Cyclic AMP-Independent Activation of Neutrophil-like HL-60 Cells by Prostaglandin E2

Cell. Signal. Vol. 9, No. 7, pp. 531–537, 1997 Copyright  1997 Elsevier Science Inc.

ISSN 0898-6568/97 $17.00 PII S0898-6568(97)00047-8

Cyclic AMP-Independent Activation of Neutrophil-like HL-60 Cells by Prostaglandin E2 ZhiHui Xie and Helen Wise* Department of Pharmacology, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, N. T., Hong Kong

ABSTRACT. The role of cAMP in mediating prostaglandin E2 (PGE2)-stimulated aggregation of neutrophillike HL-60 cells has been investigated. Although the EP2 receptors appear to couple to Gs-proteins, PGE2 stimulated HL-60 cell aggregation appears to be a cAMP-independent process. This response to PGE2 is independent of calcium and tyrosine kinase activity, appears to involve activation of phosphatidylinositol 3-kinase which is negatively regulated by phosphatidic acid generated from phospholipase D activity, and is partially dependent on protein kinase C activity. In contrast, although the chemotactic peptide N-formyl-methionyl-leucyl-phenylalanine (FMLP) produces a similar aggregation response to PGE2, FMLP uses a distinct intracellular signalling pathway. The aggregation response to FMLP involves activation of Gi-proteins, is partially dependent on extracellular calcium, is negatively regulated by protein kinase C, and is independent of phosphatidylinositol 3-kinase, phospholipase D and tyrosine kinase activity. The possibility exists that EP2 receptor activation leads to Gs-dependent, but cAMP-independent, stimulation of phosphatidylinositol 3-kinase activity in HL-60 cells. cell signal 9;7:531–537, 1997.  1997 Elsevier Science Inc. KEY WORDS. HL-60 cells, Neutrophil aggregation, Prostaglandin E2, EP2 receptors, cAMP, PI 3-kinase

INTRODUCTION Human promyelocytic leukemic HL-60 cells can be differentiated by dimethyl sulfoxide (DMSO) into neutrophillike cells [1], and are widely used as a model system for studying all aspects of neutrophil activity [2–3]. Although prostaglandins such as PGE2 consistently show inhibitory effects on human neutrophil responses stimulated by chemotaxic agents [4–9], we recently reported a stimulatory response of neutrophil-like HL-60 cells to PGE2 [10]. PGE2 produces a reversible aggregation response in HL-60 cells which, although identical in its aggregometer profile to the response produced by the chemotactic peptide FMLP, differs with regard to its lack of dependency on changes in intracellular calcium. PGE2 mediates this HL-60 cell aggregation response by activation of the EP2 receptor subtype [10]. PGE2 can stimulate four subtypes of EP receptor, known as EP1, EP2, EP3, and EP4 [11, 12], with the EP2 receptor coupling exclusively to stimulation of adenylyl cyclase via a Gs-protein. Therefore in the present study we have examined the role of cAMP in mediating the aggregation response to PGE2, and would suggest that activation of HL-60 cells by PGE2 is independent of cAMP. Despite the initial similarity between the profile of the aggregation response of HL-60 cells to *Author to whom all correspondence should be addressed. E-mail: helen [email protected] Received 13 December 1996; accepted 3 February 1997.

PGE2 and FMLP, a more detailed examination indicates their use of independent intracellular signalling pathways. MATERIALS AND METHODS Materials Rolipram was a gift from Schering AG (Germany). 12-Epi scalaradial, erbstatin, H-89, LY294002, Rp-cAMPS, SpcAMPS and SQ 22536 were purchased from Biomol (Australia). 8-[3H]adenine (specific activity 27 Ci/mmol) and [adenine-U-14C]cAMP (specific activity 278 mCi/mmol) were purchased from Amersham International (U.K.). All other materials were obtained from Sigma (U.S.A.). Methods HL-60 cells were a gift from Dr. K. N. Leung (Chinese University of Hong Kong) and were grown in RPMI 1640 medium containing L-glutamine (2 mM), 10% (v/v) heat-inactivated foetal calf serum, penicillin (100 i.u./ml) and streptomycin (100 mg/ml) at 378C in 5% CO2, 95% air. HL-60 cells were differentiated to granulocytes by incubation for 5–7 d in medium containing 1.3% (v/v) DMSO as described previously [10]. Differentiated HL-60 cells were washed once in Hepes-buffered saline (HBS: NaCl, 145 mM; KCl, 5 mM; MgCl2, 1 mM; glucose, 10 mM; Hepes acid, 10 mM; pH 7.55) and then resuspended in HBS supplemented with indomethacin (3 mM) 30 min before use. CELL CULTURE.

