IL-3-dependent leukotriene synthesis

Eur. J. Immunol. 1992. 22: 2907-2913

Martin Krieger, Vinzenz von TscharnerA and Clemens A. Dahinden Institute of Clinical Immunology, Inselspital and Theodor Kocher InstituteA, University of Bern, Bern

2907

Signal transduction for interleukin-3-dependent leukotriene synthesis in normal human basophils: opposing role of tyrosine kinase and protein kinase C* The intracellular signaling pathways regulating the synthesis of leukotrienes by myeloid cells are largely unknown. In addition, the signal transduction mechanisms utilized by the cytokine receptor family are still poorly understood. The fact that in mature human basophils the synthesis of leukotriene C4 (LTC4) induced by C5a is strictly dependent on a short preincubation with the cytokine interleukin-3 (IL-3), allowed us to investigate the metabolic requirements for LTC4 synthesis, and also to provide some information on early signal transduction mechanisms of IL-3 in these differentiated, non-dividing blood leukocytes. IL-3 itself does not alter intracellular free calcium concentration ([Ca2+Ii)in basophils, whereas C5a induces a transient rise independent of IL-3 pretreatment, indicating that the priming effect of IL-3 cannot be explained by alterations in [Ca2+]ichanges. The protein kinase C inhibitor staurosporine did not inhibit C5a-induced histamine release nor IL-3-dependent LTC4 formation in contrast to the IgE receptordependent basophil response. Activation of protein kinase C (PKC) by phorbo1-12-myristate-13-acetate(PMA) induced histamine release without leukotriene formation. PMA-treated basophils did not produce LTC4 in response to C5a. Rather, PMA blocked the IL-3 effect on C5a-induced LTC4 synthesis. Only the C5a signal but not the IL-3 effect was pertussis toxin sensitive. Two unrelated tyrosine kinase inhibitors, tyrphostin RG-50864 and herbimycin A , were both very efficient blockers of IL-3-dependent lipid mediator formation whereas C5a-induced histamine release was preserved. Thus LTC4 formation does not require activation of a staurosporine-sensitive serinehhreonine kinase. To the contrary, IL-3-dependent LTC4 formation appears to be regulated by serinelthreonine and tyrosine phosphorylation in an antagonistic manner.

1 Introduction Leukotrienes are potent pro-inflammatory mediators formed by oxidation of free arachidonic acid via the 5-lipoxygenase pathway [1, 21. They are formed mainly by different myeloid cell types which produce them in large amounts in response to calcium ionophore treatment. In neutrophils, LTB3 is the major primary product [3] whereas leukotriene C4 (LTC4) is mainly synthesized by eosinophils and basophils [2,4]. The facts that calcium ionophores are potent inducers of leukotriene synthesis in all cell types studied, and that protein kinase C (PKC) activation by phorbol esters has been shown to lower the calcium ionophore requirement in neutrophils, led to the concept that the level of intracellular free calcium and PKC activity may regulate leukotriene formation [5]. However, the definition of physiological triggers of leukotriene synthesis has been difficult. In particular, chemotactic granulocyte

agonists acting through G protein-coupled receptors, such as the complement-derived C5a [6], which are known to increase intracellular calcium concentration ([Ca2+],) and activate PKC through the generation of inositol-1,4,5trisphosphate (IP3) and diacylglycerol by phospholipase C, are generally ineffective triggers of leukotriene synthesis [7,81.

Studies from several laboratories have shown that exposure of different mature granulocytic effector cell types to appropriate hematopoietic growth factors renders the cells capable of producing leukotrienes in response to various soluble agonists [8-121. Particularly high quantities of LTC4 can be produced by human basophils as compared to other mature granulocytes. Activation of human basophils by cross-linking of the high-affinity receptor for IgE (FcERI) leads to the release of preformed mediators by degranulation as well as de novo synthesis of lipid mediators. By contrast, C5a is a potent degranulating agent but does not promote LTC4 synthesis, indicating that the release of these [I 107651 two types of mediators is regulated differently. After a short preincubation of 5 to 10 min with the hematopoietic growth * This work was supported by the Swiss national Science Founda- factor IL-3, IL-5 or granulocyte macrophage colonytion, Grant No. 3132470.91. stimulating factor (GM-CSF), however, basophils strongly respond with LTC4 synthesis to a C5a challenge [lo, 11,131. Correspondence: Clemens A. Dahinden, M.D., Institute of ClinThis experimental model is particularly attractive for ical Immunology, Inselspital, CH-3010 Bern, Switzerland dissecting the mechanisms for degranulation and LTC4 Abbreviations: LTC4: Leukotriene C4 or (55 6R)-5-hydroxy- formation since two different sequential signals can be 6-S-glutathionyl-7,9-trans-12,14-cis-eicosatetraenoic acid [Caz+li: analyzed separately, and since LTC4 production is depenIntracellular free calcium concentration fura-2/AM: [fura-2-pen- dent on IL-3 in an all-or-nothing fashion. In addition to takis-(acetoxymethyl)ester] providing information about the regulation of lipid media0 VCH Verlagsgesellschaft mbH. D-6940 Weinheim, 1002

