THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1992 hy The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 1, Issue of January 5, pp. 358-363,1992 Printed in U.S.A.

Regulation of Formyl Peptide Receptor Expression and Its mRNA Levels during Differentiation of HL-60 Cells*

(Received for publication, July 22, 1991)

H. Daniel Perez$, Edward Kelly, and Richard Holmes From the Rosalind Russell Arthritis Research Laboratory, University of California, Sun Francisco, and The Medical Service, San Francisco General Hospital, San Francisco, California 94110

When incubated with M-Z’-O-dibutyryladenosine 3’,5‘-cyclic monophosphate (dbcAMP), HL-60 cells ex- pressed formyl peptide receptor (FPR) (as assessed by ligand binding) and FPR transcripts in a time- and concentration-dependent fashion. Experiments using dbcAMP analogs modified at either the C-6 or C-8 position indicated that the process was mediated by a protein kinase A type I, and protein kinase A type I activity was isolated from undifferentiated HL-60 cells by DEAE-Sephacel chromatography. Forskolin mim- icked the effects of dbcAMP. Forskolin and dbcAMP- dependent expression of FPR and FPR transcript was inhibited by staurosporine. Retinoic acid (but not ret- inal or retinol) was capable of inhibiting dbcAMP- dependent expression of FPR and FPR transcripts without affecting FPR mRNA half-life. Dexametha- sone enhanced the effects of dbcAMP and blocked the inhibitory effect of retinoic acid on expression of FPR and FPR transcripts. Phorbol 12-myristate 13-acetate (PMA) alone (1.5-15 nM) failed to induce HL-60 to express FPR and FPR transcripts. Low concentrations (1.5 nM) of PMA enhanced the ability of dbcAMP to induce HL-60 cells to express FPR and FPR transcript, whereas high (15 nM) concentrations of PMA inhibited dbcAMP effects. These results indicate that expression of FPR and FPR transcripts by HL-60 cells can be up- and down-regulated by agents that induce HL-60 cells to differentiate and that a “cross-talk” effect exists between protein kinase A and protein kinase C that modulates FPR gene transcription (and receptor expression) by these cells.

Binding of formyl peptide to its specific receptor(s) on human neutrophils and monocytes stimulates these cells to migrate in a directed fashion, selectively releases a portion of their lysosomal contents, and generates highly reactive oxy- gen-derived free radicals (1). Neutrophils are short lived, terminally differentiated cells. Thus, they are not good models to study formyl peptide receptor gene expression.

The human promyelocytic leukemia cell line HL-60 has

*This work was supported by Grants AM-28566 and AI-28290 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ T o whom correspondence should be addressed Bldg. 30, Rm. 3300, San Francisco General Hospital, 1001 Potrero Ave., San Fran- cisco, CA 94110. Tel.: 415-821-8189; Fax: 415-648-8425.

been used as a model to study formyl peptide receptor (FPR)’ modulation (2, 3). HL-60 cells can be induced to differentiate into either neutrophils or monocyte-like cells by a variety of agents (4). For example, planar solvents (ie. dimethyl sulf- oxide), dibutyryl CAMP, and retinoic acid are capable of inducing HL-60 cells into neutrophil-like cells. Phorbol esters on the other hand, induce HL-60 cells to differentiate into a monocyte-like cell. Differentiation of HL-60 cells into neutro- phil-like cells is accompanied by the expression of functional receptors for chemotactic peptides (and lipids), including FPR (2, 3, 5). It should be noted however, that not all stimuli that promote differentiation of HL-60 cells into neutrophil-like cells induce expression of FPR. While planar solvents and dibutyryl bcAMP do, retinoic acid does not (4, 6). Retinoic acid differentiated HL-60 cells however acquire markers of mature neutrophils, such as expression of adhesion-related glycoproteins (7) and oxidase activity (8). Interestingly, phor- bo1 ester-differentiated HL-60 cells acquire markers of mature monocytes but they do not express FPR (whereas normal human monocytes do) (6, 7). These results suggest that a complex regulation is involved in mediating FPR expression. The mechanisms involved in regulating FPR gene transcrip- tion and FPR expression have not been studied. We have begun to address some of these questions, and in this report we present data indicating that transcription of the FPR gene (and expression of receptors) in HL-60 cells can be up- and down-regulated by agents that induce differentiation. Fur- thermore, it appears that a “cross-talk” effect exists between protein kinase A (PKA) and protein kinase C (PKC) that modulates FPR gene transcription (and receptor expression) in HL-60 cells.

EXPERIMENTAL PROCEDURES

Reagents and Supplies-IP“2’-O-dibutyryl adenosine 3’:5’-cyclic monophosphate (dbcAMP), NG-benzoyl CAMP, NG-monobutyryl CAMP, 8-bromo CAMP, 8-(6-aminohexyl)amino CAMP, 8-(4-choro- phenylthio) CAMP, 3-isobutyl-1-methylxanthine (IBMX), forskolin, staurosporine, histone type 11-A, retinoic acid, retinol, retinal, dexa- methasone 21-phosphate, synthetic rabbit protein kinase A inhibitor, bovine insulin, and 4n-phorbol 12,13-didecanoate were from Sigma.

