Inhibition of the calcium store-operated calcium entry pathway by ...

2 downloads 0 Views 996KB Size Report
by emptying with thapsigargin the Ca2+ stores of HL60 ... this store-operated Ca2+ entry pathway (SOCP), suggest- ...... cells at this stage of differentiation.
VOl. 268,No.36,Issue of December 25, PP . 26911-26919, 1993 Printed in U.S.A.

JOURNAL OF BIOLQCICAL CHEMISTRY 0 1993 by The American Society for Biochemistry and Molecular Biology, Inc. THE

Inhibition of the Calcium Store-operated Calcium Entry Pathway by Chemotactic Peptide and by Phorbol Ester Develops Gradually and Independently along Differentiationof HL60 Cells* (Received forpublication, May 19, 1993, and in revised form, July 26, 1993)

Mayte Montero,Javier Garcia-SanchoS, and JavierAlvarez From the Departamento de Bioquimica y Biologia Molecular y Fisiologia, Facultad de Medicina, Universidad de Valladolid, 47005-Valladolid, Spain

N-Formyl-methionyl-leucyl-phenylalanine (fMLP) in- brane Ca2+ pathway (5-7). We shall use the term store-operhibited transiently the entry of Ca2+ and Mn2+ inducedated Ca2+ entry pathway (SOCP)2 to refer to this Ca2+ entry by emptying with thapsigargin the Ca2+stores of HL60 mechanism (8). cells differentiated toward granulocytes. Phorbol12,13- We have recently reported that,before stimulating Ca2+ endibutyrate (PDB) produced a permanent inhibition of try, fMLP, but not platelet-activating factor, induces a short this store-operated Ca2+ entry pathway (SOCP), suggestphase of inhibition of SOCP in human neutrophils(8).Inhibiing that inhibition was due to protein phosphorylation tion of SOCP by fMLP was maximal within 10 s and then PDB gradually decreased with a half-time of about 1min. Only then, mediated by protein kinase C (PKC). Inhibition by was prevented by the PKC inhibitors staurosporin and once this inhibition is relieved, the well-known fMLP-induced chelerythrine. Inhibition byfMLP was prevented by stimulation of Ca2+ entry becomes apparent. Regarding the chelerythrine but only partially by staurosporin. The mechanisms, we showed that a verysimilar inhibition in terms characteristics of the inhibition were similar to those of intensity, [Ca2+Ji dependence, and time needed to develop M., Alvarez,J., reported in human neutrophils (Montero, and Garcia-Sancho, J. (1!4!33)J. Biol. Chern. 268, 13066 could be obtained with phorbol 12,13-dibutyrate (PDB). Inhi13061). Neither fMLPnorPDB inhibited significantly bition by PDB was, however, long-lasting, as expected for a SOCP in undifferentiated HL60 cells. Single-cell [Ca2+li nonmetabolizable protein kinase C (PKC)activator. These findby fMLP could be mediated measurements at different stages of differentiation ings suggested that inhibition showed that inhibitionby fMLP and PDBdeveloped in- through activation of PKC, leading to phosphorylation of the SOCP protein itself or of a SOCP regulatory protein (8). The dependently, suggesting differentinhibitorymechanisms. The simplest explanation would be that inhibi- only important discrepancy was the higher sensitivity of the tion by fMLPtakes place throughactivation of a protein inhibition by PDB to the PKC inhibitor staurosporin, ascompared to the inhibition by fMLP. This difference could be atkinase distinct fromPKC and that the PDB-activated PKC isoform necessary to phosphorylateandinhibit tributed to thedifferent mechanism of PKC activation in each SOCP is expressed only along differentiation. Addition-case. ally, inhibition by bothfMLP and PDBdeveloped gradu- We have studied here the development of the inhibition of ally. At intermediate stages of differentiation,PDB was SOCP by fMLP and PDB during differentiation of the human able to produce a partial and maintained inhibition and promyelocytic leukemia cell line HL60 toward granulocytes. fMLP a partial and short-lived inhibition of SOCP. HL60 cells differentiated withMezSO for 7 days behaved similarly to neutrophils with respect to the inhibitionof SOCP by fMLP and PDB. In undifferentiated HL60 cells, however, neiAgonists such as chemotactic peptide N-formyl-methionylther fMLP nor PDB was able to inhibit SOCP. The lack of effect leucyl- phenylalanine(fMLP)l or platelet-activating factor of fMLP on undifferentiated cells was expected as these cells increase the cytosolic Ca2+ concentration ([Ca2+li) in human develop high-affinity receptors to fMLP only during differeninositol tiation (see“Discussion”),but thelack of inhibition by PDB was neutrophils (1-5). Thisincrease is initiallydueto 1,4,5-trisphosphate-mediatedCa2+mobilization fromthe intra- surprising. HL60 cells have been shownto possess PKC activity cellular calcium stores, which is followed by an increased Ca2+ (9) although this activity increases along differentiation (10). entry from the extracellular medium through the plasma mem- On the other hand, HL60 cells are able to respond t o phorbol brane. The mechanism for the Ca2+ entry is still unknown, esters, which induce differentiation towardmonocytes (11, 12). although there is strongevidence in neutrophils and othercell We find that, at intermediate stages of differentiation, some types suggesting that the emptying of the Ca2+ stores signals cells responded only to fMLP or only to PDB, and some reby some unknown mechanism the opening of a plasma mem- sponded partially toboth in an additive way. This suggests that inhibition of SOCP by fMLP is not mediated by PKC, but prob*This investigation was supported by Grant PB92-0268 from the Spanish Direccibn General de Investigacih Cientifica y TBcnica. The ably by a different Ca2+-sensitive protein kinase and that incosts of publication of this article were defrayed in part by the payment hibition by PDB occurs through activationof a PKC isoform not of page charges. This article must thereforebe hereby marked “aduer- present in undifferentiated HL60 cells but expressed during tisement” in accordance with 18 U.S.C. Section 1734 solely t o indicate this fact. $ To whom correspondence and reprint requests should beaddressed. *The name SOCP was chosen by analogy tovoltage-operated Tel.: 34-83-423084;Fax: 34-83-423085. (VOCC), receptor-operated (ROCC), or second messenger-operated The abbreviations used are: fMLP, N-formyl-methionyl-leucyl-phe- (SMOCC) CaZ+ channels. AlthoughSOCPcouldbe a subclass of nylalanine; [Ca2+l,,cytosolic free calcium concentration; platelet-acti- SMOCC, we want to stress with this name that this Ca2+entry pathway vating factor, l-O-alkyl-2-acetyl-sn-glycero-3-phosphorylcholine; PDB, opens as a consequence of the emptying of the intracellular calcium phorbol 12,13-dibutyrate;SOCP, store-operated Ca2+ entry pathway; stores rather than by the action of a second messenger generated diPKC, protein kinase C; Me2S0, dimethyl sulfoxide. rectly on interaction of an agonist with its membrane receptor.