532

HL-60 cell aggregation was measured with a Chrono-log platelet aggregometer as described previously [10]. The cell suspension (1 3 107 cells/ ml) was incubated at 378C in siliconized cuvettes in incubation buffer (HBS supplemented with indomethacin (3 mM) and CaCl2 (1 mM)) for time periods from 10–30 min. For incubation periods of 4 h or longer, cells were resuspended in fresh cell culture medium containing indomethacin (3 mM) and were then resuspended in incubation buffer immediately prior to transfer into the aggregometer. After a 3 min equilibration period in the aggregometer, PGE2 or FMLP was added to the stirred cell suspension and the maximal change in light transmission recorded in arbitrary units (mm deflection of pen recorder). HL-60 CELL AGGREGATION.

[3H]cAMP ACCUMULATION. Cyclic AMP production was assayed by measuring the conversion of [3H]ATP to [3H]cAMP as described previously [13]. Briefly, HL-60 cells were incubated with [3H]adenine (10 mCi/ml) for 60 min at 378C in buffer (HBS supplemented with glucose (10 mM), CaCl2 (1 mM) and bovine serum albumin (1%, w/v)) at a cell density of 3 3 107 cells/ml. Labelled cells were washed and incubated in triplicate at 1 3 107 cells/ml in buffer containing rolipram (100 mM) and indomethacin (3 mM). After 5 min incubation at 378C, test drugs were added for a further 10 min when the reaction was terminated by the addition of ice-cold 7.5% trichloroacetic acid. Results are expressed relative to the basal activity represented by [3H]cAMP generated in the absence of exogenous stimuli. DATA ANALYSIS. EC50 and IC50 values were calculated using the curve fitting program ALLFIT [14]. Results are expressed as means with S.E.M. of (n) independent experiments. Statistical analysis was performed using an unpaired Student’s t-test.

RESULTS AND DISCUSSION The Role of cAMP in PGE2-stimulated HL-60 Cell Aggregation Having previously identified the EP2 receptor as responsible for the PGE2-mediated activation of HL-60 cells [10], and since the EP2 receptor subtype is coupled exclusively to a Gs-protein [11, 12], we attempted to establish a role for cAMP in mediating the response to PGE2. In the presence of the type IV phosphodiesterase (PDE) inhibitor rolipram (100 mM), PGE2 produced a concentration-dependent increase in [3H]cAMP accumulation with a 5-fold stimulation at 10 mM (Fig. 1). However, PGE2 is able to stimulate an aggregation response at concentrations which have little effect on cAMP levels. Thus the EC50 value for stimulation of adenylyl cyclase activity is at least 100-fold higher than that for stimulation of the aggregation response where PGE2 has an EC50 value of 38 6 13 nM (n 5 5, Fig. 1). In contrast to PGE2, FMLP (10 mM) caused no significant stimulation of [3H]cAMP accumulation (103 6 4% control, n 5 3).

