Quantitative and Qualitative Differences in Guanine Nucleotide Binding Protein Activation by Formyl Peptide and Leukotriene B4 Receptors*

Formyl peptides and leukotriene B4 (LTB,) stimulate disparate neutrophil functional responses and second messenger generation. The hypothesis that differences in receptor-guanine nucleotide-binding proteins (G protein) interaction account for the disparate re- sponses was examined using HL-60 granulocyte plasma membranes. The quantity of receptor-coupled G proteins was determined by guanosine 5’-(-y- thio)triphosphate (GTP-yS) equilibrium binding in the presence or absence of f-Met-Leu-Phe and/or LTB,. About one-third of the total GTPyS binding sites were coupled to f-Met-Leu-Phe receptors, to LTB, recep- tors, and to receptors when both ligands were added simultaneously. The dissociation constant of GTP-yS- binding sites in the presence of LTB, was significantly greater than that in the presence of f-Met-Leu-Phe. f- Met-Leu-Phe shifted the GDP dose-inhibition curve for GTPyS binding further to the right than did LTB,. The apparent initial rate of GTP hydrolysis and GTP-yS binding stimulated by f-Met-Leu-Phe was significantly greater than that stimulated by LTB,. There were significantly more formyl peptide receptors than LTBI receptors, however, formyl peptide and LTB, receptor density did not differ under GTPyS binding assay conditions. The rate of GTP hydrolysis stimulated by LTB, was not increased in membranes containing twice

Chemoattractants are a heterogeneous group of agonists which stimulate directed migration and release of lysosomal enzymes and reactive oxygen metabolites by neutrophils. Peptide chemoattractants consist of formylated peptides, C5a, and interleukin 8, while lipid chemoattractants include leukotriene B4 (LTB,) and platelet-activating factor. Although all chemoattractants stimulate directed migration, they differ in their ability to stimulate release reactions. Formylated peptides stimulate a vigorous release of lysosomal enzymes and reactive oxygen metabolites. On the other hand, LTB, stimulates weak and short-lived release reactions (7,(24)(25)(26)(27)(28)(29). The disparate functional responses are associated with differences in second messenger generation. Second messenger responses, including the calcium transient, diacyglycerol generation, phosphatidic acid formation, and inositol trisphosphate generation, are more transient and reduced in magnitude following LTB, stimulation compared to the responses to formyl peptides (25,26,(29)(30)(31)(32)(33). The molecular basis for these different functional and biochemical responses to formyl peptides and LTB, remains unknown.
The ability of formyl peptide and LTB, receptors to stimulate formation of the same second messengers, but with different kinetics, suggested to us that differences in receptor-G protein interaction are the most likely basis for the disparate biochemical and cellular responses. The present work was initiated to examine three hypotheses which could explain the ability of formyl peptides and LTB, to induce different functional and biochemical responses. First, formyl peptide and LTB, receptors could be coupled to different G proteins, which have different capabilities of activating the same effector enzymes. Second, formyl peptide and LTB, receptors could differ quantitatively in their activation of the same G proteins. Third, formyl peptide and LTB, receptors could differ qualitatively in their activation of common G proteins leading to disparate signal transmission to effector enzymes. Our results indicate that formyl peptide receptors activate greater quantities of common G proteins than LTB, receptors. These quantitative differences result from differences in receptor density and/or qualitative differences in receptor-G protein interactions identified by different affinity states for guanine nucleotides.

G Protein Activation by Chemoattractant Receptors
[-y-32P]GTP was from ICN Biomedicals, Inc. (Irvine, CA). Cell Culture-HL-60 cells, obtained from American Type Culture Collection (Rockville, MD), were grown in suspension culture in RPMI 1640 media supplemented with 10% (v/v) horse serum, 1% (v/ v) nonessential amino acids, 1 mM L-glutamine, 50 units/ml of penicillin and 50 pg/ml streptomycin. Cells were grown in a humidified atmosphere with 8% CO, at 37 "C. To induce myeloid differentiation, cells were seeded at a density of lo6 cells/ml and cultivated for 5-6 days in medium containing 1.25% (v/v) dimethyl sulfoxide (34).