26911

26912

Inhibition of Ca2+Entry by fMLP and Phorbol Ester

differentiation. On the other hand,inhibition by fMLP or PDB progresses gradually instead of all-or-none at the single-cell level suggesting that, as differentiation goes on, more SOCP proteins are susceptible to inhibition by either fMLP or PDB.

RESULTS

We have firstcharacterized the inhibition by fMLP andPDB of the SOCP in HL60 cells differentiated for 7 days with Me2S0. Differentiated HL60 cells were treated with the endomembrane Ca2+-ATPaseinhibitor thapsigargin(18)for 10 min MATERIALSANDMETHODS prior to fMLP or PDB addition in order to deplete completely Cell Culture-HL6O human leukemia cells were obtained from Dr. the Ca2+stores and to activate SOCP. Since the fMLP-induced Faustino Mollinedo, Centro de Investigaciones Biologicas del Consejo release of Ca2+ from the stores cannot take place with this SuperiordeInvestigacionesCientificas,Madrid,Spain.Theywere experimental design, the inhibition of Ca2+ entry by the chepropagated and cultured as described elsewhere (13). HL60 cells were motactic peptide can be evidenced most clearly. differentiated toward granulocytes by culture in medium containing The left part in Fig. 1 (truces A to F ) illustrates theeffects of 1.3% dimethyl sulfoxide (14) during2-7 days.3 fMLP on Ca2+ entry through SOCP, measured as the [Ca2+li Measurements oflCa2+l,-All the measurements were performed in fura-2-loaded cells. For experiments with cell suspensions, the cells increase induced by addition of Ca2+ to thapsigargin-treated as Ca2+(truce B ) were collected and washedtwice with standard medium of the following cells. fMLP wasadded either at the same time composition (in mM): CaCl, 0.2; NaCl, 145; KCl, 5; MgCI, 1;glucose, 10; or preceding Ca2+by different time periods (C-F). When fMLP Na-HEPES, 10,pH, 7.4. Then they were resuspendedat 2% cytocrit in and Ca2+ were addedsimultaneously, there was some Ca2+ the same medium, and fura-WAM (final concentration, 4p ~ was ) added entry during the first few seconds followed bynet Ca2+ exitand from a 2 mM stock in dimethyl sulfoxide. After 30-45 min at room a subsequent slow increase of [Ca2+Ii(truce B ) . Complete inhitemperature, the cells were washed twice with standard medium in bition of Ca2+ entry was obtained when fMLP preceded Ca2+ order to remove extracellular fura-2, resuspendedat 1%cytocrit in the same medium, and storedroom at temperature until used. Fluorescenceaddition by 10 s (truce C). This inhibition was transient as evidenced by the slow increase of [Ca2+libeginning about 30 s measurements were performed in 0.5-ml aliquots of the cell suspension addition. When Ca2+was delayed 60 s with regard to and under magnetic stirring in a fluorescence spectrophotometer con- after Ca2+ structed by Cairn Research Ltd. (Newnham, Sittingbourne, Kent, UK), fMLP (truce E ) , the rate of Ca2+ entry, as estimated from the which allows quasisimultaneous(30-300 Hz) measurementsof fluorescence excited at up to six different wavelengths. The emission wavelength was set at 530 -c 10 nm, and readings were integrated a t I-s intervals. [Ca2+li was estimated from the ratio of the fluorescences excited at 340 and 380 nm (&0/3,0, Ref. 15). Measurements were performed a t 37 "C.4 For single-cell measurements, HL60 cells wereloaded with fura-2 as described above. The cells were then allowed t o attach to polylysinecoated glass coverslips mounted underthe microscope (Nikon Diaphot) in a chamber thermostatted a t 36 "C for 10 min. Cells were then perfused at 3 mumin with medium containing 0.5 mM Ca2+ and epi-illuminated alternately at340 and 380 nm. Light emitted above 520 nm