Z. Xie and H. Wise

If PGE2-stimulated [3H]cAMP accumulation is responsible for the aggregation response, then cAMP mimetics should also cause HL-60 cell aggregation. However, the cell permeable analogues of cAMP such as dibutyryl cAMP (1 mM), 8-bromocAMP (100 mM) and Sp-cAMPS (100 mM) all failed to stimulate HL-60 cell aggregation (n 5 3). The adenylyl cyclase activator forskolin (100 mM) also had no effect on HL-60 cell aggregation (n 5 3) despite increasing [3H]cAMP accumulation to a level (136 6 5% control, n 5 3) similar to that produced by 100 nM PGE2. Although forskolin is also a weak activator of adenylyl cyclase in rat peritoneal neutrophils [13], it is still capable of inhibiting FMLP-stimulated aggregation in those cells [15]. Human neutrophils contain predominantly the type IV PDE [16], and any response mediated by cAMP would be expected to be potentiated by inhibition of this PDE activity. However, 10 min pre-incubation of HL-60 cells with the type IV PDE inhibitors rolipram (10 mM) or Ro-20-1724 (10 mM), or with the non-selective PDE inhibitor 3-isobutyl-1-methyl xanthine (IBMX; 100 mM) had no effect on the concentration-response curve to PGE2 (n 5 5, Fig. 2). Furthermore, 30 min pre-incubation of HL-60 cells with an inhibitor of protein kinase A activity, such as Rp-cAMPS (10 mM), failed to attenuate the concentration-response curve to PGE2 (n 5 5, Fig. 3). Another protein kinase inhibitor H-89 (10 mM) and the adenylyl cyclase inhibitor SQ 22536 (100 mM) also failed to attenuate the aggregation response to PGE2 (n 5 3, Fig. 3). Taken together, the above results suggest a lack of involvement of cAMP in mediating the aggregation response to PGE2. Although PGE2 can also stimulate adenylyl cyclase activity in human neutrophils [8, 17] and rat neutrophils [13], not all PGE2-mediated responses in neutrophils are dependent on cAMP. For example, it has been known for some time that there is a poor correlation between receptoractivated adenylyl cyclase activity and inhibition of neutrophil function [18]. Indeed it has recently been shown that PGE2-mediated inhibition of FMLP-stimulated chemotaxis in human neutrophils is independent of cAMP production [8]. Less clearcut is the dependency on cAMP production for PGE2-mediated inhibition of FMLP-stimulated aggregation of rat neutrophils [19]. In contrast, PGE2-mediated inhibition of FMLP-stimulated superoxide anion production by human neutrophils is clearly dependent on cAMP production [9]. The results of the present study are similar to those described for inhibition of FMLP-stimulated chemotaxis [8], and suggest that PGE2-stimulated aggregation of HL-60 cells is independent of cAMP production. In order to help identify the G-protein coupling the EP2 receptor to the cellular response, HL-60 cells were incubated with cholera toxin or pertussis toxin, as described in Table 1. Pertussis toxin catalyses the ADP-ribosylation of Gi-proteins, whereas cholera toxin catalyses the ADP-ribosylation of Gs-proteins [20]; both toxins will uncouple the respective G-proteins from activation by membrane-bound receptors. Pertussis toxin significantly inhibited FMLP-

EP2 Receptor Coupling in HL-60 Cells

533

FIGURE 1. The ability of PGE2

to stimulate HL-60 cell aggregation and [3H]cAMP accumulation. The aggregation response of HL-60 cells (m) and [3H]cAMP accumulation (d) was measured as described in the Materials and Methods section. Results are expressed as means 6 S.E.M. of five separate experiments for aggregation, and of three experiments, each performed in triplicate, for [3H]cAMP accumulation. The aggregation results are shown as a percentage of the maximum response to PGE2 which was 7.0 6 0.5 mm. PGE2-stimulated [3H]cAMP accumulation is shown as a percentage of basal [3H]cAMP production calculated as percentage conversion of [3H]ATP to [3H]cAMP (0.117 6 0.022%).

stimulated HL-60 cell aggregation (Table 1), confirming the role for Gi-proteins in mediating many of the neutrophil responses to FMLP [21–23]. In contrast, pertussis toxin had no effect on PGE2-mediated HL-60 aggregation (Table 1), thus excluding a role for Gi-protein coupling to the EP2 receptor. Cholera toxin had no significant effect on FMLPmediated HL-60 cell aggregation but abolished the response to PGE2 (Table 1). Cholera toxin did not cause HL-60 cell aggregation, giving further support to the idea that aggregation is a cAMP-independent process. As cholera toxin can uncouple Gs from its receptors [24, 25], then the failure of PGE2 to activate cholera toxin-pretreated HL-60 cells suggests that the EP2 receptor is coupled to a Gs-protein.

