintains uterine quiescence and gestation. PGF₂ is also responsible for upregulating the uterine oxytocin receptors that facilitate parturition (74); this also contributes to the failure of initiating labor in FP^/ mice.

Ablation of the gene for the DP receptor (132) suggests a role for DP in allergic asthma. DP^/ mice demonstrated no difference compared with wild-type mice in total IgE after sensitization to ovalbumin. Howeve