Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The physiologic and pharmacologic actions of PGs mediated by distinct G-protein coupled membrane-bound receptors having some degree of selectivity for different classes of PGs have been studied extensively in various in vitro and in vivo models (Coleman et al., 1990, 1994; Mitchell et al., 1994). The characterization of this important family of receptors by molecular pharmacology techniques (i.e., with selective agonists and a more limited number of selective antagonists) (Coleman et al., 1990, 1994) and by molecular biology techniques has revealed intriguing structural similarities and differences among members of the PG receptor family (Narumiya, 1994; Coleman et al., 1994; Abramovitz et al., 1994; Boie et al., 1995; Namba et al., 1994), and has stimulated further investigations into the functional relationships among these receptors.

The FP receptor has generated considerable interest as an ophthalmic therapeutic target because PGF₂ and other FP receptor agonists lower intraocular pressure in humans and primates (Bito, 1997; Wang et al., 1990). The intraocular pressure-lowering actions of PGs may involve the ciliary muscle, and possibly other tissues, in the eye (Yousufzai et al., 1988; Csukas et al., 1993; Goh and Kishino, 1994). Regarding the tissue distribution of the FP receptor, pharmacological studies with the endogenous FP receptor agonist PGF₂ have often been ambiguous due to the relatively low selectivity of PGF₂ for the family of PG receptors and the presence of several classes of PG receptors in many tissue preparations. There is evidence for the FP receptor in uterine smooth muscle (Senior et al., 1993) and on corpus lutea of many species (Davis et al., 1987; Chegini et al., 1991). In ocular tissues, the FP receptor appears to mediate the contractile response of the isolated cat iris sphincter muscle (Goh and Kishino, 1994) and Mukhopadhyay et al., 1997 recently detected the FP receptor mRNA in both human nonpigmented ciliary epithelial cells and human ciliary muscle cells.

Although the FP receptor in isolated tissue preparations has been characterized by various pharmacological methods (Powell et al., 1976; Coleman et al., 1990, 1994; Chen et al., 1995; Woodward et al., 1995; Griffin et al., 1997), and by molecular cloning techniques (Abramovitz et al., 1994; Sugimoto et al., 1994; Sakamoto et al., 1994; Lake et al., 1994; Graves et al., 1995), questions related to species differences in FP receptor structure, function and tissue distribution have not yet been resolved. The FP receptor has been identified in only a few cell types, and its functional properties have not been fully characterized (Woodward et al., 1990; Nakao et al., 1993; Adams et al., 1996; Griffin et al., 1997). With the recent report of a second FP receptor isoform with a truncated carboxy terminus (Pierce et al., 1997), important questions about functional differences between the two known FP receptor isoforms have been raised.

FP receptor mRNA was recently detected in the A7r5 rat thoracic aorta smooth muscle cell line (Adams et al., 1996), but functional properties of the protein expressed by this message were not reported. Therefore, in our study we undertook a detailed characterization of the molecular pharmacology of the FP receptor in the A7r5 cell line using two different functional assays, activation of PLC and mobilization of intracellular calcium. These experiments have demonstrated similarities between FP receptors in the A7r5 cell line and Swiss mouse 3T3 fibroblasts (Griffin et al., 1997), as assayed by PLC activation using a variety of PG agonists. Moreover, the results of this study have suggested that extracellular calcium contributes to the intracellular calcium mobilization response stimulated by FP receptor agonists in the A7r5 cell line.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Cell culture. A7r5 rat vascular smooth muscle cells were grown in DMEM containing 4.5 g/liter glucose and 110 mg/liter sodium pyruvate, which was supplemented with 2 mM L-glutamine, 10 µg/ml gentamicin sulfate and 10% fetal bovine serum. Cells were passaged at 5- to 7-day intervals before attaining confluence by treatment with 0.05% trypsin/0.53 mM EDTA (ethylenediamine tetraacetic acid) and were used in these studies within 10 passages after thawing the frozen cells from the vendor. The agonist stimulation experiments were performed on cells grown to confluence in 24-well uncoated plastic plates.

