chemoattractant receptors - Europe PMC

3 downloads 180 Views 1MB Size Report
Jul 22, 1991 - regulatory protein (G protein) to activate phospholipase C, resulting inthe formation of ... mediated signals through interactions with G proteins (8). Attenuation ...... Cotecchia, S., Kobilka, B. K., Daniel, K. W., Nolan, R. D., Lapetina,. E. Y., Caron, M. G. ... Rollins, T. E. & Springer, M. S.(1985) J. Biol. Chem. 260 ...
Proc. Nati. Acad. Sci. USA Vol. 88, pp. 11564-11568, December 1991 Medical Sciences

Receptor class desensitization of leukocyte chemoattractant receptors (formylpeptide/C5a/receptor expression/guanine JOHN R. DIDSBURY*t, RONALD J. UHING*, ERIC TOMHAVE*, CRAIG GERARD§, NORMA GERARD§, AND RALPH SNYDERMAN* Departments of *Medicine and *Pathology, Duke University Medical Center, Durham, NC 27710; and IIa Sue Prlinutter Research Laboratory, Children's Hospital, Pulmonary Division, Beth Israel Hospital, Department of Pediatrics, Harvard Medial School, Boston, MA 02115

Communicated by Robert J. Lefkowitz, September 6, 1991 (received for review July 22, 1991)

ABSTRACT To better define their regulation,. tide and CSa chemoatacnt receptor cDNAs were t t expressed with high efficiency (...35-54%) in human kidney cefls. As in neutrophils, both receptors were active In eleva intracellular calcium (EDse 0.51 nM). Agonistrpecfc desnitization for calcium elevation was observed ft bath chensoattractant receptors at doses of 1 nM. el redesensitbaon of formylpeptide, C5a, and a cptors- required high doses of phorbol ester (1iN M phebol 12-nyrsa 13-acetate). To fbrther study the of des ation, forylpetde and C5a receptor CDNAS were cotranfected resulting in "80% of receptor-positive ael pesing both receptors. he ells odn nous a1-adrenergc re s. -Interesng, ch tt receptors were c l by pretreatment d wbIlw m o u doses of either CSa or f (1 (up nM) but not by the ea ric aOt to 10pM). Neither ctdesensitized a

receptors. This penom was reproduced in hunan _mterzd trophils. These data sugest a previously u mechanism of receptor regulation, which is intermediate between homologous and het gous desensitization. Class desensitIzation of ch t receptors is less e ve a _ but is far more efient and spocific Itgzao than heterologous dnsiiation. Receptor class d ation may affect functional classes of receptors via ofeither the receptor or the shared guanine n regulatory protein.

Phagocytes accumulate at sites of inflammation by migrating along gradients of chemoattractants (1). At higher concentrations, chemoattractants stimulate the cell's cytotoxic responses (activation of the respiratory burst and exocytosis of lysosomal enzymes). Specific receptors for chemotactic factors are present on phagocytic leukocytes (i.e., neutrophils and mononuclear phagocytes) and initiate signal transduction via a pertussis toxin-sensitive guanine nucleotide-binding regulatory protein (G protein) to activate phospholipase C, resulting in the formation of diacylglycerol and inositol phospholipids (2, 3). The best-characterized chemoattractant receptor is the one that binds formylpeptides. Recently, cDNAs encoding receptors for three chemoattractants, formylpeptide [e.g., formylmethionylleucylphenylalanine (fMet-Leu-Phe)] (4, 5), C5a (6), and platelet activating factor (7) have been cloned. These receptors are 20-35% identical in their amino acid sequence and have the characteristic structure of a diverse family of receptors that contains seven putative membrane-spanning domains and transduce ligandmediated signals through interactions with G proteins (8).