was recorded by an extended ISIS-M camera(Photonic Science, Robertbridge, East Sussex, UK) and analyzed using Applied an Imaging Magical image processor (Sunderland, U K ) with 32-megabyte video RAM. Thirty-two video frames of each wavelength were averaged by hardware, withan overall time resolution of about 6s for each pairof images at alternate wavelengths. The resulting images were ratioed(340/380) pixel by pixel to produce ratio images, and [Ca2+li was calculated by interpolation into a look-up table constructed by imaging fura-2 solutions with known Ca2+ concentrations using the same microscope objective. Additions of thapsigargin, fMLP, or PDB were made directly into the incubation chamber. Before the first addition (thapsigargin), perfusion was stopped and 50 pl of a 3-fold concentrated solution of thapsigargin prepared in the same perfusion medium was added over the chamber containing about 100 pl of medium (final concentration0.5 PM). Subsequent additions of fMLP (final concentration 1 p ~ or) PDB (final concentration 100 nM) were performed similarly. Chemicals-Fura-B/AM was obtained from Molecular Probes. fMLP, PDB, andstaurosporinwereobtainedfromSigma,Madrid,Spain. Thapsigargin was obtained from Calbiochem. Chelerythrine was obtained from Alomone Laboratories, Jerusalem, Israel. Other chemicals were obtained eitherfrom Sigma or from Merck, Darmstadt, Germany.

400

0

:

'

B

2

0

30

0

60

90

1

1 0

Time (s)

0

E 4001

4200 0 O c - l '

04-3-2-1

"

0

1

:

:,

2 3

Time (rnin)

FIG.1. Time course of the inhibition by fMLP of the [Ca*+li increase (let3panels) and the M n z + entry (right panel) in thapExpression of formyl-peptide receptors along differentiation was monitored by measuring the maximum height of the [Ca2+l, increase sigargin-treated differentiatedHL60 cells. HL60 cells differentiloaded with fura-2 and treated with following addition of 1 p~ fMLP. The values, normalized to 100% with ated with Me2S0 for 7 days were 0.2 m~ (left regard to the [Ca2+], obtained after 7 days differentiation, were (mean * 0.5 p~ thapsigargin for 10 min in medium containing either S.E.; n = 3-6): nondifferentiated, 3.5-c 0.2; 1 day, 22 ? 1; 2 days, 48? 2; panels) or 0.5 m~ (right panel) Ca2+. In the leftpanels, the [Caz+li was monitored and,a t t = 0 in thefigure, 0.5 m~ Ca2+ was added either alone 3 days, 66 ? 7; 4 days, 94 f 7. It has been reported that compartmentation of the dye within or- (panel A), simultaneously with 1 p~ fMLP (panelB ) or 10 s ( p a d C), 30 s (panel D),1 min (panelE ) , or 3 min (panel F ) aRer the addition ganella may cause serious artifacts in [Ca2+],measurements (16). We of 1 p~ fMLP, as shown by the arrows. In the right panel, F3M)was did not find evidence suggesting fura-2 compartmentation in the HL60 cells used here: (i) resting [Ca2+li levels (before thapsigargin addition) monitored to follow quenching of fura-2 fluorescence by Mn2+and, at t = 0 in the figure, 0.2 mM Mn2+ was added either alone (empty circles), were below 100 nM (not shown here, but see Ref. 17);(ii) on plasma simultaneously with 1 p~ fMLP (filled circles), or 10 s (empty trimembrane permeabilization with low concentrations of digitonin, unable to permeabilize endomembranes, more than 95% of the loaded dye angles), 30 s (filled triangles), 1 min (empty squares), or 3 min (filled of 1 p~ fMLP. Data were normalized to 100% was rapidly quenchedby added Ni2+;(iii) no indication of nonhomoge- squares) after the addition of three similar neousfluorescencedistributionwas observed in single-cell imaging F360 at t = 0. The experiments shown are representative ones. measurements in fura-2-loadedcells.