Phorbol12-myristate 13-acetate (PMA) was from Calbiochem. Cell culture media, buffer, and supplements were purchased from GIBCO. [Y-~’P]ATP, [a-32P]dCTP, and carrier-free Na-’” were from Amer- sham Corp. DEAE-Sephacel was from Pharmacia LKB Biotechnol- ogy Inc.

Cell Culture-HL-60 cells were maintained in suspension culture

The abbreviations used are: FPR, formyl peptide receptor; PKA, protein kinase A; PKC, protein kinase C; PMA, phorbol12-myristate 13-acetate; FP, N-formyl-Nle-Leu-Phe-Tyr; dbcAMP, N6-2’-0-di- butyryl adenosine 3’:5’-cyclic monophosphate; IBMX, 3-isobutyl-l- methylxanthine; ST, staurosporine; RA, retinoic acid; Cpt CAMP, 8- (4-chlorophenylthio) CAMP; Mnb CAMP, P-monobutyryl CAMP; CRE, dbcAMP response element; CREB, CRE-binding protein; RIn, PKA regulatory subunit a.

358

Formyl Peptide Receptor Gene Expression by HL-60 Cells 359 in RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum, penicillin (100 units/ml), streptomycin (100 pglml), L-gluta- mine (2 mM), and bovine insulin (10 mg/l). For differentiation exper- iments, cells were seeded at a density of 3.0 X 106/ml (20 ml) in 150- cm2 flasks, and 20 ml of media containing the desired reagents were added. Cells were incubated at 37 "C, 5.0% C02, 95% humidity for varying periods of time. Differentiation was assessed morphologically (8), cell numbers were determined by triplicate counts using a Sysmex cell counter, and viability was assessed by trypan blue dye (0.1%) exclusion.

Binding Studies-N-Formyl-Nle-Leu-Phe-Tyr (FP) (Sigma) was radioiodinated by the chloramine-T method, as described (9). Specific activity of lZ5I-FP was 1700 Ci/mmol. Equilibrium binding (4 "C, 20 min) of Iz5I-FP (8.0 nM) to suspended HL-60 cells (2 X lo6) was assessed as described previously (9). Nonspecific binding (i.e. binding in the presence of 1000-fold excess unlabeled FP) did not account for more than 3-7% of total binding.

Generation of an FPR Probe-HL-60 cells were induced to differ- entiate by incubation with 90 mM dimethyl formamide and 500 nM dexamethasone for 72 h (6). Poly(A)+ RNA from differentiated HL- 60 cells (expressing FPR as assessed by binding of 12'I-FP) was obtained using the Fast Track isolation kit (Invitrogen) and used to construct a cDNA library in pcDNAl (Invitrogen), according to the manufacturer's instructions. Two oligonucleotides corresponding to the 5' (sense) and 3' (antisense) sequences of published HL-60 FPR (10) containing EcoRI sites at their ends were synthesized and used to amplify the HL-60 FPR open reading frame from the differentiated HL-60 library by the polymerase chain reaction. The 1050-base pair fragment generated was purified by agarose gel electrophoresis, sub- cloned into M13, and sequenced by the dideoxy method, using Se- quenase 2.0 (U. S. Biochemical Corp.). Sequence of the probe revealed a perfect match with the previously reported sequence (10).

Northern Blot Analysis-Poly(A)+ RNA (0.5 pg) was electropho- resed through 1.0% agarose formaldehyde gels, blotted to Zeta-Probe (Bio-Rad) membranes using a positive pressure blotter (Stratagene) followed by UV-cross-linking (UV Stratalinker, Stratagene). The blots were probed using the FPR open reading frame probe described above that had been labeled with [m3']dCTP by random priming. Hybridization was done at 42 "C for 18 h. After hybridization, the blots were washed once at room temperature with 2 X SSPE, 0.1% sodium dodecyl sulfate for 20 min followed by a wash using 0.2 X SSPE, 0.1% sodium dodecyl sulfate at 42 "C for 20 min. After washing, the blots were exposed to X-Omat AR5 film at -70 "C overnight with an intensifying screen, after which they were stripped and reprobed with a 32P-labeled human &actin (Clontech Laboratories, Inc.) probe, under identical conditions. Films were scanned using a Hoefer GS 300 densitometer (Hoefer Scientific Instruments). FPR mRNA levels represent the densitometric reading obtained using the specific FPR probe divided by the densitometric readings obtained using the actin probe. Actin message expression was not altered under the experi- mental conditions used here.

Isolation of HL-60 PKA-Undifferentiated HL-60 cells (4.0 X 10' cells) were suspended in 4 ml of 10 mM KP04, 10 mM EDTA, 0.5 mM IBMX, pH 6.8 (buffer A) at 4 "C, homogenized with four strokes of a Dounce homogenizer and then centrifuged at 27,000 X g for 20 min at 4 "C. The supernatant was applied to a 1.0 X 4.0-cm column packed with DEAE-Sephacel and equilibrated in buffer A (11). The column was then washed with 50 ml of buffer A, after which a linear gradient of NaCl (0-0.4 M in buffer A) was started. Fractions (4 ml) were collected and a 20-p1 aliquot of each assayed for PKA activity, as described by Corbin et al. (ll), in the presence and absence of 2.0 p~ dbcAMP. The activity ratio was determined as described (11). Active fractions were retested in the presence of 500 ng of rabbit PKA inhibitor to determine specificity of the reaction.