Membrane Preparation-HL-60 cell membranes were prepared as described by Gierschik et al. (35), except that centrifugation through a Percoll gradient was omitted. In brief, HL-60 cells were centrifuged for 30 min at 3000 rpm. All centrifugations and subsequent procedures were performed at 4 "C. The cell pellet was resuspended in a buffer (30 ml) containing 10 mM triethanolamine-HC1, pH 7.4, and 140 mM NaCl and centrifuged at 4000 rpm for 30 min. Cells were resuspended in a buffer (30 ml) containing 250 mM sucrose, 20 mM Tris-HCI, pH 7.5, 1.5 mM MgC12, 1 pM leupeptin, 1 mM ATP, 1 mM phenylmethylsulfonyl fluoride, 3 mM benzamidine, and 2 pg/ml soybean trypsin inhibitor. The resuspended cells were placed in a nitrogen cavitation bomb for 30 min at 25 bar (375 psi). Membranes were collected into 375 pl of 100 mM EGTA, centrifuged at 4,000 rpm for 45-60 s, and then passed through a double layer of cheesecloth to remove nuclei. Subsequently, membrane preparations were centrifuged for 15 min at 15,000 rpm, and the pellets were resuspended in buffer containing 20 mM Tris-HC1, pH 7.5, 1 mM EDTA, 1 mM dithiothreitol, 3 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, 10 p~ leupeptin, and 2 pg/ml soybean trypsin inhibitor. Membranes were centrifuged and resuspended three times before being resuspended in 1-2 ml of buffer, and 100-pl aliquots were snap frozen in liquid nitrogen. Membrane preparations were stored at -70 "C. Initial rates of GTP-yS binding were determined by removal of 50pl aliquots from a 350-pl reaction mixture after incubation for 20, 40, 60, 80, 100, and 120 s, filtered through Whatman GF/C filters, and washed three times with 2.5 ml of buffer containing 50 mM Tris-HC1, pH 7.5, and 5 mM MgCI,. Specific binding was determined as above. Apparent initial rate of binding was calculated by obtaining the slope from the linear portion of each curve and expressed as femtomoles [35S]GTPyS bound/milligram membrane protein/minute. G Protein Content-The quantity of G proteins coupled to each receptor was determined using a modification of the [35S]GTP-yS binding assay, as described by Gierschik et al. (36). Reactions were performed in the presence of increasing concentrations of GTP-yS and in the presence or absence of M f-Met-Leu-Phe, 10"j M LTB4, and 0.5 p~ GDP. The reaction was allowed to proceed for 60 min at 30 "C before filtering and counting.

GTP-yS
GTPase Assay-Hydrolysis of [-y-32P]GTP was determined in a reaction mixture (100 pl) containing 50 mM triethanolamine-HC1, pH 7.4, 1 mM dithiothreitol, 1 mM EDTA, 5 mM MgCl,, 100 p M ATP, 100 mM NaCl, 0.4 mg/ml creatine kinase, 5 mM creatine phosphate, 2 mg/ml bovine serum albumin, 100 nM GTP, and 0.02-0.05 pCi/ tube of [-y-32P]GTP. The reaction was initiated by the addition of 2-10 pg of membrane protein to each tube. The reaction was carried out at 30 'C for 10 min, and terminated by addition of 700 pl of icecold activated charcoal as described (37). Assay tubes were centrifuged for 30 min at 4 "C at 2000 rpm. 500 pl of supernatant were counted to determine the release of 3zPi. High affinity GTPase activity was determined after subtracting the amount of '*Pi released in the presence of 50 p~ GTP from the total amount of ''PI released. Results were expressed as picomoles of P, released/minute/milligram of membrane protein.
Initial rates of hydrolysis of [y-"P]GTP were determined after 1-, 2-, 3-, 4-, and 5-min incubations at 30 "C. Rate of hydrolysis was calculated from the slope of the linear portions of the curve and expressed as pmol of free Pi released/minute/milligram of membrane protein.