Attenuation of signaling (i.e., desensitization) has been most extensively studied in (-adrenergic receptors (flARs) and rhodopsin. Two types of rapid desensitization of OARs, homolous and heterologous, have been identified and result from p rybktion of the agonist-occupied rcep0 protein G. (9). tor, which uncouples itofim the Homologous desensitizaton occurs imi response to occupancy of a specific agonist, which s at a MR-speciflc kinase (PARK) to phosphoryble specii sit on the fiAR (10-12). 1Heterologous deeoi ARs occurs in response to agonists for other coupled to activaion of adenylate cyclase and result in this class of receptors being-phosphorylated by cAMP-dpendet protein kinase (PKA) (13, 14). The formylpeptide cbtaut receptor, which transsnitive G-ie G duces signals through a pertssi toxin protein (15, 16), can alsoObe ds tz X presezeof agonist (e.g., fMet-Leu-Phe) (17, 18). Ato s, there is evidence so_ that poor desensitization may result from in receptor-Gprotein interaction (17). Whether phosphorylation of receptor is a mechanism remains to be determined. Exposure to agonist also leads to formylpeptide receptor internalization (19, 20), and inhibition of receptor internalization attenuates desensitization (21). Given the magnitude of the cytotox*c potential of leukocyte responses to chemoattractants, it is likely that these activities are tightly regulated. The availability of cloned cDNAs encoding chemoattractant receptors allows the determination of their structure-function relationships and provides a means to better understand receptor desensitization mechanisms. We have cotransfected and expressed multiple chemoattractant receptor cDNAs in the same cell. Analysis of these cotransfectants indicates three types of rapid chemoattractant receptor desensitization: (i) homologous (e.g., Met-Leu-Phe --+fMet-Leu-Phe), (ii) heterologous [e.g., phorbol 12-myristate 13-acetate (PMA) -. fMet-LeuPhe], and (iii) a third type, which we call receptor class desensitization [e.g., (fMet-Leu-Phe * C5a) a,adrenergic or Prpurinergic]. MATERIALS AND METHODS Matril PMA, norepinephrine, carbachol, and C5a were from Sigma. fMet-Leu-Phe was purchased from Peninsula Laboratories. The acetoxymethyl ester of indo-1 was obAbbreviations: fMet-Leu-Phe, formylmethionylleucylphenylalanine; G protein, guanine nucleotide-binding reuaory protein; PAR, receptor; DARK, BAR-specific kdnase; PMA, phorbol P-adrenergic 13-acetate-; 12-myristate G, stimulatory G protein; G,. inhibitory G protein; PKA, cAMP-dependent protein kinase; [Ca2+]i, intracellular calcium concentration. tTo whom reprint requests should be addressed at: Department of Medicine, Box 3680, Duke University Medical Center, Durham, NC 27710.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

11564

Medical Sciences: Didsbury et al.

tained from Molecular Probes. ATP was from -Pharmacia LKB. Tissue culture media were purchased from GIBCO. DNA Cloning and Plasmid Constructions. Reverse transcription PCR of differentiated HL-60 cell mRNA to obtain formylpeptide receptor-encoding cDNA was carried out essentially as described (22) by using an antisense primer (5'-ClTTGCCTGTAACGCCACCTC-3') and amplified with antisense and sense primer (5'-ATGGAGACAAATTCCTCTCTCC-3') using Vent polymerase (New England Biolabs) for 29 cycles (denaturation at 940C for 30 sec, annealing at 550C for 30 sec, and extension at 720C for 30 sec). The resulting cDNA fragment was labeled by nick-translation and used to screen a AgtlO HL-60d library (23) under low stringency. Formylpeptide receptor-positive clones were confirmed -by double-stranded DNA sequencing as described (24). The open reading frame, devoid of nontranslated sequences, of the formylpeptide receptor cDNA from a receptor-positive clone was generated by PCR to contain a unique EcORI site immediately 5' of the start codon and a HindIII site immediately 3' to the stop codon by using the cloning primers described above, which had an additional 12 bases at their 5' end encoding EcoRI and HindIII sites, respectively. The EcoRI/HindIII-digested PCR-generated receptor cDNA was then directionally cloned into EcoRI/HindIII-cut pRK5 plasmid DNA (25). The EcoRI-digested cDNA fragment encoding the C5a chemoattractant receptor (6) was inserted into the same pRK5 vector. Cells, Cell Culture, and Tranifection. Adenovirus type 5-transformed human embryonic kidney 293 cells were from American Type Culture Collection (ATCC CRL 1573). TSA cells (a clonal variant of 293 cells stably expressing viral large tumor antigen) were kindly provided by G. Rice (Genentech). Cells were maintained in -DMEM/F12 (50:50) supplemented with 10% fetal bovine serum, 2 mM glutamine, penicillin (100 units/ml), and streptomycin (100 gg/ml). Calcium phosphate-mediated transfection of cells was carried out as described (26) by using 8-9 ,ug of CsCl-purified plasmid DNA as a calcium phosphate precipitate in 0.5 ml of 0.82% (wt/vol) NaCI/0.6% (wt/vol) Hepes, pH 7.1/0.02%' (wt/vol) Na2HPO4/0.25 M CaCI2 added to 5 ml of complete medium per 60-mm plate. Cells were exposed to DNA for 18 h, glycerol shocked by exposure to 15% (vol/vol) glycerol in Dulbecco's phosphate-buffered saline (PBS) without Ca2+ and Mg2+ for 15 sec, washed with PBS, incubated in complete medium for 48 h, and analyzed. Transfection efficiency was monitored by flow cytometry and radioligand binding. The approximate degree of formylpeptide and' C5a receptor coexpression was determined by flow cytometric analysis of fluoresceinated formylpeptide and C5a binding in conjunction with the degree of cross-desensitization between fMetLeu-Phe- and C5a-elicited calcium responses in the cell population. Human neutrophils were isolated from peripheral blood of normal volunteers as described (27). Cytosolic Calcium Measurements. Transfected cells were removed from cell culture dishes with Versene (GIBCO), washed once with 10 mM Hepes-buffered Hanks' balanced salt solution (HHBSS) at room temperature, and resuspended in 1.2 ml of HHBSS (5-6 x 106 cells per ml). Cells were loaded with the acetoxymethyl- ester of indo-1 at 1 ,uM for 20 min at room temperature, washed twice with HHBSS, resuspended in 1.2 ml of HHBSS, and placed in a cuvette. The cuvette was placed into a heated (37°C) cuvette holder of a Perkin-Elmer fluorescence spectrophotometer (model 65019). Calcium analyses were carried out, after equilibration of the cells to 37°C (5 min), with an excitation wavelength of 355 nm and an emission wavelength of 405. nm. Maximal and minimal fluorescence were determined in the presence of 0.02% digitonin and 20 mM Tris, pH 8/5 mM EGTA, respectively. Intracellular calcium concentrations ([Ca2+]i) were measured using the following formula [Ca2+]j = Kd(F -