Phosphoinositide turnover experiments. Previously published procedures were used to measure [³H]-IPs produced by agonist-mediated activation of phospholipase C (Sharif et al., 1994; 1996a; Griffin et al., 1997). Tritium labelling of the phosphatidylinositol pool was achieved by a 24- to 30-hr incubation of cells with 1.0 to 1.5 µCi [³H]-myo-inositol (18.3 Ci/mmol) in 0.5 ml DMEM. After a single rinse with DMEM/F-12 containing 10 mM LiCl, 0.5 ml of the same medium was added to each well and the agonist (or solvent) was then incubated with the cells for 60 min at 37°C, in triplicate experiments. All agonists were typically dissolved in ethanol (final concentration of 1%). Compounds evaluated as potential antagonists were "preincubated" with the cells for no more than 15 min before adding the agonist. The reaction was quenched by aspirating the medium and immediately adding 1 ml of 0.1 M formic acid (kept at 4°C) to lyse the cells. The plates were kept cold and stored frozen for up to 1 wk. For chromatographic separation of radiolabeled components, the thawed cell lysates (0.9 ml) were loaded on columns packed with approximately 1 ml AG 1-X8 anion exchange resin. After washing the columns with 10 ml of H₂O and 8 ml of 50 mM ammonium formate, the total [³H]-IPs fraction was eluted with 4 ml of 1.2 M ammonium formate containing 0.1 M formic acid into a scintillation vial (Berridge et al., 1982). The eluate was mixed with 15 ml of scintillation fluid and the total [³H]-IPs were determined by scintillation counting with a beta-counter. Data were analyzed by the sigmoidal fit function of the Origin Scientific Graphics software (Microcal Software, Northampton, MA) to determine agonist potency (EC₅₀ value) and efficacy, relative to the standard cloprostenol.

Measurement of intracellular calcium concentration. Intracellular calcium concentration was assayed as described (Markwardt et al., 1996; Sharif et al., 1996b). Briefly, cells grown on a coverslip were incubated in a dye-loading buffer (NaCl 125 mM, KCl 5 mM, CaCl₂ 1.8 mM, MgCl₂ 2 mM, NaH₂PO₄ 0.5 mM, NaHCO₃ 5 mM, HEPES 10 mM, glucose 10 mM, bovine serum albumin 0.1%, fura-2 acetoxymethyl ester 5 µM, pH 7.2) for 60 min at room temperature. After rinsing the coverslip with an assay buffer (dye-loading buffer without bovine serum albumin and fura-2 acetoxymethyl ester), the coverslip was mounted in a chamber on the stage of a microscope (Nikon Diaphot, Nikon, Garden City, NY). The chamber was filled with 2 ml of the assay buffer and kept at room temperature during the experiment. Single cell intracellular fluorescence intensities at 510 nm emission wavelength excited by alternating 340 and 380 nm excitation wavelengths were measured by a ratio fluorometer fitted with a photon counting photomultiplier detector (DeltaScan 4000, Photon Technology International, South Brunswick, NJ). Intracellular calcium concentration was calculated from the intensity ratio of fluorescence at these two excitation wavelengths as described (Grynkiewicz et al., 1985). Various drugs in a volume of 20 µl each were added into the chamber as indicated. To remove the test agents, the chamber was washed by completely replacing the assay buffer a minimum of three times. In some experiments, extracellular calcium was eliminated by substituting CaCl₂ with 3 mM EGTA in the assay buffer.

Materials. A7r5 rat vascular smooth muscle cells (CCL-1444) were purchased from the American Type Culture Collection, Rockville, MD. Tissue culture and other reagents purchased from Life Technologies, Grand Island, NY included: DMEM, DMEM/F-12, glutamine, gentamicin and trypsin/EDTA. Fetal bovine serum (HyClone, Logan, UT) was heat-inactivated at 56°C for 30 min and stored at 20°C. Formic acid, ammonium formate and LiCl were supplied by Sigma Chemical Co., St. Louis, MO. Amersham Corp., Deerfield, IL was the source of [³H]-myo-inositol. Fura-2 acetoxymethyl ester was purchased from Calbiochem. AG 1-X8 anion exchange resin was a product of Bio-Rad, Hercules, CA. Ecolume scintillation fluid was supplied by ICN Biomedicals, Costa Mesa, CA. U73122 and U73343 were purchased from Biomol Research Laboratories, Inc., Plymouth Meeting, PA. AVP and the V₁ antagonist [d(CH₂)₅, Tyr(Me)², Tyr(NH₂)⁹]-Arg⁸-vasopressin were purchased from Peninsula Laboratories, Inc, Belmont, CA. All PGs were purchased from Cayman Chemical Co., Ann Arbor, MI or synthesized by published methods.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