Receptor Binding Assays-f-Met-Leu-Phe and LTB, binding assays were performed in a reaction mixture ( Protein Assay-Membrane protein was determined using the method described by Bradford (39), with bovine IgG as the standard.
Statistical Analysis-Data were analyzed by Student's t test or by one-way analysis of variance and Student-Newman-Keuls post-test using the SPSS Statistical Package (SPSS Inc. Chicago, IL). The 95% confidence limits were taken a priori as statistically significant.

RESULTS
We have previously reported evidence that formyl peptide and LTB4 receptors share some common G proteins (7). To estimate the size of the G protein pool accessible to formyl peptide and LTB, receptors, the effect of f-Met-Leu-Phe (low5 M) and LTB, M) was examined in a GTPyS saturation experiment performed in the presence of 0.5 ~L M GDP. GTPyS binding was also determined in the absence of GDP. Preliminary studies determined that f-Met-Leu-Phe-and LTB,stimulated GTPyS binding reached equilibrium by 60 min, as previously described (36). Transformation of the binding data in the absence of GDP according to Scatchard resulted in a linear plot, indicating a single high affinity for GTPyS binding sites (Fig. 1). Addition of 0.5 p~ GDP reduced the apparent affinity of GTPyS-binding sites but did not alter their density. In the presence of GDP and lod5 M f-Met-Leu-Phe or 1O"j M LTB,, the plots were curvilinear, indicating two distinct classes of binding sites, as previously described (36). Agonist activation resulted in an increase in apparent GTPyS affinity of a portion of the GTPyS-binding sites. Thus, in the presence of GDP G proteins coupled to formyl peptide or LTB, receptors demonstrated a higher affinity for GTPyS, while the affinity for GTPyS of G proteins not coupled to these receptors was not affected. The higher affinity portion of this biphasic curve was used to calculate the density of receptor-coupled G proteins for formyl peptide and LTB4 receptors and their affinities for GTPyS in the presence of 0.5 pM GDP. Fig. 2 shows a portion of a Scatchard plot of GTPySbinding data demonstrating only the higher affinity GTPySbinding sites found in the presence of 0.5 p~ GDP and M f-Met-Leu-Phe, M LTB4, or both ligands, following subtraction of GTPyS binding to the lower affinity binding sites. Ligand concentrations were at least 1000 times greater than the dissociation constants of their respective receptors (see Table 111). In this single experiment, the density of Binding data obtained in the absence of agonist or GDP is labeled control. In the absence of GDP a single, high affinity for GTPyS is seen, while the affinity is reduced in the presence of GDP. Addition of GDP and f-Met-Leu-Phe or LTB, results in a biphasic curve in which a two-site model provides a significantly better fit. From the high affinity portion of the biphasic curve, the density and affinity of receptor-coupled GTPyS-binding sites are obtained (see Fig. 2). receptor-coupled G proteins is similar, while the affinity for GTPyS is lower in the presence of LTB,. From 10 separate experiments in three different membrane preparations the density of total GTPyS-binding sites, the density of GTPySbinding sites coupled to each receptor, and their affinity for GTPyS were calculated (Table I). The density of G proteins coupled to formyl peptide receptors, LTB, receptors, or both receptors activated simultaneously was not significantly different. The G proteins coupled to these receptors represented about 30% of the total GTPyS-binding sites in HL-60 membranes. The dissociation constant for GTPyS was significantly greater following LTB, stimulation than that produced by f-Met-Leu-Phe stimulation or addition of the two agonists simultaneously.

1-Met-Leu-Phe
The competition between GDP and GTPyS for guanine nucleotide-binding sites of formyl peptide and LTB, receptor- Significantly different from the affinities of the total G protein pool and G proteins coupled to f-Met-Leu-Phe receptors and to both receptors ( p < 0.05).