Proc. Natl. Acad. Sci. USA 88 (1991)

11565

Fmin)A(Fmax - F) (28), where F is the observed fluorescence and Flin and Fmax are the minimal and maximal fluorescence, respectively. RESULTS The human formylpeptide chemoattractant receptor was transiently expressed in 293 human kidney cell lines with high efficiency (40-55%; unpublished results). By using intracellular calcium elevation as a measure of receptor action, cells transfected with formylpeptide receptors were exposed to doses of fMet-Leu-Phe followed 5 min later by a second 10 nM dose of fMet-Leu-Phe (Fig. 1A). The- formylpeptide receptors underwent homologous desensitization with an ED" of 1-5 nM fMet-Leu-Phe. Norepinephrine, acting through endogenous a1-adrenergic receptors, had no effect on the calcium response elicited by fMet-Leu-Phe, although a-adrenergic receptors were homologously desensitized (Fig. 1B). To determine if chemoattractant receptors, which as a class utilize a pertussis toxin-sensitive G protein to affect calcium elevation (2, 3), could more effectively desensitize one another, formylpeptide and C5a receptor (6) cDNAs were cotransfected in the same cell population. Expression levels of both chemoattractant receptors, determined by flow cytometric analysis taken together with the degree of crossdesensitization observed (Fig. 2), indicated a maximum of -80%o of 'receptor-expressing cells potentially expressing both formylpeptide and C5a receptors. Cotransfected cells were first exposed to a 10 nM dose of fMet-Leu-Phe and subsequently (5 min later) exposed to either 10 nM fMetLeu-Phe, 10 nM C5a, or 10 uM norepinephrine. As expected, fMet-Leu-Phe pretreatment desensitized by =90%o the secondary response to fMet-Leu-Phe, with no desensitizing effect on norepinephrine-mediated calcium responsiveness. Interestingly, the calcium responsiveness to C5a was inhibited by 72-76% by fMet-Leu-Phe pretreatment (Fig. 2A). Conversely, cotransfectants exposed first to 10 nM C5a desensitized by 91-93% the secondary response to C5a, also with no effect on norepinephrine-mediated calcium responsiveness. The secondary response to fMet-Leu-Phe, however, was inhibited by 70-76% by C5a pretreatment (Fig. 2B). Assuming that 80o of positive transfectants expressed both formylpeptide. and C5a receptors, then the degree of crossdesensitization would approach 90%. At these doses norepinephrine pretreatment had no desensitizing effect on either fMet-Leu-Phe- or C5a-mediated calcium responses. Heterologous desensitization was demonstrated by pretreatment with 100 nM PMA, which lowered by 94-97% the calcium responsiveness to fMet-Leu-Phe, C5a, and norepinephrine (Fig. 2D). To determine whether the phenomenon of receptor class desensitization could be demonstrated in human neutrophils, which endogenously contain these receptors, the ability of fMet-Leu-Phe and C5a to cross-desensitize the calcium responsiveness of each othet was examined. Human neutrophils were exposed to primary doses of fMet-Leu-Phe ranging from 1 to 100 nM and followed 5 min later by a second dose of 1 nM C5a (this dose elicits a calcium response that is -75% of the maximum, whereas a 10 nM dose is =75% of the maximum in transfected cells). A dose-dependent desensitizing effect was observed with >80% inhibition of the C5a response by pretreatment with 100 nM fMet-Leu-Phe (Fig. 3B). Conversely, a similar dose-dependent desensitizing effect of C5a on intracellular calcium elevation by 1 nM fMet-Leu-Phe was apparent (Fig. 3A). Statistical nonpaired i-test analysis of data points at 10 and 100 nM revealed significant desensitizing dose-response differences of fMetLeu-Phe vs. C5a on fMet-Leu-Phe-elicited intracellular calcium elevation [P < 0.01 at 10 nM, P < 0.001 at 100 nM (Fig.