Therefore as the EP2 receptors in HL-60 cells appear to be coupled to Gs, yet PGE2-mediated HL-60 cell aggregation is independent of cAMP production, we attempted to identify an alternative signalling pathway for EP2 receptors. Although not observed in our HL-60 cells, it has been shown that DMSO-differentiated HL-60 cells contain two adenylyl cyclase stimulatory receptors coupled to [Ca21]i mobilization [26]. In these cells, histamine and PGE2 increased cAMP and both released calcium from intracellular stores by a cholera toxin-sensitive pathway. Moreover, a detailed study of the histamine H2 receptor-mediated mobilization of [Ca21]i indicated that this process was independent of cAMP. More recently it has been shown that, in chick myo-

FIGURE 2. The effect of phosphodiesterase inhibitors on PGE2-stimulated HL-60 cell aggregation. Cells were pre-incubated with con-

trol buffer (s), rolipram (10 mM, d), Ro-20-1724 (10 mM, j) or IBMX (100 mM, m) for 10 min at 378C. The aggregation response to PGE2 was then measured. Results are expressed as means 6 S.E.M. of five separate experiments. The results are shown as a percentage of the control response to PGE2 (10 mM) which was 3.8 6 0.6 mm.

534

Z. Xie and H. Wise

FIGURE 3. The effect of cAMP antagonists on PGE2-stimulated HL-60 cell aggregation. Cells were pre-incubated with control buffer

(s), Rp-cAMPS (10 mM, d), H-89 (10 mM, j) or SQ 22536 (100 mM, m) for 30 min at 378C. The aggregation response to PGE2 was then measured. Results are expressed as means 6 S.E.M. of five separate experiments for Rp-cAMPS, and three experiments for H-89 and SQ 22536. The results are shown as a percentage of the control response to PGE2 (10 mM) which was 3.5 6 0.4 mm.

cytes, exogenous Gsa can couple the adenosine A2a receptor to a cAMP-independent stimulation of the calcium channel [27]. However, it remains possible that there may also exist an effector role for the bg subunits of Gs, in addition to the a subunits, in DMSO-differentiated HL-60 cells. Since the recognition in 1987 that bg subunits could directly activate the muscarinic K1 channel in chick heart [28], there have been many reports describing an effector role for bg subunits [29, 30], but all of these studies have used bg subunits from Gi, Go or Gt-protein. Further studies are therefore needed to distinguish between a cAMP-inde-

pendent function of Gsa, and a possible effector role for bg subunits from Gs in HL-60 cells. The Role of Protein Kinase C in HL-60 Cell Aggregation Neutrophil-like HL-60 cells contain a phospholipase C (PLC) isoform capable of being stimulated by bg subunits purified from bovine retina (bgt) and bovine brain [23]. Previous results suggest though that PGE2 does not activate PLC in HL-60 cells because, in contrast to FMLP, PGE2 produces no increase in intracellular calcium and the aggregation re-

TABLE 1. Effect of drug pretreatments on HL-60 cell aggregation responses to PGE2 and FMLP

Response (% control) PGE2 (10 mM)

FMLP (10 mM)

Control (4 h) Cholera toxin (150 ng/ml, 4 h) Pertussis toxin (100 ng/ml, 4 h)

100 6 18 0 6 0*** 104 6 7

100 6 12 78 6 14 10 6 4***

Control (10 min) Staurosporine (1 mM, 10 min)

100 6 7 42 6 11**

100 6 6 169 6 29*

Control (10 min) 12-epi scalaradial (1 mM, 10 min) p-BPB (100 mM, 10 min)

100 6 13 99 6 13 107 6 8

100 6 15 113 6 15 84 6 10

Control (10 min) Ethanol (1% v/v, 10 min) Propranolol (200 mM, 10 min)

100 6 9 79 6 6 5 6 5***

100 6 15 97 6 9 103 6 9

Control (10 min) LY294002 (100 mM, 10 min) Wortmannin (1 mM, 10 min)

100 6 6 38 6 13** 10 6 5***

100 6 7 106 6 4 77 6 11

Control (30 min) Erbstatin (50 mM, 30 min)

100 6 11 117 6 8

100 6 7 87 6 4

Pretreatment

HL-60 cell aggregation was measured as described in the Materials and Methods section using 1 3 107 cells/ ml of incubation buffer. The results are presented as means 6 S.E.M. derived from 5 separate experiments. Control aggregation responses in these experiments ranged between 4 and 6 mm. *P , 0.05, **P , 0.01, ***P , 0.001 compared to the respective control response.