LTB, coupled G proteins was examined by adding increasing concentrations of GDP in the presence of [35S]GTPyS and f-Met-Leu-Phe or LTB, (Fig. 3). GDP inhibited basal [35S] GTPyS binding half-maximally and maximally at concentrations of about 500 nM and 10 p~, respectively. The doseinhibition curve was shifted to the right in the presence of M LTB,. In the presence of M f-Met-Leu-Phe the curve was shifted further in the right, so that over 10-fold higher concentrations of GDP were required for half-maximal inhibition of [35S]GTPyS binding.
The different affinities for GTPyS following f-Met-Leu-Phe and LTB, stimulation in the presence of GDP suggest that the rate of guanine nucleotide exchange stimulated by f-Met-Leu-Phe should be more rapid than that stimulated by LTB,. Figs. 4 and 5 show examples of the experimental data from which apparent initial rate of GTP hydrolysis between 0 and 5 min and apparent initial rate of GTPyS binding between 20 and 120 s were calculated. Table I1 summarizes the calculations of apparent initial rates. The initial rates of GTP hydrolysis and GTPyS binding were significantly less following LTB, stimulation, while f-Met-Leu-Phe and f-Met-Leu-Phe plus LTB, stimulated similar initial rates of GTP hydrolysis. Quantitative differences in G protein activation between the two receptor types are shown by a significantly greater quantity of GTPyS binding following f-Met-Leu-Phe stimulation at all time points (Fig. 5).
The rate of G protein activation is affected by receptor number (40). Therefore, the density and affinity of f-Met-Leu-Phe and LTB, receptors in Me2SO-differentiated HL-60 granulocyte membranes were determined.  bound was determined at 20-5 intervals between 20 and 120 s. Rates of binding were calculated from the slopes. In this experiment the basal rate was 52.5 fmol of G T P r S bound/mg of membrane protein/ min, the f-Met-Leu-Phe stimulated rate was 150.4 fmol/mg/min, the LTB, stimulated rate was 75.1 fmol/mg/min, and the rate following simultaneous addition of f-Met-Leu-Phe and LTB, was 192.5 fmol/ mg/min. Apparent initial rate was calculated by subtracting the basal rate from the rate in the presence of M f-Met-Leu-Phe and/or 10+ M LTB,. and LTB4, respectively, to HL-60 cell membranes and Scatchard transformation of the binding data. A non-linear, computer analysis of these data indicated that the binding of f-Met-Leu-Phe was best represented by a two-site model. On the other hand, LTB, binding was best described by a onesite model. Analyzing the results of multiple experiments, the total number of f-Met-Leu-Phe receptors was about 4-fold greater than LTB, receptors (Table 111).
To examine the role of differences in receptor number on the different initial rates of G protein activation, membranes from HL-60 cells differentiated with retinoic acid were employed. HL-60 cells were cultivated with 1O"j M retinoic acid for 5 days, and membranes were prepared as described. LTB, receptors were present at a density of 800 fmol/mg. Therefore, these membranes contained over twice the LTB, receptor density found in membranes from MezSO-differentiated cells ( Table 111). The apparent initial rate of GTP hydrolysis in membranes from retinoic acid-differentiated cells following    Significantly different from f-Met-Leu-Phe receptor density, p < 0.05.
ligand binding was performed. Consequently, formyl peptide and LTB, receptor density and affinity were determined in the presence of 0.5 PM GDP and 150 mM NaC1. Under these conditions receptor density was reduced to 180 fmol of formyl peptide receptors/mg membrane protein and 170 fmol of LTB, receptors/mg. Dissociation constant for LTB, receptors was 0.7 nM. Formyl peptide receptors were best described by a one-site model with a dissociation constant of 5 nM. These data indicate that receptor density under the assay conditions for GTPyS binding was similar for formyl peptide and LTB, receptors.