Medical Sciences: Didsbury et al.

11566

Proc. Nad. Acad Sci. USA 88 (1991)

B

21284-

A

0-

25,

100 nM fMLP

_.

10 nM fMLP

10 AM Nor

a

10

a;

304-

X 801 20-

C.)

0

10 nM fMLP

10 nM fMLP

80-

4

4 Nor

.0 co 0

.0

61Cal 54-

0-

0Amd Nor

100 nM fMLP

1.0 nM fMLP

10 nM fMLP

0.1 nM fMLP

10 nM4fMLP

'a245-

-96-

K| \_nephrine

35 -

0-

0-

1

3

5

Time,

min

7

"WI

I1 nM4 fMLP

10 AM Nor ..

1

~3

5

-Time, min

3A)] and on C5a-elicited intracellular calcium elevation [P < 0.001 at 10 nM, P < 0.02 at 100, nM (Fig. 3A)]. F variance analysis, comparing the homologous with the receptor class desensitization curves of Fig. 3 A and B from 10 to 100 nM doses, indicated a significant difference between the two with a P < 0.025. To distinguish receptor class desensitization from heterologous desensitization, we measured the effect of ATP, a P2-purinergic receptor agonist, on COa- and fMetLeu-Phe-mediated calcium elevation and vice versa. Pretreatment with 100 AM ATP, which elicits a calcium response at >100 nM doses of chemoattractants, did not inhibit the subsequent calcium response to 1 nM fMet-Leu-Phe or C5a and actually "primed" the response to these chemoattractants (Fig. 3 A and B). Conversely, fMet-Leu-Phe pretreatment had no desensitizing effect on ATP-mediated calcium responses, whereas C5a had a small desensitizing effect at high concentrations (100 nM) (Fig. 3C). All three agonists at these concentrations (1 nM fMet-Leu-Phe, 1 nM C5a, and 1 pM ATP) gave a submaximal calcium response (d75%, 75%, and 45%, respectively). Calcium elevation via P2purinergic receptors was equally as sensitive to inhibition by pertussis toxin treatment (data not shown) as were chemoattractant receptors (unpublished results). DISCUSSION Leukocyte activation by a single dose of chemoattractant is transient in nature. Chemoattractant-stimulated increases in intracellular calcium are maximum within 20 sec and are at basal levels by 5 min. Moreover, exocytosis and respiratory burst activation are not sustained beyond 2-5 min. These observations suggest that mechanisms exist that terminate the chemoattractant signal. The greatest insight into the mechanism(s) of desensitization of G protein-coupled receptors has resulted from studies of PARs and rhodopsin. f3ARs, which couple to adenylate cyclase by activating G., undergo two types of rapid desensitization, agonist-specific homologous desensitization and agonist-nonspecific heterologous desensitization (i.e., attenuation of adenylate cyclase responsiveness by other hormones) (9-12). The primary mechanism of rapid desensitization involves phosphorylation of the receptor, which uncouples it from GC. Heterologous desensitization occurs at doses of agonists (10 nM), which, by

FIG. 1. Desensitization of formylpeptide receptors. (A) Intracellular calcium elevation in formylpeptide receptor-transfected 293 cells. Indo-l-loaded formylpeptide receptor-transfected 293 cells were exposed to varying doses of fMet-Leu-Phe followed 5 mm later by a second 10 nM dose, and [Ca2+Ji was measured as described in Materials and Methods. (B) Intracellular calcium elevation in formylpeptide receptor-transfected 293 cells. Experiments were performed as in A using the a, agonist norepi(Nor). Representative tracings of three experiments are shown. Intracellular calcium levels measured using the equation [Ca2+]j = Kd(F - Fmn)/(Fm - F) (28) result in a nonlinear y-axis. fMLP, fMetLeu-Phe.