Inhibition of ea2+ Entry

by fMLP and

[Ca2+Iirise, was aboutone-half of that observed in the control with no fMLP (trace A). Three minutes after the addition of fMLP (truce F ) , Ca2+ entry had returned near the control values. These results suggest that inhibition of Ca2+ entry by fMLP is completed within 10 s and then reverses with a halftime of about 1 min. The rcght part of Fig. 1 shows the effects of fMLP on Mn2+entry in thapsigargin-treated cells. As before, fMLP wasadded to thecells at different times before Mn2+.The same transientinhibition reversing with a half-time of about 1 min was also found. Therefore, fMLP similarly prevented the stimulation of Ca2+ andMn2+ entry induced by thapsigargin, suggesting that, as documented previously in human neutrophils (8), both cations are able to pass through the SOCP. Fig. 2 shows that PDB was also able to inhibit both Ca2+ (upper punel)and Mn2+(lowerpanel)entry throughSOCP. The inhibition by PDB differed from the one effected by fMLP in being sustained, a t least within 5 min after addition of PDB (results not shown).The timeneeded for full inhibition was, as for fMLP, about 10 s, as evidenced by the timeneeded to inhibit Mn2+entry maximally when PDB and Mn2+were added simultaneously (curve B in the lower panel of Fig. 2 ) . The fMLPconcentration needed t o produce 50%inhibition of Mn2+entry (measured30 s after Mn2+addition) was near 1nM, which produced in two experiments 49% and 52% of the maximum inhibition (obtained with 1 p~ fMLP). 10 nM fMLP produced in two experiments 84% and 93% of the maximum inhibition. The PDB concentration needed to produce 50% inhibition was between 0.2 nM (which produced in two experiments 22% and 37% of the maximum inhibition, obtained with 100 nM PDB) and 1nM (which produced in two experiments 53% and 69% of maximum inhibition). Fig. 3 illustrates that, as reported previously for human neutrophils (81, the inhibition of Mn2+entry by fMLP and PDB is dependent on the[Ca2+li.Intheseexperiments, cells were

Phorbol Ester [Ca2f]i

204 0

20

=

26913

40 nM

[Ca2+]i = 420 nM

I

40

0

20

40

I

60

Time (s) Time (s) d ' entry in thapFIG.3.Effect of [Ca2+lion the inhibition of M sigargin-treated differentiated HLBO cells by fMLP and PDB. HL60 cells differentiatedwith MezSO for 7 days were loaded with fura-2 and treated with 0.5 thapsigargin for 10 min either in Ca2+free medium (containing50 PM EGTA, left panel) or in medium containing 0.5 m Ca2+(right panel). The [Ca2+l, by the endof the thapsigargin treatment is shown at top of each panel. Then, at t = 0 in thefigure, 0.25 mM Mn2+was added either alone (empty circles),simultaneously with 1 fMLP (filledcircles), or simultaneouslywith 100 nM PDB (triangles). Data were normalized to 100% F36nat t = 0.The experiments shown are representative of three similar ones.