EP2 Receptor Coupling in HL-60 Cells

535

FIGURE 4. The effect of wort-

mannin on PGE2-stimulated HL60 cell aggregation. Cells were pre-incubated with various concentrations of wortmannin, for 10 min at 378C, before testing the aggregation response to PGE2. Results are expressed as means 6 S.E.M. of five separate experiments, each performed in duplicate. The results are shown as a percentage of the control response to PGE2 (10 mM) which was 3.7 6 0.5 mm.

sponse to PGE2 is independent of extracellular calcium [10]. However, we have shown here that phorbol 12-myristate 13-acetate (PMA; 100 nM) can produce a large, irreversible aggregation response in HL-60 cells (51 6 5 mm, n 5 4) with a time lag of about 30–60 s. This response compares with the more rapid, reversible effect of both PGE2 and FMLP which typically produce a maximum response of 7–10 mm. Therefore we have examined the role of protein kinase C (PKC) in the aggregation response of HL-60 cells. The PKC inhibitor staurosporine, at a concentration which abolished the aggregation response to 100 nM PMA (data not shown), significantly inhibited PGE2-stimulated HL-60 cell aggregation (Table 1), suggesting that PKC mediates at least part of the aggregation response to PGE2. In contrast, staurosporine significantly enhanced the aggregation response to FMLP (Table 1) suggesting that PKC plays an inhibitory feedback role in FMLP-mediated HL-60 cell aggregation.

The Role of Phosphatidylinositol 3-kinase (PI 3-kinase) in HL-60 Cell Aggregation A range of chemoattractants such as FMLP, platelet activating factor and leukotriene B4 can stimulate PI 3-kinase activity in human neutrophils [31] and inhibition of PI 3-kinase attenuates FMLP-stimulated superoxide anion production in human neutrophils [32, 33] and guinea pig neutrophils [34], and FMLP-stimulated aggregation of rat neutrophils [19]. Although PI 3-kinase is involved in mediating responses following activation of tyrosine kinase-receptors [35], incubation of HL-60 cells with the tyrosine kinase inhibitor

erbstatin had no effect on either PGE2 or FMLP-mediated aggregation (Table 1). Interestingly though, PI 3-kinase can also be activated by G-protein-coupled receptors in neutrophils [31]. In the present study we found that the PI 3-kinase inhibitors wortmannin and LY294002 significantly inhibited PGE2-mediated HL-60 cell aggregation, but had no significant effect on the response to FMLP (Table 1). Wortmannin inhibited PGE2-mediated HL-60 cell aggregation in a concentrationdependent manner with an IC50 value of 57 nM (Fig. 4). Although wortmannin also inhibits phospholipase D (PLD), PLC and phospholipase A2 (PLA2) [36], we are confident that the target for wortmannin in our assay is PI 3-kinase as PGE2 does not appear directly to activate PLC (see section above) or PLD (see following section). PGE2-stimulated PLA2 activity would be expected to liberate arachidonic acid which could be converted by cyclooxygenase to prostanoids, or by lipooxygenase to leukotrienes. As our assays are performed in the presence of indomethacin, we can exclude the possibility of prostanoid production. Products of the lipooxygenase pathway such as leukotriene B4 are chemotaxic for neutrophils, and it is possible that wortmannin is preventing the production of these compounds. However, this seems unlikely as we found that two PLA2 inhibitors, 12-epi scalaradial and p-bromophenacyl bromide (pBPB), had no significant effect on either PGE2 or FMLP-mediated aggregation (Table 1). Therefore the ability of wortmannin, and the more selective inhibitor of PI 3-kinase, LY294002, to markedly inhibit PGE2-mediated HL-60 cell aggregation indicates a prime role for PI 3-kinase in the intracellular signal transduction pathway.