7

activating adenylate cyclase, stimulate PKA-mediated phosphorylation of OARs. This process does not require occupancy of the nonhomologous receptors Homologous desensitization occurs with higher doses of specific agonist (2 AM), where OARs are phosphorylated by a (ARK (10), with- the phosphorylated receptor then associating with (3-arrestin (9) to further impair receptor/G-protein coupling (12). Phosphorylation of 3AR by J3ARK occurs at sites different from those phosphorylated by PKA. The mechanisms involved in the desensitization of chemoattractant receptors are not yet well defined. Formylpeptide receptors are desensitized (i) by exposure to formylpeptides as well as by elevations in intracellular cAMP (29-31), which inhibit calcium influx (32), (ii) by PMA (33), which disrupts coupling of the activated G protein to phospholipase C (34), or (iii) by pertussis toxin (15, 35-38), which disrupts receptor coupling to its G protein (34). Structural similarities (namely, serine/threonine-rich cytoplasmic tails) between chemoattractant receptors and other G protein-coupled receptors suggest that receptor phosphorylation may also play an important role in desensitization of chemoattractant signals. McLeish et aL (17) have demonstrated that formylpeptide receptors on differentiated HL-60 cells are homologously desensitized and suggest two potential mechanisms, loss of membrane receptors and functional alteration in receptor G-protein interaction, analogous to BAR desensitization mechanisms. The present studies sought to define the mechanisms of chemoattractant receptor regulation by analysis of cloned and expressed formylpeptide and C5a receptors. Desensitization of these receptors was analyzed by using calcium elevation as a measure of receptor action. Both homologous and heterologous desensitization was observed. Homologous desensitization was seen with as little as 1 nM fMet-Leu-Phe pretreatment. The a1-adrenergic agonist norepinephrine, at doses that elicited calcium responses greater than those evoked by a 100 nM dose of fMet-Leu-Phe, did not affect fMet-Leu-Phe responsiveness, indicating the specificity of homologous desensitization. These observations also indicated that homologous desensitization of formylpeptide receptors was not due to depletion of intracellular calcium stores. Unlike chemoattractant receptors, a1-adrenergic receptors couple to a pertussis toxin-insensitive G protein to .

Medical Sciences: A

1st dIOSC: 1lIM fMlI .I"

Didsbury et al. B

Proc. Natl. Acad. Sci. USA 88 (1991)

lSt dJOSe: 10 IIM ( §.u 1000-i 1O

7-A

tNILP CO5a NLr -1ld d(i)Se

C

2nd dlose

norepinephrine D 1st dose: 100 nM PMA 100 75 -

0

0

0 0

0

O 50-

4-4

0

el

25 -

(96) (94) (97) 2nd dose

fMLP C5a Nor 2nd dose

FIG. 2. Cross-desensitization of formylpeptide and C5a receptors coexpressed in transfected TSA cells. TSA cells cotransfected with formylpeptide and C5a receptor cDNAs were exposed to an initial dose of either 10 nM fMet-Leu-Phe (A), 10 nM C5a (B), 10 ,M norepinephrine (C) or 100 nM PMA (D), and the [Ca2+]i was measured. Five minutes later, cells were exposed to a second dose (same concentration as the primary dose) of the indicated ligand, and the [Ca2+], was measured. The secondary response was calculated as a percent of the primary response evoked by the same ligand. Each point (bar) is the mean of duplicate determinations; the maximum variability was ±4%. The average degree of inhibition (%) of the response to a second dose of ligand is indicated in parentheses. fMLP, fMet-Leu-Phe; Nor, norepinephrine.

stimulate calcium elevation (refs. 39 and 40; data not shown). Heterologous desensitization was observed in chemoattractant receptor-transfected cells by PMA (100 nM), which