treated with thapsigargin for 10 min either in Ca2+-free medium (containing50 J ~ MEGTA) or in medium containing0.5 mM Ca2+. The [Ca2+Ii values measured at the end of this period were (mean 2 S.D.) 40 2 5 nM ( n = 10) and 420 2 60 nM ( n = 231, respectively. At this time, either Mn2+ alone or Mn2+together with fMLP orPDB was added. It is apparent that inhibition the by both fMLP and PDB is stronger and fasterin the cells with the higher [Ca2+Ii(right panel). Inhibition by fMLP, calculated from the first 30 s after Mn2+ addition, was 32 -t 4% (mean 2 S.D., n = 4) in thecells treated with thapsigargin in the Ca2+free medium and 68 2 7% (mean 2 S.D., n = 3) in the cells treated in the Ca2+-containing medium. The inhibition by PDB was 1 2 2% (mean 2 S.D., n = 3) in the cells treated with thapsigargin in the Ca2+-free medium and 50 f 4% (mean 2 S.D., n = 3) inthose treated in theCa2+-containing medium.It seems clear that theinhibition by PDB was more sensitive to \5 I 400 decreasing [Ca2+li than the inhibition by fMLP, which was never entirely abolished in the low-[Ca2+licells. As reported previously for human neutrophils(B), staurosporin, a potent nonspecific PKC inhibitor (19, 20), antagonized the inhibition of SOCP in differentiated HL60 cells by both 01 : I -60 60 120 180 PDB and fMLP, but the sensitivity was very different. Fig. 4 shows that theinhibition of Ca2+(panel A ) or Mn2+(panel C ) Time (s) entry by PDB was largely prevented by 100 nM staurosporin. The inhibition of Ca2+entry wasprevented by 89 2 13%(mean -t S.D., n = 4, measured 2 min after Ca2+ addition), and the 1004. I inhibition of Mn2+ entry was prevented by 93 -t 7% (mean 2 S.D., n = 3, measured 30 s after Mn2+addition). The halfmaximal effect, measured from the [Ca2+lilevel obtained 2 min after Ca2+ addition, was obtained with 2 0 4 0 nM staurosporin (20 m staurosporin prevented the inhibition of Ca2+ entryby 40.32 2 11%,mean 2 S.D., n = 3, and 40 nM staurosporin by 62 2 20 E % , mean 2 S.D., n = 3). In contrast, theinhibition of Ca2+ and 30 60 90 120 Mn2+entry by fMLP wasonly slightly affected by 100 nM stauTime (s) rosporin (Fig. 4, panels B and D).The values estimated in FIG.2. Effect of PDB on CaZ' (upper panel)and M n 2 + (lower several similar experiments were (mean 2 S.D.) 23 2 1%for panel) entry in differentiated HL60 cells. HL60 cells differentiated Ca2+ entry ( n = 4) and 20 2 2% for Mn2+ entry ( n = 3), both 0.5 m with MezSO for 7 days were loaded with fura-2 and treated with entries were measured 30 s after theaddition of either Ca2+or thapsigargin for 10 min in medium containing either 0.2 mM Ca2+(upper panel)or 0.5 m Ca2+(lowerpanel).In theupperpanel, the [Ca2+li Mn2+. In order to obtain further evidence for the involvement of was monitored and, a t t = 0 in thefigure, 0.5 m Ca2+ was addedeither alone (trace A ) , simultaneously with 100 nM PDB (trace B ) , or 1 min phosphorylation in theinhibition of SOCP by PDB and fMLP, after adding 100 I ~ MPDB (trace C). In thelower panel, F360was moniwe have tested theeffects of chelerythrine. This drug has been tored and, at t = 0 in the figure, 0.2 mM Mn2+was added either alone reported to be a potent and specific inhibitor of PKC (21). Fig. (trace A ) , simultaneously with 100 nM PDB (trace B ) , or 1 min after chelerythrine prevented the adding 100 nM PDB (trace C). Data were normalized to 100%F36nat 5 shows that treatment with 5 t = 0. Data are representativeof four similar experiments. inhibition of SOCP by both fMLP andPDB. ICso of this drugfor A

6oov :

3

Inhibition of Ca2' Entry by fMLP and Phorbol Ester

26914

FIG.4. Effect of staurosporin on the inhibition by fMLF' (right panela) and PDB (leftpuneZs)of the Caa+entry (upperpanels)and the M n a + entry (lower panels) in thapsigargintreated differentiated HL60 cells. HL60 cells differentiated with MezSO for 7 days wereloaded with fura-2 and treated with 0.5 p~ thapsigargin for 10 min either in medium containing 0.2 m~ Ca2+(upperpanels)or 0.5 rn Ca2+(lower panels). In the upper panels, the [Ca2+Ji was monitored and, at t = 0 in the figure, either 0.5 rn Ca2+(Control), 0.5 rn Ca2+ + 1p~ fMLP ( f M L P ) ,0.5 rn Ca2++ 1p~ fMLP after a 5-min preincubation with 100 m staurosporin (st+fMLP), 0.5 m~ Ca2++ 100 I"PDB (PDB),or 0.5 rn Ca2+ + 100 m PDB after a5-min preincubation with either 100 m staurosporin (stIOO+PDB), 40 IIM staurosporin (stlO+PDB), or 20 I" staurosporin (st20+PDB)were added. In the lowerpanels, F360was monitored, and the same additions (except those with low staurosporin concentrations) were made except the last addition of 0.5 m~ Ca2+which was replaced by the addition of 0.2 m~ Mn2+. Ca2+ Staurosporin (100m) did not modify or Mn2+entry in thecontrols (not shown). This experiment is representative of three similar ones.