RESULTS

Effect of dbcAMP on FPR Expression by HL-60 Cells- Initially, we sought to determine the effect of dbcAMP on transcription and translation of FPR by HL-60 cells. HL-60 cells were incubated with increasing concentrations of dbcAMP for 72 h, after which we determined both their ability to bind l2'1-FP (Fig. LA) and express FPR transcript (Fig. IC). dbcAMP induced HL-60 cells to express FPR i n a concentration-dependent fashion (Fig. LA). Very little FPR was expressed at concentrations of 0.05-0.1 mM dbcAMP,

1500 IA. I Kb ''

I 0.05 0.1 0.2 0.3 0.5 $:;

N

2 cAMP(mM)

D. 1

0 24 46 72 Time (h)

FIG. 1. Effect of dbcAMP on FPR expression by HL-60 cells. A, specific binding of lZ5I-FP by HL-60 cells that had been incubated with increasing concentrations of dbcAMP for 72 h at 37 "C (mean f S.E., n = 3). B, specific binding of "'I-FP by HL-60 cells that had been incubated (37 "C) with dbcAMP (0.2 mM) for increasing periods of time (mean f S.E., n = 3). C, expression of FPR transcript by HL-60 cells that had been incubated with increasing concentrations of dbcAMP for 72 h at 37 "C. Top, FPR transcript; bottom, actin control. D, expression of FPR transcript (depicted as mRNA levels, see "Experimental Procedures") by HL-60 cells that had been incubated (37 "C) with dbcAMP (0.2 mM) for increasing periods of time. Determinations of '251-FP-specific binding (total binding minus nonspecific, 4 "C for 20 min) and expression of FPR transcript (ie. mRNA isolation) were performed simultaneously. Kb, kilobases.

whereas 0.2 mM dbcAMP-induced FPR expression was ap- proximately 65-70% of maximal. Maximal FPR expression was observed at 0.3-0.5 mM dbcAMP (Fig. LA). This concave upward concentration curve has been observed in other cell types (12). The shape of the curve was unaffected by IBMX (0.25 mM) (although the concentration responses were shifted to the left, not shown), suggesting that neither phosphodies- terase nor inhibitory adenosine receptors play a major role in this process (12). dbcAMP-induced expression of FPR was associated with expression of a 1.6-kilobase FPR transcript (Fig. IC). FPR transcript was barely detectable at 0.1 mM dbcAMP and reached maximum levels at 0.3-0.5 mM dbcAMP. As reported in other systems (13), dbcAMP had no effect on actin mRNA levels (Fig. IC). Studies were performed to determine the time course of dbcAMP-induced FPR expres- sion by HL-60 cells. Incubation of HL-60 cells with 0.2 mM dbcAMP resulted in a time-dependent expression of FPR. Maximal expression of FPR (Fig. 1B) and FPR transcript (Fig. 1D) was observed at 48-72 h. Longer incubations (96 h) resulted in a 15-20% diminution in the ability of HL-60 cells to bind I2'I-FP and a 30-40% diminution in FPR transcript (not shown).

Studies were also performed using other membrane-perme- able dbcAMP analogs. HL-60 cells were incubated with ana- logs modified at the C-6 (0.2 mM) or C-8 (0.2 mM) position or with a combination of C-6 and C-8 analogs (0.2 mM each) for 72 h, after which their ability to bind "'1-FP and expression of FPR transcript were determined. When tested alone, C-6 analogs induced both FPR expression (Fig. 2 A ) and FPR transcript (Fig. 2B). dbcAMP was the most effective, followed

360 Formyl Peptide Receptor Gene Expression by HL-60 Cells

0 1

C6 7 1

c8 Analogs

C6cC8 - 1 I

I I I I - I l l l"l cAMP Ez

"t Mnb

rr- -L

J - -

Er Am Ez Bz Mnb Mnb None

A'm & A', 1.7 2.38 4.18

+ Br

1 .A6 "" - 8.

Analogs c6 C8

7n "' CB+CS I

I n I

+ + + + Br Am 01 Am

FIG. 2. Effect of dbcAMP analogs on expression of FPR by HL-60 cells. A , specific binding of "'I-FP by HL-60 cells that had been incubated with cAMP analogs modified at either the C-6 or C- 8 position (0.2 mM each) or with a combination of C-6 and C-8 analogs (0.2 mM each) for 72 h at 37 "C. A synergism quotient was determined as described (Ref. 12). B, expression of FPR transcript by HL-60 cells that were incubated as described in A . Determinations of '2,51-FP binding and expression of FPR transcripts were performed simultaneously. CAMP, dibutyryl CAMP; Bz, W-benzoyl CAMP; Mnb, W-monobutyryl CAMP; Br, 8-bromo CAMP; Am, 8-(6-amino- hexy1)amino CAMP.