nearly completely desensitized the formylpeptide and C5a chemoattractant receptors as well as a1-adrenergic receptors. To determine whether chemoattractant receptors as a class were regulated coordinately, C5a and formylpeptide receptors were expressed in the same cells. Unexpectedly, cells expressing both C5a and formylpeptide receptors efficiently cross-desensitized each other. The desensitization was receptor class selective since neither C5a nor fMet-Leu-Phe desensitized norepinephrine-elicited calcium responses or vice versa. In contrast, PMA heterologously desensitized all three receptor types. The observation of receptor class desensitization in this receptor expression system led us to examine the phenomenon in neutrophils. The same receptor class desensitization was observed in these cells. The activity and potency of the receptor class desensitization process was less than homologous desensitization as evidenced by the lower ability of 10 nM and 100 nM doses of each chemoattractant to desensitize the other chemoattractant type, whereas nearly complete desensitization was seen to the same chemoattractant (Fig. 3). Statistical analysis of the dose-response curves distinguishing homologous from receptor class desensitization (Fig. 3 A and B) revealed them to be significantly different. In neutrophils, receptor class desensitization was distinguished from heterologous desensitization by the fact that ATP, acting through P2-purinergic receptors, did not desensitize the response to fMet-Leu-Phe or C5a or vice versa, in agreement with previous observations (41). At high doses of C5a (100 nM), there may have been some heterologous desensitization of P2-purinergic receptors. Of note was the potentiating effect of ATP pretreatment on C5a- and fMet-Leu-Phe-mediated calcium responses, which may relate to the previously observed priming effect of ATP on fMet-Leu-Phe-mediated superoxide production (42, 43) and secretion (44). Thus, there appears to be at least three types of chemoattractant receptor desensitization, homologous (i.e., fMetLeu-Phe M fMet-Leu-Phe), heterologous (i.e., PMA M fMetLeu-Phe), and receptor class desensitization [i.e., (fMetLeu-Phe * C5a) a1-adrenergic or P2-purinergic]. Unlike homologous desensitization, which for ,B-adrenergic receptors results in phosphorylation of only agonist-occupied receptors (9), chemoattractant receptor class desensitization does not require agonist occupancy, since neither formylpeptide nor C5a bind to each other's receptor (45, 46) but do efficiently desensitize each other. Receptor class desensitization occurred with relatively low doses of chemoattractant

}

B

125

11567

C

125ci, 0

U, 0

0

"

100

20

7s

100

< 75

co

'-4

50-

"

0. 04 0

10 Pretreatment dose

25-

0

100

1 Pretreatment dose

25] i

Pr tr ae

dose

10 Pretreatment dose

100

FIG. 3. Cross-desensitization of formylpeptide and C5a receptors in human neutrophils. Human neutrophils were loaded with the calcium indicator indo-1 and exposed to the indicated nanomolar concentrations of fMet-Leu-Phe (e) and CSa (-) and micromolar concentrations of ATP (A) (pretreatment dose), followed 5 min later by a second dose of agonist [1 nM fMet-Leu-Phe (A), 1 nM C5a (B), or 1 ,uM ATP (C)], and the maximum [Ca2+J1 was measured. The response to the second dose of agonist is expressed as a percent of the maximum calcium response in the absence of agonist pretreatment. Each point is the mean of two to six determinations. fMLP, fMet-Leu-Phe.

11568

Medical Sciences: Didsbury et al.

(e.g., 10 nM), whereas even high doses of fMet-Leu-Phe (-100 nM) caused no heterologous desensitizing effect on other calcium-elevating receptors (e.g., a1-adrenergic receptors on TSA cells or P2-purinergic receptors on neutrophils). The mechanism for receptor class desensitization is not known, but one possible mechanism could involve a modification of chemoattractant receptors themselves. By analogy with the .AR system, this could be mediated by phosphorylation of chemoattractant receptors by a specific kinase. However, unlike the actions of PARK, which requires agonist occupancy and is receptor specific, phosphorylation of two types of chemoattractant receptors following occupancy of one would require activation of a receptor classspecific kinase that phosphorylates unoccupied as well as agonist-occupied receptors. In this regard, it is noteworthy that alignment of the human formylpeptide and C5a receptors reveals that 7 out of 11 potential phosphorylation sites in the cytoplasmic tails of these receptors are identical. If such a kinase does modify chemoattractant receptors, then how is it activated? It appears that none of the classical products of the phospholipase C pathway are sufficient (i.e., calcium, inositol phosphates, or protein kinase C activation), since there is no desensitizing effect ofnorepinephrine or ATP on chemoattractant responses, even though they are equally effective in activating phospholipase C (40, 44). An alternative mechanism for receptor class desensitization is the activation of a Ga covalent modifier (e.g., a kinase) that affects specific Ga subunits or only free Ga subunits. In this regard, phosphorylation of Gi.2 has been associated with loss of its function and in receptor desensitization in hepatocytes (47, 48). Ligand occupancy of the cAMP chemoattractant receptor in Dictyostelium also results in phosphorylation of its transducing G.2 protein (49). Activation of one type of chemoattractant receptor could activate the pool of relevant G proteins utilized by the entire class of receptors and thereby expose the Ga subunits to a modifying enzyme activated concomitantly with cellular activation. This hypothesis is compatible with the observation that chemoattractant receptors utilize a pertussis toxin-sensitive G protein to elevate calcium, whereas a1-adrenergic and M1muscarinic-cholinergic receptors utilize a pertussis toxininsensitive G protein to initiate this process. However, P2-purinergic receptors do require a pertussis toxin-sensitive mechanism in human neutrophils (41). Whether P2 receptors couple to a different pertussis toxin-sensitive G protein than do chemoattractant receptors is unknown. The data presented here suggest a level of cellular downregulation intermediate between homologous and heterologous desensitization. Homologous desensitization is the most specific in that only agonist-occupied receptors are affected. Heterologous desensitization affects different classes of receptors whether or not they have encountered ligand. Receptor class desensitization appears to fall between homologous and heterologous desensitization and provides a mechanism for negative modulation of functional groups of receptors. Whether this phenomenon applies to receptors other than those for chemoattractants remains to be determined. We acknowledge Drs. G. Rice and A. Ashkenazi for providing 293 cell lines. This work was supported by Grants DE-03738, HL-36162, and CA-29589 from the National Institutes of Health. 1. Snyderman, R. & Uhing, R. J. (1988) in Irflammation: Basic Principles and Clinical Correlates, eds., Gallin, J. I., Snyderman, R. & Goldstein,