PI turnover response. Figure 1 shows the concentration-dependent accumulation of [³H]IPs in A7r5 cells stimulated by several PGs having some degree of selectivity at each of the major PG receptors. The most potent compound evaluated in this cellular assay was cloprostenol (EC₅₀ = 0.84 ± 0.06 nM, n = 34). Under the conditions of the assay, the response to cloprostenol was linear for 75 min and the ratio of the maximal cloprostenol-stimulated response to the basal response (produced by ethanol alone) was typically more than 10. The high potencies of cloprostenol and fluprostenol (EC₅₀ = 4.45 ± 0.19 nM, n = 19) (both full agonists), compared with the other PGs (fig. 1; table 1) indicate that this response is mediated by the FP receptor. The activity of PGF₂ (EC₅₀ = 30.9 ± 2.82 nM, E_max = 86.4% of cloprostenol, n = 7) is also quite consistent with the known activity of this endogenous FP ligand at its own receptor (Davis et al., 1987; Nakao et al., 1993). By contrast, compounds known to have relatively greater activity at other PG receptors had significantly lower potencies and efficacies (relative to cloprostenol): PGD₂ (EC₅₀ = 222 ± 71.4 nM, E_max = 50.9%, n = 5), PGE₂ (EC₅₀ = 2607 ± 270 nM, E_max = 63.1%, n = 3) and the TP receptor ligand U-46619 (EC₅₀ = 5900 ± 1230 nM, E_max = 63.4%, n = 5), or were inactive, e.g., the IP/EP₁ receptor ligand iloprost.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Concentration-dependent stimulation of [3H]IPs formation in A7r5 cells by compounds with known selectivity for different PG receptors. Each plot is representative of several such experiments for each agonist: () cloprostenol; () fluprostenol; () PGF2; () PHXA85; () PGD2; () PGE2; () U46619; and () iloprost. Each point is the average of triplicate determinations in a single experiment, with S.E.M. shown. Data from numerous experiments of this type are shown in table 1.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Correlation of functional activities (PI turnover assay) of a group of PGs evaluated with the A7r5 rat aortic smooth muscle cells and Swiss mouse 3T3 fibroblasts. A, Functional potencies (log EC50): slope, 0.92 and correlation coefficient, 0.98. B, Efficacies (Emax, % relative to cloprostenol standard), slope 1.0 and correlation coefficient, 0.97. Data for all 13 active PGs in table 1 are included in A and B; Iloprost is included only in B. Data for Swiss 3T3 cells taken from Griffin et al. (1997).

2

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Concentration-dependent effects of U73122 and U73343 on [3H]IPs formation in A7r5 cells stimulated by fluprostenol and PGF2. After a 15-min preincubation with various concentrations of U73122, the cells were exposed to agonist for 1 hr: () 107 M fluprostenol and () 5 × 106 M PGF2. Also shown are the control experiments, with 1% ethanol plus each agonist (data points at 1011 M conc.: , fluprostenol; , PGF2) or 105 M U73122 plus 1% ethanol (, both experiments). Statistical significance of U73122 effects compared to the respective control without U73122 (analysis of variance with Bonferroni t test comparison): P < .05 for 105 M and 106 M U73122, with both fluprostenol and PGF2. U73343 was evaluated at 2 × 105 M, in the absence () or presence () of 107 M fluprostenol, in an identical protocol in the same study. The responses to 107 M fluprostenol in the presence () or absence () of U73343 were not significantly different. The data shown are mean response ± S.E.M. for triplicate determinations and are representative of three experiments.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of EGTA on the PI turnover response of A7r5 cells to fluprostenol and PGF2. EGTA was either present at a concentration of 3 mM (, fluprostenol; , PGF2) or absent (, fluprostenol; , PGF2) in the standard assay. The mean responses ± S.E.M. for triplicate determinations are shown. Statistical significance of the effects of EGTA compared to the respective control without EGTA: P < .05 for all concentrations of PGF2 of more than 107 M and all concentrations of fluprostenol of more than 1010 M.