by W-benzoyl cAMP (Bz) and W-monobutyryl cAMP (Mnb). The C-8 analogs 8-bromo cAMP (Br ) and 8-(6-amino- hexy1)amino cAMP (Am) were inactive both as inducers of FPR expression (Fig. 2 A ) and FPR transcript (Fig. 2B). This pattern of activity was suggestive of a PKA type I-mediated response (12). If so, C-8 analogs should potentiate the effects of C-6 analogs. To quantify further the effects of analog combinations, we used a synergism quotient, defined as the net effect of the analog combination on '"I-FP binding di- vided by the sum of the net individual analogs effects, as described by Beebe et al. (12). C-8 analogs uniformly increased the ability of C-6 analogs to induce FPR expression (Fig. 2 A ) and FPR transcript (Fig. 2B). The most potent combination was Mnb + Am (synergism quotient, 4.18) followed by Mnb + Br (2.38), Bz + Am (1.7, and Bz + Br (1.46). These results indicated that the effect of dbcAMP was mediated by a PKA type I. To confirm these results, we sought to determine if undifferentiated HL-60 cells contain PKA type I. HL-60 cell PKA activity has never been characterized. Isolation of HL- 60 PKA on DEAE-Sephacel (Fig. 3) demonstrated that PKA activity eluted in two peaks: one, as soon as the salt gradient was started, consistent with PKA type I (11) and the second a t 0.15-0.18 M NaC1, consistent with PKA type I1 (11). PKA activity could not be detected in the absence of dbcAMP. Incubation of active fractions with synthetic rabbit PKA inhibitor abolished their activity, confirming that the activity was dependent upon intact PKA (i.e. regulatory catalytic subunits) (11). Experiments were performed to determine if the effects of dbcAMP on FPR expression and expression of FPR transcript required a phosphorylation step. HL-60 cells were incubated for 24 h in the presence of either dbcAMP (0.2 m) + IBMX (0.25 mM) or dbcAMP + IBMX and increas- ing concentrations of the protein kinase inhibitor staurospor-

1.3 , I 0.4

Fraction number

FIG. 3. Isolation of PKA activity from undifferentiated HL- 60 cells. DEAE-Sephacel chromatography of PKA in extracts of undifferentiated HL-60 cells. A-A, PKA activity in the absence of PKA inhibitor; A-A, PKA activity in the presence of PKA inhibitor.

0.25

O ' I

+ + + + 0 10 25 50

+ ST(nM) 0 10 25 50 F K + + + +

ST(nM1 0 FK +

50

FIG. 4. Staurosporine inhibits dbcAMP and forskolin (FK)- induced FPR expression by HL-60 cells. HL-60 cells were incu- bated (37 "C) for 24 h with either dbcAMP (0.2 mM) and IBMX (0.25 mM) ( A ) or forskolin (25 p ~ ) and IBMX (0.25 mM) ( B ) in the absence or presence of increasing concentrations of ST. At the end of incubation, their ability to bind '*'I-FP specifically was determined. C, FPR mRNA levels from experiments depicted in A. D, FPR mRNA levels from experiments depicted in B. Determinations of "'I-FP binding and expression of FPR transcripts were performed simulta- neously.

ine (ST) (10-50 nM). ST is a potent inhibitor of PKA (as well as PKC) (14, 15). A 24-h incubation period was chosen to prevent possible effects of ST on HL-60 growth (15). IBMX was used to maximize dbcAMP effects. After 24 h, cells incubated with dbcAMP + IBMX in the presence and absence of ST exhibited the same kinetics of growth (as determined by cell counts) and were differentiated to the same extent (as determined by morphology). ST did not alter viability of HL- 60 cells. Presence of S T inhibited, in a concentration-depend- ent fashion, the expression of both FPR (Fig. 4A) and FPR transcript (Fig. 4C) induced by dbcAMP and IBMX. The terpenoid forskolin directly activates adenylate cyclase. In- cubation of HL-60 cells for 24 h with a combination of forskolin (25 PM) + IBMX (0.25 mM) induced expression of FPR (Fig. 4B) and FPR transcript (Fig. 4D), a process that was inhibited by ST (Fig. 4, B and D ) .

Effect of Retimic Acid and Dexamethasone on Expression of FPR by HL-60 Cells-Incubation of HL-60 cells for 72 h with increasing concentrations of retinoic acid (RA) (0.2-5.0 kh4) induced these cells to differentiate into neutrophil-like cells