I. M. (Raven, New York), pp. 309-323. 2. Snyderman, R., Perianin, A., Evans, T., Polakis, P. & Didsbury, J. R. (1990) in ADP-Ribosylating Toxins and G Proteins: Insights into Signal Transduction, ed. Moss, J. (Am. Soc. Microbiol., Washington), pp. 295-323. 3. Allen, R. A., Traynor, A. E., Omann, G. M. & Jesaitis, A. J. (1988) Hemat. Oncol. Clin. N. Am. 2, 33-59.

Proc. Nad. Acad. Sci. USA 88 (1991) 4. Boulay, F., Tardif, M., Brouchon, L. & Vignais, P. (1990) Biochem. Biophys. Res. Commun. 168,1103-1109. 5. Boulay, F., Tardif, M., Brouchon, L. & Vignais, P. (1990) Biochemistry 29, 11123-11133. 6. Gerard, N. P. & Gerard, C. (1991) Nature (London) 349, 614-617. 7. Honda, Z., Nakamura, M., Miki, I., Minami, M., Watanabe, T., Seyuma, Y., Okado, H., Toh, H., Ito, K., Miyamato, T. & Shimuzi, T. (1991) Nature (London) 349, 342-346. 8. Dohhnan, H. G., Caron, M. G. & Lefkowitz, R. J. (1987) Biochemistry 26, 2657-2664. 9. Leflowitz, R. J., Hausdorff, W. P. & Caron, M. (1990) Trends Pharmacol. Sci. 11, 190-194. 10. Benovic, J. L., Strasser, R. H., Caron, M. G. & Lefkowitz, R. J. (1986) Proc. Natd. Acad. Sci. USA 83, 2797-2801. 11. Lohse, M. J., Benovic, J. L., Codina, J., Caron, M. G. & Lefkowitz, R. J. (1990) Science 248, 1547-1550. 12. Bouvier, M., Leeb-Lundberg, L. M. F., Benovic, J. L., Caron, M. G. & Lefkowitz, R. J. (1987) J. Biol. Chem. 262, 3106-3113. 13. Hausdorff, W. P., Bouvier, M., O'Dowd, B. F., Irons, G. P., Caron, M. G. & Letkowitz, R. J. (1989) J. Biol. Chem. 264, 12657-12665. 14. Clark, R. B., Kunkel, M. W., Friedman, J., Goka, T. J. &Johnson, J. A. (1988) Proc. Natl. Acad. Sci. USA 85, 1442-1446. 15. Smith, C. D., Lane, B. C., Kusaka, I., Verghese, M. W. & Snyderman, R. (1985) J. Biol. Chem. 260, 5875-5878. 16. Spiegel, A. M. (1987) Mol. Cell. Endocrinol. 49, 1-16. 17. McLeish, K. R., Gierschik, P. & Jakobs, K. H. (1989) Mol. Pharmacol. 36, 384-390. 18. Simchowitz, L., Atkinson, J. P. & Spilberg, I. (1980) J. Clin. Invest. 66, 736-747. 19. Jesaitis, A. J., Naemura, J. R., Painter, R. G., Sklar, L. & Cochrane, C. G. (1983) J. Biol. Chem. 258, 1968-1977. 20. Sklar, L. A., Finney, D. A., Oades, Z. G., Jesaitis, A. J., Painter, R. G. & Cochrane, C. G. (1984) J. Biol. Chem. 259, 5661-5669. 21. Jesaitis, A. J., Tolley, J. 0. & Allen, R. A. (1986) J. Biol. Chem. 261, 13662-13669. 22. Lee, C. C. & Caskey, C. T. (1989) in Genetic Engineering: Principles and Methods, eds. Setlow, J. K. & Hollaender, A. (Plenum, New York), Vol. 11, pp. 159-170. 23. Didsbury, J. R. & Snyderman, R. (1987) FEBS Lett. 219, 259-263. 24. Chen, E. Y. & Seeburg, P. H. (1985) DNA 4, 165-170. 