Intracellular calcium mobilization response. The mean calculated resting intracellular calcium concentration of A7r5 cells was 52 ± 2 nM (mean ± S.E.M., n = 27). Exposure of the cells to 1 µM fluprostenol induced a rapid increase of intracellular calcium, which reached a peak value of 347 ± 79 nM (n = 12) within the first 30 sec, followed by a significant elevation that persisted as long as the prostanoid was present in the buffer (fig. 5A). The intracellular calcium level slowly (within 1-3 min) returned to the basal level after the removal of fluprostenol. Interestingly, repeated administration of the agonist resulted in significantly diminished responses (fig. 5B), suggesting rapid desensitization or down-regulation of this response. The diminished response persisted even after a 30-min duration between the successive additions of fluprostenol (fig. 5B).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of fluprostenol on calcium mobilization in A7r5 cells. A, A typical time course response to 1 µM fluprostenol. B, Repeated treatment with fluprostenol induced diminished responses. Flu = fluprostenol.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Effect of calcium-free buffer on the calcium mobilization response of A7r5 cells to fluprostenol. The small, sustained calcium response produced by fluprostenol addition to cells in calcium-free buffer was significantly increased by replacing this buffer with normal assay buffer containing 1.8 mM calcium. Similar results were obtained in three independent studies.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Effects of U73122 on the calcium response of A7r5 cells to fluprostenol. Addition of U73122 subsequent to fluprostenol effectively quenched the calcium mobilization response (compare with fig. 5A).

2

2

2

2

2

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Concentration-dependent mobilization of intracellular calcium in A7r5 cells by fluprostenol and PGF2. A and C, Typical photomultiplier tracings showing responses produced by increasing concentrations of fluprostenol or PGF2. B and D, Dependence of peak calcium response on the cumulative concentration of the respective agonist. Symbols represent mean values and S.E.M.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Effect of sequential addition of fluprostenol and PGF2 on calcium mobilization in A7r5 cells. A, Addition of fluprostenol after a maximally effective concentration of PGF2 increased the calcium response. B, Addition of PGF2 after fluprostenol did not produce any additional calcium mobilization. Similar results were obtained in four independent studies.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The results of this study have demonstrated that the A7r5 rat aortic smooth muscle cell line contains a functional FP receptor. The coupling of this receptor to PLC activation was characterized by evaluating the cellular PI turnover response stimulated by a group of PGs with known selectivity at the FP receptor and other classes of PG receptors. These data demonstrated that PGs previously demonstrated to be relatively potent and selective FP receptor agonists are the most potent and efficacious stimulators of PI turnover in A7r5 cells, with rank order of potency as follows: cloprostenol > 16-phenoxy PGF₂ > 17-phenyl PGF₂ > fluprostenol > PGF₂ > PhXA85. Other PGs had considerably lower potency and efficacy; these included PGD₂, PGE₂, 17-phenyl PGE₂, enprostil, sulprostone, U-46619 and the inactive iloprost. The responses to 17-phenyl PGE₂ (with relative selectivity for the EP₁ receptor) or the TP receptor ligand U-46619 were not inhibited by AH6809 (non-selective antagonist at the EP₁, EP₂ and DP receptors) or by SQ-29,548 (selective TP receptor antagonist), respectively (data not shown). These results appear to rule out any significant role for these receptors in the PG-stimulated activation of PLC in A7r5 cells.

To provide further support for the identity of the PLC coupled receptor in A7r5 cells, we compared agonist potency and efficacy data obtained in this study (table 1) and in our recent study of the FP receptor in the Swiss mouse 3T3 cell line (Griffin et al., 1997). The correlation of both potency and efficacy over a wide range of these parameters for compounds evaluated in both cell types was remarkably high (r  0.97) and provided compelling evidence for functional similarity of the FP receptors in these cells. We previously showed a similar degree of correlation (r = 0.94) between potency in the PI turnover assay in Swiss 3T3 cells and binding affinity at the FP receptor of a bovine corpus luteum membrane preparation for a series of PGs (Griffin et al., 1997), including many tested in our study. Additionally, the published rank order of potencies of selected PGs that cause contraction of the isolated cat iris sphincter muscle (Goh and Kishino, 1994), a response mediated by the FP receptor, correlated well with their potencies in the PI turnover assay (table 1). Collectively, these data demonstrate that the binding affinities of structurally diverse PGs for the FP receptor correlate well with their potencies for activation of the signal transduction cascade coupled to this receptor. Moreover, such data obtained with FP receptors from several different species provide valuable pharmacological evidence for the highly conserved gene sequence and function of the single FP receptor cloned from many species and the major FP receptor isoform of sheep (Abramovitz et al., 1994; Sugimoto et al., 1994; Sakamoto et al., 1994; Lake et al., 1994; Graves et al., 1995).