536

Although not proven here, it is possible that EP2 receptor activation may lead to the release of bg subunits from Gs which then directly activate PI 3-kinase, resulting in calcium-independent (and cAMP-independent) HL-60 cell aggregation. In contrast, FMLP receptor stimulation leads to activation of a Gi-dependent, extracellular calciumdependent, PI 3-kinase-independent pathway in HL-60 cells. It is interesting to note that there are other reports of both calcium-independent wortmannin-sensitive, and calcium-dependent activation of neutrophils. For example in guinea pig neutrophils, platelet-activating factor activates mitogen-activated protein kinase through two distinct pathways; one is calcium-dependent, whereas the other is calcium-independent and wortmannin-sensitive [37]. Further studies are required to prove the direct activation of PI 3-kinase following EP2 receptor stimulation in HL-60 cells. The Role of PLD in HL-60 Cell Aggregation As we have shown that HL-60 cells can aggregate in response to PMA, yet PGE2-mediated HL-60 cell aggregation is independent of PLC, we investigated the possible role of PLD as the source of diacylglycerol (DAG) in these cells. In the absence of a specific inhibitor of PLD, we have used a high concentration of propranolol (200 mM) to inhibit phosphatidate phosphatase activity, and ethanol (1%, v/v) to promote the formation of phosphatidylethanol from phosphatidic acid (PA). Propranolol increases PA and decreases DAG in FMLP-stimulated human neutrophils, whereas ethanol decreases both PA and DAG [38]. Although ethanol had no significant effect on either PGE2 or FMLP-mediated HL-60 cell aggregation (Table 1), propranolol dramatically inhibited the response to PGE2, leaving the FMLP response unaffected. If propranolol increases PA levels in HL-60 cells and inhibits PGE2-stimulated HL-60 cell aggregation, then it is possible that PA functions as an inhibitory regulator of the aggregation response following EP2 receptor activation. Such speculation is supported by the recent observation that PA can directly inhibit PI 3-kinase purified from rabbit platelets [39]. It is possible therefore that PLD is activated by PI 3-kinase leading to the production of both PA and DAG. The DAG proceeds to activate PKC with the consequent production of a staurosporine-sensitive aggregation response. Accumulation of PA will lead to feedback inhibition of PI 3-kinase activity and inhibition of the aggregation response. As both wortmannin and propranolol can abolish PGE2-mediated HL-60 cell aggregation, yet staurosporine produces only 58% inhibition, then it is likely that PI 3-kinase is upstream of PLD and PKC and controls two independent pathways. Therefore to summarize our findings, FMLP stimulates HL-60 cell aggregation via a signalling pathway involving a pertussis toxin-sensitive Gi-protein. This response to FMLP is partially dependent on extracellular calcium [10], is negatively regulated by PKC, and is independent of PI 3-kinase, PLD and tyrosine kinase activity. In contrast, stimulation of HL-60

Z. Xie and H. Wise

cell aggregation by activation of EP2 receptors involves a signalling pathway coupled to a Gs-protein, but appears to be independent of the generation of cAMP. This response to PGE2 is independent of calcium [10] and tyrosine kinase activity, appears to involve activation of PI 3-kinase which is negatively regulated by PA generated from PLD activity, and is partially dependent on PKC activity. Further studies are now clearly needed to demonstrate if EP2 receptor activation can stimulate PI 3-kinase activity and to determine whether this response is mediated by Gsa or Gsbg. This study was supported by the Research Grants Council of Hong Kong (CUHK24/93M).