120

60

0

-80

120

60

180

Time (s)

Time (s)

60"

0

10

20

30

Time (s)

Control

I

I 0

-60

40

50

0

10

20

30

Time

40

50

60

(5)

plasma membrane permeability to this water-soluble drug. In order to investigate further themechanisms of the inhibition of SOCP by fMLP andPDB, we have studied its appearance andevolution during differentiation of HL60 cells. Fig. 6, left, shows the inhibition of Mn2+entry in thapsigargin-treated cells produced by either fMLP, PDB, or both together, at different stages of differentiation. No inhibition was observed in undifferentiated cells by either fMLP or PDB (A), but inhibition appeared and progressed with differentiation (B-E). It is also apparent that the inhibitory effects of PDB and fMLPwere additive at intermediate differentiation stages (B-D until the o l : : : : : : ! : : : : : : l time at which one or both of them produced near maximum -90 -60 -30 0 30 60 90 -90 -60 -30 0 30 60 90 120 inhibition, which was thecase of fully differentiated cells ( E ) .A similar pattern was observed when the Ca2+ entry through Time ( s ) Time ( s ) SOCP was studiedalong differentiation. The entryof Ca2+was FIG.5. Effect of chelerythrine on the inhibition of Caa+ entry estimated from the [Ca2+li increase on Ca2+ addition to thapby fMLP and byPDB in thapsigargin-treated differentiated HL80 cells. HL60 cells differentiated with MezSO for 7 days were sigargin-treated cells. Fig. 6, right, shows that, in undifferenloaded with fura-2 and treated with 0.5 p~ thapsigargin for 10 min in tiated cells (A), fMLP had no effect and PDB produced only a medium containing 0.1 m~ Ca2+.Then, 0.5 rn Ca2+was added a t t = 0 minor decreasein the[Ca2+liincrease. Inhibition appeared and in the figure (shown by the arrow). PDB (100m) or fMLP (1 p ~ was ) progressed with differentiation (B-E). Additivity between the added 1 min or 10 s before the CaZ+addition, respectively. In the experiments shown in panel B , chelerythrine (5 p)was added 20 min effects of fMLP and PDB was also clear in these experiments, before thapsigargin. This experiment is representative of three similar particularly at the beginning of differentiation ( B and C ) . It is ones. also apparent that the inhibition by PDB developed faster than that by fMLP. different PKC substrates hasbeen shown to be near 1 PMcomAdditivity was better seen when the decrease of [Ca2+liinpared to 100 PM for other kinases (21). The concentration used duced by addition of fMLP or PDB to cells treated with thapin Fig. 5, 5 p ~ should , ~ inhibit PKC by about 80% (21). Preinsigargin in Ca2+-containing medium was studied. In this excubation of the cells with chelerythrinefor 30 min was required perimental design, the cells were first incubated with to obtain the full inhibitory effect, probably because of poor thapsigargin in Ca2+-containing medium for 10 min. This resultsin a maintained increase of [Ca2+li reflecting a new This drug hasendogenous fluorescence which interferes with furs-2 steady-state with increased Ca2+ influx and Ca2+ extrusion. measurements, thus preventing the use of larger concentrations in our Inhibition of Ca2+ influx is revealed by a decrease of [Ca2+li. experiments. For [Ca2+licalculations in chelerythrine-treated cells, a Fig. 7 illustrates, using this experimentaldesign, the effect of calibration was performedwithin the same cell sample at the end of the consecutive additions of fMLP andPDB or vice versa, incells at experiment.

"_

Chelervthrine

I

I

26915

Inhibition of Ca2+Entry by fMLP and Phorbol Ester 1200-

A: Undif. 800

400

800

0 1200

800

400

800

400

0 1200

0 100 80

800 60

40

400

20

0

2010

30

40

Time (s)

50

60

-. -120

-60

0

60

120

160

Time (s)

FIG.6. Effects of fMLF' and PDB on the entryof lVbe* (kflpumds)and [Caz*],(right panels) in thapsigmgin-treatedm o cells in of differentiation.Either undifferentiated HL60 cells (panel A ) or HL60 cells differentiated with Me2S0 for 2 days (panel B ) , different stages 3 days (panel C), 4 days (panel D ) , or 7 days (panel E ) were loaded with fura-2 and treated with 0.5 PM thapsigargin for 10 min in medium containing either0.5 m~ (leftpanels)or 0.2 m~ Ca2+(rightpanels).In theleftpanels, F3- was monitored and, a t t = 0 in thefigure, either 0.2 mM Mn2+(empty circles), 0.2 mM Mn2+ 1 min after 100 I"PDB (filled circles),0.2 m~ Mn2+ 10 s after 1 p~ fMLP (empty triangles),or 0.2 mM Mn2+ after addition of both 100 nM PDB 1min before and 1 fMLP 10 s before (filled triangles)were added. In theright panels, [Ca2+Iiwas monitored and, a t t = 0 in thefigure, either 0.5 m~ Ca2+(empty circles),0.5 m~ Ca2+1min after 100 nM PDB (filled circles),0.5 m~ Ca2+10 s after 1 pbf fMLP (empty triangles), or 0.5 m~ Ca2+ after addition of both 100 nM PDB 1 min before and 1 PM fMLP 10 s before (filled triangles)were added. The arrows indicate the times that either PDB ( t = -60 s), fMLP ( t = -10 s ) , or Ca2+ ( t = 0) was added to the corresponding traces. Data are representative of three to five similar experiments.