The potency of AVP (EC₅₀ = 2.69 ± 0.33 nM) determined in this study agrees well with published data on the binding affinity of AVP for A7r5 cell membrane preparations (K_i = 1.3-2.1 nM) (Thibonnier et al., 1991). We confirmed that the response to Arg⁸-vasopressin could be inhibited by the V₁ AVP antagonist [d(CH₂)₅, Tyr(Me)², Tyr(NH₂)⁹]-Arg⁸-vasopressin (Thibonnier et al., 1991), which, however, had no effect on the fluprostenol-stimulated response (unpublished observations). Also, the serotonin-stimulated increase in intracellular calcium concentration in these cells mediated by the 5-HT₂ receptor has a reported potency of 550 nM (Weintraub et al., 1994), similar to our EC₅₀ value of 121 ± 10.7 nM for the PI turnover response. These data further define the selectivity profile of receptors coupled to PLC activation in A7r5 cells.

The mobilization of intracellular calcium in single A7r5 cells by fluprostenol and PGF₂ was shown to behave generally as expected from their activities in the PI turnover assay, namely fluprostenol was both more potent and more efficacious than PGF₂. As observed in other studies of FP-receptor activated mobilization of intracellular calcium (Woodward et al., 1990; Woodward and Lawrence, 1994; Griffin et al., 1997), the FP receptor of A7r5 cells also displays agonist-dependent desensitization of the calcium response. The PLC inhibitor U73122 inhibited fluprostenol-dependent mobilization of intracellular calcium at a concentration (3 µM) quite consistent with its IC₅₀ value (1.25 µM) in the PI turnover assay, suggesting that PI turnover is necessary for the calcium response. Although U73122 pretreatment could completely prevent the intracellular calcium mobilization and [³H]IPs accumulation responses, both responses were also significantly inhibited by depletion of calcium in the assay buffers. Similar effects were observed in a comprehensive study of calcium mobilization in A7r5 cells activated by the PLC-coupled V₁-type AVP receptor (Hughes and Schachter, 1994). In that study, the calcium response was shown to involve intracellular calcium stores and a large contribution from extracellular calcium, and both components were significantly inhibited by U73122 (Hughes and Schacter, 1994). These effects, which have also been observed for other PLC-coupled receptors in various cell types, have been attributed to the PLC-linked formation of an unknown signal(s) that increases the permeability of the cell membrane to calcium (Fasolato et al., 1994). Previous studies of FP receptor functional responses apparently did not characterize FP receptor activated influx of extracellular calcium and thus did not investigate its dependence on PLC activity (Woodward et al., 1990; Nakao et al., 1993; Griffin et al., 1997). Our results have provided the first evidence that PLC-linked influx of extracellular calcium may play a potential role in modulating FP-receptor mediated signaling responses of A7r5 cells and possibly other cell types.

The pharmacological characterization of an FP receptor in the A7r5 rat aortic smooth muscle cell line described herein complements the recent report of the presence of the FP receptor mRNA in this cell line (Adams et al., 1996), and similar unpublished observations that we have made. Our data indicated that A7r5 cells do not contain the TP receptor, because the modest activity of the TP receptor agonist U-46619 was not antagonized by the selective TP receptor antagonist SQ-29,548. However, primary vascular smooth muscle cells contain both the TP receptor and FP receptor, but can be subcultured for a limited time only before loss of their phenotypic and functional characteristics (Dorn et al., 1992). Because the A7r5 vascular smooth muscle cell line has retained the phenotypic features and the FP receptor characteristics of primary smooth muscle cells but apparently has no other PG receptor, coupled to either PLC (this study) or adenylyl cyclase (Senchyna M, Griffin BW, Crankshaw DJ and Sharif NA, unpublished observations), this cell line should prove valuable for further studies of FP receptor function and pharmacology.

In conclusion, our results demonstrate that A7r5 rat vascular smooth muscle cells possess functionally coupled PG receptors of the FP type. The detailed pharmacological studies described herein established that the FP receptor in these cells is coupled to PLC and thus responsible for generating IPs, which subsequently mediates the influx of extracellular calcium. Further studies to link these second messenger responses to other relevant physiological/biochemical functions in these cells are warranted and are currently in progress.