Formyl Peptide Receptor Gene Expression by HL-60 Cells 361

(as assessed by morphology), similar to previous reports (7, 8). Although HL-60 cells differentiated in response to RA, they failed to express both FPR (not shown) and FPR tran- script (Fig. 5) (6). RA, however, induced expression of CD18 transcript (Fig. 5), as reported previously (7). RA has been reported to decrease binding of [3H]fMet-Leu-Phe by HL-60 cells induced to differentiate with planar solvents (6). Con- sequently, we examined the effect of RA on dbcAMP-me- diated induction of FPR and FPR transript by HL-60 cells. HL-60 cells were incubated for 72 h with either dbcAMP (0.2 mM) or dbcAMP and increasing concentrations of RA (0.1- 10 p ~ ) , after which their ability to bind T - F P and expression of FPR transcript were determined. RA induced a concentra- tion-dependent diminution in both expression of FPR (Fig. 6A) and expression of FPR transcript (FIg. 6B). To study further the effect of RA on dbcAMP-induced HL-60 FPR expression, we performed time course analysis of the RA effect. HL-60 cells were incubated with dbcAMP (0.2 mM) in the presence and absence of 1.0 p~ RA for varying periods of time, after which we determined the ability of the cells to bind 1251-FP and express FPR transcript. Incubation of HL- 60 cells with dbcAMP induced a time-dependent expression of FPR and FPR transcript (Fig. 7, closed circles) that was maximal after 72 h of incubation. Incubation of HL-60 cells with dbcAMP and RA resulted in significant diminution in both expression of FPR and FPR transcript (Fig. 7, open triangles). After 24 h, RA had minimal effect on both expres- sion of FPR and FPR transcript. Maximal effects of RA on

A. B. FPR CD1S

-9 - - =.rm'' -24h-

... -48 h- - - -72h- - -

FIG. 5. Effect of RA and dbcAMP on expression of FPR and CDlS transcripts by HL-60 cells. HL-60 cells were incubated (37 "C, 24-72 h) with either RA (1.0 pM) or dbcAMP (0.2 mM), after which their ability to bind Iz5I-FP was determined and mRNA ob- tained. mRNA (0.2 pg) was blotted to Zeta-Probe membranes using a slot-blot apparatus. Membranes were processed as described under "Experimental Procedures" and incubated with either labeled FPR probe or a labeled 1.8-kilobase insert probe for CD18 (7) prior to autoradiography. A, left, mRNA obtained from control (undifferen- tiated) and RA-differentiated (24-72 h) HL-60 cells hybridized with the FPR probe. Right, identical experiment using mRNA from control and dbcAMP-differentiated HL-60 cells. E, identical experiment as described in A, except that the blot was hybridized with the CD18 probe. dbcAMP-differentiated HL-60 expressed FPR (as assessed by binding of Iz5I-FP), whereas RA-differentiated cells did not.

0.0 " 5.0 10 5.0 10

R A W )

FIG. 6. Effect of RA on dbcAMP-induced expression of FPR by HL-BO cells. HL-60 cells were incubated (37 "C, 72 h) with dbcAMP (0.2 mM) in the absence of presence of increasing (0.1-10 pM) concentrations of RA, after which their ability to bind Iz5I-FP specifically and to express FPR transcripts was determined. A, spe- cific binding of "'1-FP. E, FPR mRNA levels.

24 40 i 2

Time (h)

FIG. 7. Time course analysis of RA effect on dbcAMP-in- duced expression of FPR by HL-60 cells. HL-60 cells were incubated (37 "C) with dbcAMP (0.2 mM) in the absence or presence of RA (1.0 phi) for increasing periods of time, after which their ability to bind 1251-FP and express FPR transcript was determined. In some instances (arrows), RA (1.0 p ~ ) was added after preincubation of cells with dbcAMP (0.2 mM) alone for either 24 or 48 h, and the effect of RA addition on "'I-FP binding and expression of FPR transcript was determined 24 and 48 h later. A, specific binding of Iz5I-FP by HL-60 cells incubated with either dbcAMP alone (M), dbcAMP + RA (A-A), dbcAMP alone for 24 h followed by addition of RA (A-A), or dbcAMP alone for 48 h followed by addition of RA (W). B, FPR mRNA levels from experiments depicted in A.

expression of FPR were observed after 72 h (Fig. 7A, open triangles). Expression of dbcAMP-induced FPR transcript was inhibited by RA, but levels remained relatively constant during the 72 h of incubation (Fig. 7B, open triangles), sug- gesting that the half-life of FPR transcript was not affected by RA. During the same experiments, we determined the effect of RA (1.0 p ~ ) addition on expression of FPR and FPR transcript during the course of dbcAMP (0.2 mM)-induced differentiation of HL-60 cells. Addition of RA after 24 h of incubation with dbcAMP alone resulted in significant dimi- nution in expression of FPR (Fig. 7A, closed triangles). In fact, the binding measured at 48 and 72 h (24 and 48 h after addition of RA, respectively) were lower than those observed when HL-60 cells were incubated with dbcAMP plus RA from time 0 (Fig. 7A, open triangles). In contrast, expression of FPR transcript increased after 24 h of RA addition, albeit much less than in controls (Fig. 7B, closed triangles) and diminished after 48 h of RA addition. Thus, it would appear that retinoic acid modulates expression of FPR by dbcAMP- induced HL-60 cells both at the transcriptional and posttran- scriptional level. Addition of RA after 48 h of dbcAMP resulted in diminution in both expression of FPR and FPR transcript at 72 h (24 h after addition) (Fig. 7, open circles). Experiments were performed to assess directly the effect of RA on the levels of FPR transcripts. HL-60 cells were incu- bated with 0.2 mM dbcAMP in the presence and absence of 1.0 p~ RA. After 48 h, actinomycin (10 pg/ml) was added, and its effects on FPR mRNA (and T - F P binding) were monitored at time intervals (Fig. 8). Although FPR transcripts were more abundant in cells exposed to dbcAMP alone (as compared with dbcAMP + RA), the rate at which mRNA decayed was identical either in the presence or absence of RA (Fig. 8B). Interestingly, expression of FPR (as assessed by binding of 1251-FP) decreased by approximately 50% after 8 h of incubation with actinomycin (Fig. 8A). These effects of retinoic acid on expression of FPR (as assessed by binding) were specific since neither retinal ( R m l ) nor retinol (Rol) had any effect on dbcAMP (0.2 mM)-induced (72 h) expression of FPR (Fig. 9A). Next, we studied the effects of dexametha- sone ( D E X ) on expression of FPR and FPR transcript by HL-60 cells. Incubation of HL-60 cells with dexamethasone (500 nM) alone for 72 h failed to induce expression of either FPR or FPR transcript (not shown). Dexamethasone how-