0014-2980/92/1111-2907$3.50+ .25/0

2908

Eur. J. Immunol. 1992. 22: 2907-2913

M. Krieger,V v. Tscharner and C. A. Dahinden

tor synthesis, pharmacological and biochemical studies should also give insight into the signal transduction pathway of IL-3. IL-3, IL-5 and GM-CSF all interact with specific members of a novel family of receptors, the cytokine-erythropoietin receptor superfamily [14-181. The signal transduction mechanisms for this receptor family are poorly understood, partly because the classical biological responses (such as cell growth) occur late after initial ligand-receptor interaction. To our knowledge, IL-3-dependent LTC4 formation in human basophils represents by far the most rapid effect on cellular function to this group of cytokines. This facilitates the interpretation of data obtained by using drugs affecting different metabolic pathways. The results presented here indicate that IL-3dependent LTC4 formation by mature human basophils is regulated by tyrosine and serinejthreonine phosphorylation events in an antagonistic manner.

2 Materials and methods 2.1 Materials

Commercial reagents were purchased as follows: Hepes was from Calbiochem-Behring Corp. (La Jolla, CA), EDTA, fura-2/AM, PMA and staurosporine were from Fluka AG (Buchs, Switzerland), ionomycin was from Sigma (St. Louis, MO), pertussis toxin and its oligomer were from List Biological Laboratories (Campbell, CA) ,dextran T70, Ficoll-Hypaque and Percoll were from Pharmacia Fine Chemicals (Uppsala, Sweden), BSA fatty acid-freehow endotoxin content was from Boehringer Mannheim (Mannheim, FRG), Dynabeads M450 PanT and PanB were from Dynal, A. S. (Skoyen, Norway). 2.2 Cell stimuli and inhibitors

Cell purity was assessed by staining cytocentrifuge slides. Contaminating cells were small lymphocytes and monoctes that do neither release LTC4 nor histamine and do not influence the response to IL-3 and C5a [11]. Cell viability was checked with the trypan blue exclusion test (> 95% viable cells with the presented data). For [Ca2+Iideterminations, basophils were further enriched by discontinuous Percoll density gradient centrifugation and negative selection using immunomagnetic beads (Dynabeads PanT and PanB) as described in [25], resulting in basophil preparations of 85%-90% purity, contaminating cells consisting mainly of small lymphocytes. 2.4 Mediator release and inhibition assays Cells were suspended in HACM buffer (20mM Hepes, 125 mM NaC1, 5 mM KC1, 0.5 mM glucose, 0.025% BSA, 1mM CaC12 and 1m M MgC12) at a cell density of 2.5 X lo65.0 x lo6cells/ml. All release experiments were performed in a shaking water bath at 37 "C. After warming up the cells for 10 min at 37 "C, inhibitor or control buffer was added for the times indicated, followed by IL-3 or control buffer, and 10 min later C5a, anti-IgE or anti-FceRI antibodies. Twenty minutes after addition of the triggering agents, a time period previously found to be sufficient for complete release reaction to all agonists studied [13], tubes were placed on ice. Cells were pelleted by centrifugation (400 x g, 10 min, 4 "C), 0.5 ml of the supernatant was mixed with an equal volume of HACM buffer and 1ml HC104 0.8 M, centrifuged again (600 x g, 15 min, 4 "C) and stored at - 20 "C up to 2 days or it was used directly for histamine determination using an automated fluorimetric method [26]. The remaining supernatant was used without further treatment for sulfido-leukotriene determination by RIA as described elsewhere [27], using a mAb with equal reactivity towards LTC4, LTD4 and LTE4.