The effects shown in Figs. 6 and 7 could have several expladifferent stages of differentiation. In undifferentiated cells (A), there wasno response to fMLP, and PDB induced only a small nations at the single-cell level. Additivity of the inhibitions by decrease of [Ca2+li.After 2 days of differentiation ( B ) , the ef- fMLP and PDB could be due to theexistence of two subpopufects of fMLP (transient) and PDB (maintained) were clearly lations of cells responding selectively and completely either to additive. This is indicated by the fact that fMLP produced fMLP or to PDB, both subpopulations overlappingby the end of about the same decrease of [Ca2+Iiwhen it was added either the differentiation period. Alternatively, most of the cells could before or afterPDB (Fig. 7 B ) .The additivity was still apparent be able torespond to both fMLP andPDB, the inhibition being after 3 and 4 days of differentiation, although to a smaller incomplete at the early differentiation stages and increasing degree (Fig. 7, C and D ) , suggesting that there is now some gradually along differentiation. To answer this question, we overlapping between the inhibitory effects of fMLP and PDB. performed single-cell [Ca2+li measurementsinexperiments Finally, after 7 days, no effect of fMLP wasobserved after PDB with the samedesign as those of Fig. 7, i.e. sequential additions (Fig. 7 E ) , suggesting that all thecells had responded fully to of fMLP and PDB or vice versa to thapsigargin-treated HL60 the PDB added previously. It is also apparent from this figure cells, either undifferentiated or after 2, 3, or 7 days of differthat theeffect of fMLP increases withdifferentiation not only entiation. in intensity butalso in duration. Table I shows the percentage of thapsigargin-treated HL60

26916

Inhibition of Ca2+ Entryby fMLP andPhorbol Ester k Undif.

fMLP

PDB

200

FIG.7. Effects of differentiation on the [Ca2+11 decrease inducedby fMLP and PDB in thapsigargin-treated HL60 cells. Either undifferentiated HL60 cells(panel A ) or HL60 cells differentiated with MeZSOfor 2 days (panel B ) , 3 days (panel C ) , 4 days (panel D ) , or 7 days (panel E ) were loaded with fura-2 and treated with 0.5 PM thapsigargin for 10 min in medium containing 0.5 m~ Ca2+. Then, [Ca2+l,was monitored and then either 1p~ fMLP followed by100 rn PDB (left traces) or 100 nM PDB followed by 1 PM fMLP (right traces) were added, as shown by the arrows. Duces are representative of three to five similar experiments.

t fMLP

PDB

fMLP C: 3 days

fMLP 200

PDB

PDB

200t 400 200 0

fMLP 0

PDB 60 180120

PDB 240

0

80 180120

240

Time (s) TABLE I Progression of the inhibition of SOCP by fMLP or PDB during differentiation of HL60 cells analyzed at the single cell level The figures represent the percentage of thapsigargin-treated cells which decreased ICa2+liin response to either fMLP or PDB, estimated from experiments similar tothose shown in Figs. 8-11. Cells which did not increase [Ca2+l,in response to thapsigargin (1525% depending on the experiment) were excluded fromthis analysis. n corresponds to the number of cells analyzed in every condition. See text for more details. Cell condition

Responding tofMLP

Responding to PDB

riding

tofxpor

to PDB

% total cell population

Undifferentiated ( n = 135) 2 days with MezSO ( n = 130) 3 days with MEzSO ( n = 236) 7 days with MEzSO ( n = 57)

2 22 75 97"

27 24

71 55

92

5

94"

0

a Differentiation increased not only the number of cells which responded but also the amplitude (and duration, in the case of fMLP) of the response in each responding cell. See text and Figs. 8-11 for details.