362 Formyl Peptide Receptor Gene Expression by HL-60 Cells

100 8-

B 0- 2 4 6 8

Time (h) 1 2 3 4

FIG. 8. Effect of RA on levels of FPR mRNA. HL-60 cells were incubated (37 "C, 48 h) with dbcAMP (0.2 mM) in the absence or presence of RA (1.0 hM). After 48 h, actinomycin (10 pg/ml) was hdded, and its effects on the ability of HL-60 cells to specifically bind

I-FP and to express FPR transcripts was assessed at time intervals (1-8 h). A, specific binding of T - F P by HL-60 cells incubated with either dbcAMP (M) or dbcAMP + RA (A-A). B, FPR mRNA levels with dbcAMP ( C - . ) or dbcAMP + RA (A-A). FPR mRNA was determined by densitometric analysis of Northern blots and expressed as percent remaining (time 0 being 100%). Although mRNA was obtained at all time points, FPR transcript was undetectable after 4 h of incubation with actinomycin.

125

FIG. 9. Effect of dexamethasone (DEX) and RA-analogs on dbcAMP (CAMP)-induced expression of FPR by HL-60 cells. HL-60 cells were incubated (37 "C, 72 h) with either dbcAMP (0.2 mM), dbcAMP + dexamethasone (500 nM), dbcAMP + RA (1.0 p ~ ) , dbcAMP + RA + dexamethasone, dbcAMP + retinol (Rol, 1.0 PM), or dbcAMP + retinal (Rnal, 1.0 FM), after which their ability to bind "'I-FP and express FPR transcript was determined. A , specific bind- ing of '*'I-FP. B, FPR mRNA levels. Dexamethasone failed to induce expression of either FPR or FPR transcript. Neither retinol nor retinal had any effect on FPR mRNA levels.

ever, enhanced both dbcAMP (0.2 mM)-induced expression of FPR (Fig. 9A) and FPR transcript (Fig. 9B). Also, dexameth- asone partially blocked the inhibitory effect of RA on dbcAMP-induced expression of FPR and FPR transcript (Fig. 9).

Effect of PMA on Expression of FPR by HL-60 Cells- Experiments were performed in which we determined the ability of PMA to induce expression of FPR and FPR tran- script by HL-60 cells. HL-60 cells were incubated with PMA (1.5 and 15 nM) for 24 and 48 h, after which their ability to express FPR and FPR transcript was determined. Incubation with PMA induced the cells to differentiate into a monocyte- like cell line (as determined by morphology), and a population of cells became adherent to the flask. For all experiments, adherent cells were lifted using buffer containing 5.0 mM EDTA, pH 7.4, and mixed with their respective harvested nonadherent cells prior to assays. PMA (1.5 and 15 nM) alone failed to induce HL-60 to either express FPR or FPR tran- script (Fig. 10). Since PMA exerts its effects by binding and activating PKC (16, 17), we were interested to determine if PKC could modify the effects of PKA activation on HL-60 expression of FPR and FPR transcripts, Consequently, HL- 60 cells were incubated with dbcAMP (0.2 mM) for 24 and 48 h either in the absence or presence of increasing concentra- tions of PMA, and their ability to express FPR and FPR transcript was determined. As expected, dbcAMP alone in-

24 48 24 48 Time (h)

FIG. 10. Effect of PMA on dbcAMP-induced expression of FPR by HL-60 cells. HL-60 cells were incubated (37 "C) with either PMA alone or dbcAMP + PMA. After 24 and 48 h, their ability to bind "'I-FP and express FPR transcript was determined. A , specific binding of '"I-FP by HL-60 cells incubated with either dbcAMP (0.2 mM) alone (U), PMA alone (A-A) (1.5 nM) (W) (15 nM), dbcAMP (0.2 mM) + PMA (1.5 nM) (W), or dbcAMP (0.2 mM) + PMA (15 nM) (A-A). B, FPR mRNA levels from experi- ments depicted in A.

duced expression of both FPR and FPR transcript in a time- dependent fashion (Fig. 10). Co-incubation of HL-60 cells with dbcAMP (0.2 mM) and low concentrations of PMA (1.5 nM) enhanced dbcAMP-induced expression of FPR and FPR transcript significantly (Fig. 10). High concentrations of PMA (15 nM) inhibited both expression of FPR and FPR transcript (Fig. 10). These effects were not observed when the inactive PMA analog, 4a-phorbol 12,13-didecanoate was used. These results indicate that a cross-talk mechanism exists whereby PKC can regulate PKA-mediated transcription of the FPR gene.