DISCUSSION
The present study was initiated to determine if differences in formyl peptide and LTB, receptor-(; protein interactions could account for the disparate functional and biochemical responses to these agonists. Our study identifies two potential differences between formyl peptide and LTB, receptors. First, formyl peptide receptors are present in greater density on HL-60 granulocytes. Second, the two receptors interact in a qualitatively different manner with a common pool of G proteins, resulting in different affinities for guanine nucleotides. One or both of these differences lead to a greater rate of G protein activation by formyl peptide receptors. One possible explanation for different cellular responses to formyl peptides and LTB, eliminated by our studies is that these receptors are coupled to different G proteins. Our previous study provided evidence that the two receptors shared some common G proteins (7). Both receptors coupled to G proteins which were substrates for pertussis toxin and cholera toxin. Additionally, f-Met-Leu-Phe-specific uncoupling of G proteins from f-Met-Leu-Phe receptors by cholera toxin also uncoupled G proteins from LTB, receptors. Since this uncoupling was not complete, we could not determine if the two receptors shared a totally common pool of G proteins. To address this question, the density of G proteins coupled to f-Met-Leu-Phe and LTB, receptors in HL-60 granulocyte membranes was determined by GTPyS saturation binding, as described by Gierschik et al. (36). We reasoned that, if the two receptors shared a common set of G proteins, the density of GTPyS-binding sites coupled to each receptor type would be the same and would not increase with simultaneous stimulation of the two receptor populations. Our results indicate that formyl peptide and LTB, receptors are coupled to a common set of G proteins ( Fig. 2 and Table I). The pool of G proteins coupled to these receptors represents about one-third of the total G protein content of HL-60 granulocyte membranes, as measured by GTPyS binding. This finding is similar to the report of Gierschik et al. (36) that 40% of GTPyS-binding sites are coupled to formyl peptide receptors in HL-60 cell membranes.
Significant differences in the initial rate of G protein activation by f-Met-Leu-Phe and LTB, resulted in an increased quantity of GTPyS binding following f-Met-Leu-Phe stimulation. To investigate the basis for the difference in rates of activation, receptor density and receptor regulation of G protein affinity for GTPyS were determined. Tolkowsky and Levitzki (40) reported that the rate of effector enzyme activation, and presumably the rate of G protein activation, is dependent on receptor concentration. Chemical inactivation of varying amounts of P-adrenergic receptors resulted in a proportional reduction in the rate of CAMP formation. Their data are supported by several studies demonstrating that effector enzyme activity is proportional to receptor density in the presence of excess ligand (44)(45)(46). Our studies showed about four times as many formyl peptide receptors, at LTB4 receptors, on HL-60 granulocyte membranes. As reported previously (41,47), formyl peptide receptors displayed two affinity states under the conditions of our experiments. On the other hand, LTB, receptors displayed a single affinity state with a dissociation constant of about 0.1 nM. The presence of a single, high affinity state for LTB4 receptors in our study is similar to the results of recent studies (48,49).
Despite the 4-fold difference in receptor number in our study, rates of G protein activation differed by only 2-3-fold. The relative contribution of differences in receptor expression to differences in rate of G protein activation was addressed by two types of experiments. First, the initial rate of GTP hydrolysis in membranes from HL-60 cells differentiated with retinoic acid was determined. These membranes expressed about twice as many LTB, receptors as membranes from MepSO-differentiated cells. The apparent initial rate of GTP hydrolysis following stimulation with LTB, was not significantly increased in these membrane preparations, compared to membranes from MepSO-differentiated cells. Second, the number of formyl peptide and LTB4 receptors detected under the conditions of GTPyS binding was determined. Previous reports indicated that both high guanine nucleotide concentrations and the presence of sodium ions reduce the number of high affinity formyl peptide receptors and the total number of receptors detectable by ligand binding assays (41)(42)(43). In the presence of 0.5 PM GDP and 150 mM NaCl, formyl peptide and LTB, receptors were substantially reduced to about the same density (180 fmol of f-Met-Leu-Phe receptors/mg uersus 170 fmol of LTB, receptors/mg). Thus, under our assay conditions the differences in receptor density were minimized. These results suggest that factors other than differences in receptor expression produced the different rates of G protein activation upon stimulation by f-Met-Leu-Phe and LTB,.