25. Eaton, D. L., Wood, W. I., Eaton, D., Hass, P. E., Hollingshead, P., Wion, K., Mather, J., Lawn, R. M., Vehar, G. A. & Gorman, C. (1986) Biochemistry 25, 8343-8347. 26. Gorman, C. (1985) in DNA Cloning: A Practical Approach, ed. Glover, D. M. (IRL, Washington), pp. 143-165. 27. Smith, C. D. & Snyderman, R. (1987) Methods Enzymol. 141, 261-271. 28. Cobbold, P. H. & Rink, T. J. (1987) Biochem. J. 248, 313-328. 29. Simchowitz, L., Fischbein, L. C., Spilberg, I. & Atkinson, J. P. (1980) J. Immunol. 124, 1482-1491. 30. Rivkin, I., Rosenblatt, J. & Becker, E. L. (1975) J. Immunol. 115, 1126-1134. 31. Gallin, J. I., Sandier, J. A., Clyman, R. I., niello, V. C. & Vaughn, M. (1978) J. Immunol. 12E0, 492-46. 32. Takenawa, T., Ishitoya, J. & Nagai, Y. (1986) J. Biol. Chem. 261, 1092-1098. 33. Naccache, P. H., Molski, T. F. P., Borgeat, P., White, J. R. & Sha'afi, R. I. (1985) J. Biol. Chem. 260, 2125-2131. 34. Smith, C. D., Uhing, R. J. & Snyderman, R. (1987) J. Biol. Chem. 262, 6121-6127. 35. Verghese, M. W., Smith, C. D. & Snyderman, R. (1985) Biochem. Biophys. Res. Commun. 127, 450-457. 36. Brandt, S. J., Dougherty, R. W., Lapetina, E. G. & Niedel, J. E. (1985) Proc. Natl. Acad. Sci. USA 82, 3277-3280. 37. Volpi, M., Naccache, P. H., Molski, T. F. P., Shefcyk, J., Huang, C., Marsh, M. L., Munoz, J., Becker, E. L. & Sha'afi, R. I. (1985) Proc. Natl. Acad. Sci. USA 82, 2708-2712. 38. Okajima, F., Katada, T. & Ui, M. (1985) J. Biol. Chem. 260, 6761-6768. 39. Exton, J. H. (1988) FASEB J. 2, 2670-2676. 40. Cotecchia, S., Kobilka, B. K., Daniel, K. W., Nolan, R. D., Lapetina, E. Y., Caron, M. G., Lefkowitz, R. J. & Regan, J. W. (1990) J. Biol. Chem. 265, 63-69. 41. Walker, B. A. M., Hagenlocker, B. E., Douglas, V. K., Tarapchak, S. J. & Ward, P. A. (1991) Lab. Invest. 64, 105-112. 42. Kuhns, D. B., Wright, D. G., Nath, J., Kaplan, S. S. & Basford, R. E. (1988) Lab. Invest. S, 448-453. 43. Seifert, R., Burde, R. & Schultz, G. (1989) Biochem. J. 259, 813-819. 44. Cockcroft, S. & Stutchfield, J. (1989) Biochem. J. 263, 715-723. 45. Rollins, T. E. & Springer, M. S. (1985) J. Biol. Chem. 260, 7157-7160. 46. Bender, J. G., Van Epps, D. E. & Chenoweth, D. E. (1987) J. Immunol. 139, 3028-3033. 47. Bushfield, M., Murphy, G. J., Lavan, B. E., Parker, P. J., Hruby, V. J., Milligan, G. & Houslay, M. D. (1990) Biochem. J. 268, 449-457. 48. Bushfield, M., Griffiths, S. L., Murphy, G. J., Pyne, N. J., Knowler, J. T., Milligan, G., Parker, P. J., Moliner, S. & Houslay, M. D. (1990) Biochem. J. 271, 365-372. 49. Gundersen, R. E. & Devreotes, P. N. (1990) Science 248, 591-593.