DISCUSSION

Results presented here indicate that dbcAMP induces expression of FPR gene (and receptor) through activation of PKA type I. While studying dbcAMP analogs we also found that 8-(4-chlorophenylthio) cAMP (Cpt CAMP) induced expression of FPR (and transcript) by HL-60 cells, by itself. Cpt cAMP also potentiated the effect of Mnb CAMP. These results were confusing since 8-thio analogs (such as Cpt CAMP) have been reported to activate PKA type I1 (12). Recently, however, Tasken et al. (18) reported that Cpt cAMP induces PKA RIa messages (8-fold, 20 h) in the HT-29 cell line. Indeed, we have obtained similar results using HL-60 cells. These cells contain low levels of PKA RIa and RIIa mRNA. Low concentrations of dbcAMP (0.05 mM) or PMA (1.5 nM) induce a rapid (maximal 16-24 h) induction of PKA RIa (8-10-fold) mRNA without affecting the levels of PKA RIIa mRNA.2 Thus, it is possible that Cpt cAMP induces PKA type I in HL-60 cells and allows for the potentiation of Mnb dbcAMP. Interestingly, the same investigators reported that the phorbol ester 12-0-tetradecanoylphorbol 13-acetate (3.0 nM) also induces PKA RIa mRNA in HT-29 cells but had no effect on either PKA RIB, RIIa, and RIIB or Ca and Cp mRNA levels (18). This effect of 12-0-tetradecanoylphor- bo1 13-acetate was blocked by staurosporine (18). Similar effects on HL-60 cells could provide an explanation for the enhancing effect of low concentrations of PMA on dbcAMP- induced expression of FPR and FPR transcript. The effects of dbcAMP and forskolin on FPR gene expression were in- hibited in a concentration-dependent fashion by staurospor- ine, suggesting that a PKA-mediated phosphorylation event is required for FPR gene expression. The transcription-en- hancing properties of dbcAMP are mediated by distinct pro- moter elements within the target gene termed dbcAMP re-

' H. D. Perez, E. Kelly, and R. Holmes, manuscript in preparation.

Formyl Peptide Receptor Gene Expression by HL-60 Cells 363

sponse elements (CREs) (19). CREs contain sequences that are recognized by specific transcription factors known as CRE-binding proteins (CREBs). CREBs contain consensus sequences that are potential phosphorylation sites for PKA (ZO), and CREB’s phosphorylation is required for binding of CREBs to CREs (21). Thus, it is likely that a specific CREB is phosphorylated in HL-60 cells by a PKA type I, and this event is necessary for gene transcription. Indeed, we have cloned several CREB-like cDNAs from these cells. Rat CREB also has a PKC phosphorylation sequence (ZO), whereas hu- man placental CREB does not (22). We still do not know if HL-60 CREBs have a potential phosphorylation sequence for PKC. If so, it is possible that dual phosphorylation of CREB (by PKA and PKC) could enhance its transcriptional activity. Another effect of PKC is to induce AP-1 activity (19, 23). AP-1 (jun/fos heterodimer and/or junljun homodimer) binds to a distinct promoter element (the 12-0-tetradecanoylphor- bo1 13-acetate-responsive element) within target genes (24). AP-l/jun can also bind to CRE (25) and can effectively transactivate the CREs of several dbcAMP-responsive genes (25). Furthermore, the AP-1 function on a CRE can be en- hanced by forskolin (25). Recently, it has been shown that dbcAMP can enhance the expression of jun and fos in HL-60 cells, in a transient manner (26). Thus, it is possible that CREBs and AP-1 cooperate in expression of the FPR gene. Another possibility is that HL-60 CREBs form a heterodimer with jun (which is up-regulated by PMA) and the complex transactivates a CRE but not a 12-0-tetradecanoylphorbol 13-acetate-responsive element (27). These possibilities are now under investigation.

Expression of FPR and FPR transcript were modulated also by RA in a specific fashion. RA inhibited dbcAMP- induced expression of FPR and FPR transcript by 50-70%. By itself, RA did not induce expression of FPR or its tran- script but induced HL-60 to differentiate into neutrophil-like cells and increased their expression of CD18 mRNA (7). Time course analysis of the effect of RA on dbcAMP induction of expression of FPR and FPR transcript suggests that it may have a dual effect. One, RA inhibits dbcAMP-induced tran- scription of the FPR gene. The second effect of RA could be post-transcriptional. Addition of RA to dbcAMP-induced HL- 60 cells (24 h, Fig. 7) resulted in significant diminution in expression of FPR, whereas FPR transcript (although in decreased levels) continued to increase. These results suggest that RA may also interfere with either a translational or posttranslational step required for expression of FPR. Fur- thermore, these results argued against the possibility that RA may increase degradation of FPR transcript. Experiments using actinomycin D provided direct evidence indicating that RA had no effect on FPR mRNA half-life. These experiments revealed, also, that FPR expression by HL-60 cells decreased during actinomycin-mediated suppression of FPR transcrip- tion. One possible explanation for this finding is that FPR is constitutively internalized in the absence of ligand. Although some of these receptors may recycle (28), others are degraded. Inhibition of FPR synthesis over time results in decreased FPR expression.