cells which were able to respond to fMLP or to PDB by reducing their [Ca2+Ii along differentiationwith Me2S0.After 2 days of differentiation, 55% of the cells still did not decrease [Ca2+Iiin response to eitherfMLP of PDB. Only 1% of the cells responded to both, so that about20% of the cells responded only to fMLP, whereas another 20% of the cells responded only to PDB (see

below). After 3 days of differentiation, most of the cells were already able to respond to both fMLP and PDB; however, the intensity of the inhibition of SOCP was notmaximum. It will be shown below that the increased degree of inhibition by fMLP and by PDB observed at later differentiation stages (panels D and E of Figs. 6 and 7) should be attributed mainly to a gradual increase in the potency of the inhibition in every single cell rather than to anincrease of the number of responding cells. Fig. 8 shows the records of [Ca2+lifor the average of all the cells (AVG) and for three representativesingle cells ( A X )in an experiment performed with nondifferentiated HL60 cells. fMLP produced no measurable effect (only 4 out of 135 cells responded to fMLP with a decrease of [Ca2+li), whereasPDB induced a small decrease of [Ca2+Iiin theaverage. This was due to the behavior of 27% of the cells, which responded to PDB similarly to the cells shown as B and C in the figure. The rest of the cells, which did not respond to fMLP or to PDB, are represented by the cell labeled A (see also Table I). Fig. 9 shows the average (AVG) and three representative single-cell responses ( A X ) in an experiment performed with HL60 cells differentiated for 2 days.About half of the cells did not respondto fMLP or to PDB (see Table I). Regarding the rest of the cells, 22% responded only to fMLP, as represented by cells A and B, and 24% responded to PDB, as shown by cell C.

26917

Inhibition of Ca2+Entry by fMLP andPhorbol Ester

AVG

AVG

A

A B

B

C

C

FIG.8. Effects of fMLP and PDB on [Ca2+li in undifferentiated thapsigargin-treatedHLBO single cells.Single-cell [Ca2+l,analysis was performedas described under “Materials and Methods.” Additionof 0.5 V M thapsigargin (labelled Tg),1 PM fMLP, and 100 nM PDB were made directly into the chamber as described under “Materials and Methods” at the times indicated by urrows in the figure. Perfusion was stopped throughout the experiment, so that solutions added were not washed away, similarly to the cell suspension experiments shown in Fig. 6. The top trace shows the average of all the cells present under a microscope field(AVG), excluding those which did not respond to thapsigargin increasing [Ca2+l,(about 15-25% in all of the single cell experiments). Duces A, B , and C correspond tothree representative single cells. For more details. see text and Table I.

Only 1%of the cells were found to respond additively to both fMLP and PDB. Fig. 10 shows the average (AVG) and three representative single-cell traces (A“) from two experiments performed in cells differentiated for 3 days in which either first fMLP and then PDB ( l e f t )or first PDB and then fMLP(right truces)were added. In theleft part of the figure, A and B are representative of cells which responded t o fMLP with a short transient decrease in[Ca2+liand thenresponded to PDB with a long-lasting [Ca2+li decrease(71%of the cells). C is representative of 18%of the cells, which did not respond to fMLP but responded to PDB. In the rzght part of the figure, we show three representative traces corresponding to thebehavior of 36% of cells, which first responded to PDB (to a different extent) and then responded also to fMLP with a further [Ca2+lidecrease. This demonstrates that the response to PDB and fMLPis additive in many cells at this stageof differentiation. In most of the rest of cells in this experiment(56%), the [Ca2+li decrease inducedby PDB was not followed by a further [Ca2+li decrease on addition of fMLP. This was due in manyof these cells to the fact that the PDB-induced [Ca2+li decreasewas too large to allow a further drop of [Ca2+li;when fMLP was added before PDB (asin theleft truces), it effectively reduced [Ca2+liin 75% of the cells at this stage (see also Table I). Fig. 11 illustrates, finally, the behavior of HL60 cells differentiated for 7 days with Me2S0. Essentially all of the cells

Tg

PDB fMLP

FIG.9. Effects of fMLP and PDB on [Ca2+lj in thapsigargintreated single HL60 cells differentiated for2 days with Me.SO. The figure shows the average (AVG) and the [Ca2+l,record from three representative single cells, labeledA, B , and C . The arrows indicate the addition of 0.5 p~ thapsigargin (Tg), 100 n~ PDB, and 1 PM fMLP. Other details are as in Fig. 8.

responded to both fMLP and PDB (see Table I). In most of the cells, the response to fMLP was transient and the [Ca2+li went back to the levels before fMLP addition within 3 min (before PDB addition), but the time interval necessary to do it was rather variable. We show three representative single cells in which fMLP reduced [Ca2+lifor progressively greater timeperiods. Cell C did not return to the previous [Ca2+Iilevels and remained with low [Ca2+Iiat the time of PDB addition. It is apparent that the inhibitory response to fMLPobserved in Fig. 11 in cells differentiated for 7 days is more intense and longlasting that thatshown in Fig. 10 in cells differentiated for 3 days. In similar experiments in which PDB was added first, additivity could no longer be observed, as the [Ca2+li levels decreased permanently to very low values before addition of fMLP (not shown). DISCUSSION

Inhibition of SOCP in HL60 cells by fMLP and PDB had in common its fast development (completed in