We showed previously that f-Met-Leu-Phe, but not LTB,, was capable of stimulating receptor-specific cholera toxincatalyzed ADP-ribosylation (7). We concluded that formyl peptide and LTB, receptors interacted differently with their G proteins. GTPyS saturation binding data provide additional evidence that the two receptors differ qualitatively in their interaction with a common pool of G proteins. In the presence of 0.5 PM GDP, the affinity for GTPyS was significantly less when G proteins were activated by LTB, than following f-Met-Leu-Phe activation. Receptors promote GTP binding to G proteins by increasing GDP release, decreasing GDP reuptake, and/or increasing affinity for GTP by guanine nucleotide-binding sites. Recent evidence based on competition binding between GDP and GTPyS suggests that receptor activation of G proteins is related to an alteration in GDP release or GDP reuptake, rather than enhanced affinity for GTP (36). Similar to these results, we found that increasing concentrations of GDP inhibited GTPyS binding to HL-60 receptor membranes. The dose-inhibition curve was shifted dramatically to the right in the presence of f-Met-Leu-Phe, while LTB, produced only a minor shift. The shift of the curves to the right in the presence of f-Met-Leu-Phe or LTB, indicates that the agonists alter G protein affinity for GDP, rather than for GTPyS. The difference in the degree of rightward shift between f-Met-Leu-Phe and LTB, suggests that formyl peptide receptors are more efficient than LTB, receptors at inducing GDP release or inhibiting GDP reuptake. These findings, in conjunction with the differences in affinity for GTPyS in the presence of the different agonists, suggest that formyl peptide receptors induce a G protein conformation with a reduced affinity for GDP, compared to the conformational change induced by LTB, receptors. Since GDP-GTP exchange is the rate-limiting step in the cycle of G protein activation, receptor-specific differences in GDP affinity provide a plausible explanation for the different rates of G protein activation. The differences in efficiency of GDP release induced by f-Met-Leu-Phe and LTB, may also explain our previous findings with respect to receptor-specific cholera toxin labeling (7). LTB4 may be unable to stimulate cholera toxin-catalyzed ADP-ribosylation of Gi in the presence of guanosine 5'-(@,y-imido)triphosphate because of the inability to induce guanine nucleotide release, a process necessary for Gi to act as a substrate for cholera toxin.
Until recently, catalytic activation of G proteins had been shown in native membrane only for the photoreceptor rhodopsin (50,51) and in phospholipid vesicles for the P-adrenergic receptor (52,53). Gierschik et al. (36) recently reported that each formyl peptide receptor activates up to 20 G proteins on HL-60 cell membranes. Our data show catalytic activation of G proteins for both formyl peptide and LTB, receptors. We found that each formyl peptide receptor is capable of activating about four G proteins, while each LTB, receptor activated about 17 G proteins. The difference between the degree of catalytic activation by formyl peptide receptors in our study and those previously reported (36) may be due to reduced density of G proteins in the HL-60 membrane preparations we used. On the other hand, when the receptor density determined under the assay conditions for GTPyS binding is used to calculate catalytic activation in the present study, each formyl peptide and LTBl receptor is capable of activating 31-34 G proteins.
In conclusion, our study identifies two factors which may account for differences in neutrophil and HL-60 receptor responses to f-Met-Leu-Phe and LTB,. First, there is a significant difference in the density of receptors for the two agonists under optimal assay conditions. Second, receptors for the two agonists interact differently with a common pool of G proteins. These factors result in different rates of GDP-GTP exchange following addition of f-Met-Leu-Phe and LTB,. Both the quantity of activated G proteins generated and the time over which an increased concentration of (YGTP is present would be predicted to be reduced following addition of LTB,, compared to addition of f-Met-Leu-Phe. Thus, our results suggest mechanisms by which different receptors induce disparate functional responses despite stimulating generation of the same second messengers through a common pool of G proteins. Additionally, our data are consistent with the hypothesis that G proteins can exist in receptor-specific activational states identified by different affinities for guanine nucleotides.