Dexamethasone increased the expression of FPR and FPR transcript by dbcAMP-induced HL-60 cells, whereas by itself dexamethasone failed to induce either expression of FPR or FPR transcript. Thus, it appears that regulation of FPR gene

expression is a complex phenomenon that includes the con- certed interaction of transcriptional enhancers and repressors under regulation of at least two different protein kinases (A and C). Furthermore, PKA and PKC cross-talk determines the rate of transcriptional activation. We have cloned the FPR gene: expressed FPR protein, and raised specific anti- bodies directed against FPR.’ These reagents would allow us to characterize further the mechanisms involved in expression of FPR and FPR transcript by HL-60 cells.

Acknowledgments-We would like to thank Dr. Dennis Hickstein, University of Washington, Seattle, for providing the cDNA probe for CD18 and Dr. Jackie D. Corbin, Vanderbilt University, for his helpful discussions regarding dbcAMP analogs.

REFERENCES 1. Perez, H. D., Kelly, E., Elfman, F., Armitage, G., and Winkler, J.

2. Anderson, R., and Niedel, J. (1984) J. Biol. Chem. 259, 13309 3. Polakis, P. G., Uhing, R. J., and Snyderman, R. (1988) J. Biol.

4. Lubbert, M., Herrmann, F., and Koeffler, H. P. (1991) Blood 77 ,

5. Vallari, D. S., Austinhirst, R., and Snyder, F. (1990) J. Bwl.

6. Skubitz, K. M., Zhen, Y. S., and August, J. T. (1982) Blood 59 ,

7. Hickstein, D. D., Hickey, M. J., and Collins, S. J. (1988) J. Biol.

8. Collins, S. J., Robertson, K. A,, and Mueller, L. (1990) Mol. Cell.

(1991) J. Clin. Invest. 87,971

Chem. 263,4969

909

Chem. 265,4261

586

Chem. 263,13863

Biol. . l o , 2154 9. Lobo. E.. Elfman. F.. Kellv. E.. and Perez. H. D. (1990) Biochem. . .

Biophys. Acta 1055 ,193 ’

10. Boulay, F., Tardif, M., Brouchon, L., and Vignais, P. (1990) Biochem. Biophys. Res. Commun. 168, 1103

11. Corbin, J. D., Keely, S. L., Soderling, T. R., and Park, C. R. (1975) Adv. Cyclic Nucleotide Res. 5,265

12. Beebe, S. J., Holloway, R., Rannels, S. R., and Corbin, J . D. (1984) J. Biol. Chem. 259,3539

13. Collins, S., Bouvier, M., Bolanowski, M. A., Caron, M. G., and Lefkowitz, R. J. (1989) Proc. Natl. Acad. Sci. U. S. A. 86,4853

14. Herbert, J. M., Seban, E., Maffrand, J . P. (1990) Biochem. Bio- phys. Res. Commun. 171 , 189

15. Tamaoki, T., Nomoto, H., Takahashi, I., Kato, Y., Morimoto, M., and Tomita, F. (1986) Biochem. Biophys. Res. Commun. 135, 397

16. Nishizuka, Y. (1984) Science 2 4 5 , 371 17. Blumberg, P. M. (1988) Cancer Res. 48, 1 18. Tasken, K., Kvale, D., Hansson, V., and Jahnsen, T. (1990)

19. Mitchell, P. J., and Tjian, R. (1989) Science 245 , 371 20. Gonzalex, G. A., Yamamoto, K. K., Fischer, W. H., Karr, D.,

Menzel, P., Biggs, W., 111, Vale, W. W., and Montminy, M. R. (1989) Nature 337 , 749

21. Yamamoto, K. K., Gonzales, G. A., Biggs, W. H., 111, and Mont- miny, M. R. (1988) Nature 334,494

22. Hoeffler, J. P., Meyer, T. E., Yun, Y., Jameson, J. L., and Habener, J. F. (1988) Science 2 4 2 , 1430

23. Curran, T., and Franza, B. R. (1988) Cell 55,395 24. Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R. J., Rahms-

dorf, H. J., Jonat, C., Herrlich, P., and Karin, M. (1987) Cell 49 , 729

25. Sassone-Corsi, P., Ransone, L. J., and Verma, I. M. (1990) Oncogene 5,427

26. Nakamura, T., Datta, R., Sherman, M. L., and Kufe, D. (1990) J. Biol. Chem. 265,22011

27. Ivashkiv, L. B., Liou, H.-C., Kara, C. J., Lamph, W. W., Verma, I. M., and Glimcher, L. H. (1990) Mol. Cell. Biol. 10, 1609

28. Perez, H. D., Elfman, F., Marder, S., and Ives, H. (1989) J. Clin. Invest. 8 3 , 1963

Biochem. Biophys. Res. Commun. 172,409