References 1. Collins S. J., Ruscetti F. W., Gallagher R. E. and Gallo R. C. (1978) Proc. Natl. Acad. Sci. USA 75, 2458–2462. 2. Collins S. J., Ruscetti F. W., Gallagher R. E. and Gallo R. C. (1979) J. Exp. Med. 149, 969–974. 3. Harris P. and Ralph P. (1985) J. Leukoc. Biol. 37, 407–422. 4. Ney P. and Schro¨r K. (1989) Br. J. Pharmacol. 96, 186–192. 5. Gryglewski R. J. and Salvemini D. (1992) In: Prostacyclin: New Perspectives for Basic Research and Novel Therapeutic Indications (Rubanyi G. M. and Vane J., Eds), pp. 59–92. Elsevier Science Publishers, Amsterdam. 6. Nagata S., Kebo D. K., Kunkel S. and Glovsky M. M. (1992) Int. Arch. Allergy Immunol. 97, 194–199. 7. Wheeldon A. and Vardey C. J. (1993) Br. J. Pharmacol. 108, 1051–1054. 8. Armstrong R. A. (1995) Br. J. Pharmacol. 116, 2903–2908. 9. Talpain E., Armstrong R. A., Coleman R. A. and Vardey C. J. (1995) Br. J. Pharmacol. 114, 1459–1465. 10. Wise H. and Xie Z. H. (1996) Prostaglandins Leukoc. Essent. Fatty Acids 54, 199–205. 11. Coleman R. A., Smith W. L. and Narumiya S. (1994) Pharmacol. Rev. 46, 205–229. 12. Negishi M., Sugimoto Y. and Ichikawa A. (1995) J. Lipid Mediators Cell. Signalling 12, 379–391. 13. Wise H. and Jones R. L. (1994) Br. J. Pharmacol. 113, 581– 587. 14. De Lean A., Munson P. J. and Rodbard D. (1978) Am. J. Physiol. 235, E97–E102. 15. Wise H. (1996) Prostaglandins Leukot. Essent. Fatty Acids 54, 351–360. 16. Nielson C. P., Vestal R. E., Sturm R. J. and Heaslip R. (1990) J. Allergy Clin. Immunol. 86, 801–808. 17. Hecker G., Ney P. and Schro¨r K. (1990) Naunyn-Schmiedeberg’s Arch. Pharmacol. 341, 308–315. 18. Harvath L., Robbins J. D., Russell A. A. and Seamon K. B. (1991) J. Immunol. 146, 224–232. 19. Wise H. (1996) J. Leukoc. Biol. 60, 480–486. 20. Chang F.-H. and Bourne H. R. (1989) J. Biol. Chem. 264, 5352–5357. 21. Krause K. H., Schlegel W., Wollheim C. B., Anderson T., Waldvogel F. A. and Lew D. (1985) J. Clin. Invest. 76, 1348– 1354. 22. Gierschik P., Sidiropoulos D. and Jakobs K. H. (1989) J. Biol. Chem. 264, 21470–21473. 23. Camps M., Hou C., Sidiropoulos D., Stock J. B., Jakobs K. H. and Gierschik P. (1992) Eur. J. Biochem. 206, 821–831. 24. Milligan G., Unson C. G. and Wakelam M. J. O. (1989) Biochem. J. 262, 643–649. 25. Williams R. J. and Kelly E. (1994) Eur. J. Pharmacol. 268, 177–186.

EP2 Receptor Coupling in HL-60 Cells 26. Mitsuhashi M., Mitsuhashi T. and Payan D. G. (1989) J. Biol. Chem. 264, 18356–18362. 27. Liang B. T. and Morley J. F. (1996) J. Biol. Chem. 271, 18678–18685. 28. Logethetis D. E., Kurachi Y., Galper J. and Neer E. J. (1987) Nature 325, 321–326. 29. Simon M. I., Strathmann M. P. and Gautam N. (1991) Science 252, 802–808. 30. Clapham D. E. and Neer E. J. (1993) Nature 365, 403–406. 31. Stephens L., Eguinoa A., Corey S., Jackson T. and Hawkins P. T. (1993) EMBO J. 12, 2265–2273. 32. Thelen M., Wymann M. P. and Langen H. (1994) Proc. Natl. Acad. Sci. USA 91, 4960–4964. 33. Vlahos C. J., Matter W. F., Brown R. F., Traynor-Kaplan A. E., Heyworth P. G., Prossnitz E. R., Ye R. D., Marder P.,

537

34. 35. 36. 37. 38. 39.

Schelm J. A., Rothfuss K. J., Serlin B. S. and Simpson P. J. (1995) J. Immunol. 154, 2413–2422. Okada T., Sakuma L., Fukui Y., Hazeki O. and Ui M. (1994) J. Biol. Chem. 269, 3563–3567. Fry M. J. (1994) Biochim. Biophys. Acta 1226, 237–268. Cross M. J., Stewart A., Hodgkin M. N., Kerr D. J. and Wakelam M. J. O. (1995) J. Biol. Chem. 270, 25352–25355. Ferby I. M., Waga I., Sakanaka C., Kume K. and Shimizu T. (1994) J. Biol. Chem. 48, 30485–30488. Yasui K., Yamazaki M., Miyabayashi M., Tsuno T. and Komiyama A. (1994) J. Immunol. 152, 5922–5929. Lauener R., Shen Y., Duronio V. and Salari H. (1995) Biochem. Biophys. Res. Commun. 215, 8–14.