C5a was purified from yeast-activated human serum as described [19]. Recombinant human IL-3 (bioactivity 0.7 x 106U/mg, determined in a proliferation assay with leukocyte cultures from patients with chronic myeloid leukemia; endotoxin contamination of < 0.5 EU/mg according to the limulus test) was a kind gift of Sandoz (Basel, Switzerland). Anti-human IgE mAb LE27 was purified as described by C. S. Hong et al. [20]. The mAb 29C6 directed against the a chain of the high affinity receptor for IgE (anti-FceRI, Kd = 3.2nM) was a generous gift of Dr. Hakimi and Dr. Chizzonite, Hoffmann-La Roche (Nutley, NJ) [21]. The synthetic tyrosine kinase inhibitor tryphostin R G 50864 [22, 231 was kindly provided by Dr. Zilberstein of Rorer Biotechnology Inc. (King of Prussia, PA) and the benzoquinoid ansamycin antibiotic Herbimycin A [24] was a kind gift of Dr.Y, Uehara, National Institute of Health (Tokyo, Japan).

Highly purified basophils were loaded with 0.3 nmol [fura2-pentakis-(acetoxymethyl)-ester] (fura-2/AM) per lo6 cells in HACM buffer for 30 min at 37 "C. After a short centrifugation (5 min, 150 x g at room temperature) the freshly loaded cells were resuspended in prewarmed HACM buffer at a concentration of 0.5 x 106 cells/ml. Fura-2 fluorescence changes (excitation wavelength 340 nm, emission wavelength > 490 nm) of the cell suspension in response to cell agonists were continuously monitored at 0.25-s intervals and analyzed as described earlier [28]. Each measurement was standardized by adding ionomycin (5 pM final concentration) leading to 100% fura2-saturation and subsequent quenching of the fluorescence with MnClz (1 mM final concentration).

2.3 Blood donors and cell isolation

2.6 Data presentation

After informed consent was given, basophils were isolated from venous blood of unselected, healthy volunteers. The cell preparations were performed exactly as described [lo]. For mediator release studies, mononuclear cells depleted of neutrophils and eosinophils by Ficoll-Hypaque centrifugation containing 1%-8% (mean 3%) basophils were used.

Histamine release as a marker of degranulation is expressed as % of total histamine content of cells as determined in cell lysates. Leukotriene production is presented as pg LTC4/1000 basophils. Total histamine of cell lysates was used to calculate basophil numbers using 1pg histamine/cell[29]. All experiments were performed in triplicates

2.5 Determination of [Ca2+]i

Eur. J. Immunol. 1992. 22: 2907-2913

IL-3-dependent leukotriene synthesis

and repeated at least three times. Under the same experimental conditions, the pattern of mediator release was identical using cells from different donors, except that the total amount of histamine and LTC4 release varied. Thus, representative experiments are shown for each condition. SEM of triplicates for histamine release were < 2% and < 10% for LTC4 production.

2909

However, histamine release induced by C5a was not reduced by staurosporine but slightly increased. The data also demontrate that the priming effect of IL-3 for LTC4 synthesis is not suppressed by staurosporine and that LTC4 generation can proceed without the activation of a staurosporine-sensitive serinekhreonine kinase. 3.3 Effect of PKC activation by PMA

3 Results 3.1 IL-3 does not affect cytosolic free [Ca2+]i Receptor-dependent basophil degranulation is known to require the presence of extracellular calcium [30]. We previously found that IL-3 priming of basophils for C5ainduced LTC4 formation is independent of extracellular calcium [13].To examine whether IL-3 affects [Ca2+]iwhich may be independently regulated from extracellular calcium uptake, we loaded highly purified human basophils with the calcium indicator fura-2/AM and monitored fluorescence changes in response to IL-3. Fig. 1shows that no transient rise in [Ca2+],was observed after IL-3 addition in contrast to the response to the secretagogue C5a. Fig. 1 also demonstrates that the changes in [Ca2+]iobserved in response to C5a were similar with or without IL-3 preincubation, indicating that the G protein-coupled mechanisms of calcium mobilization are not affected by IL-3. 3.2 Effect of PKC inhibition by staurosporine Basophil histamine release induced by cross-linking of the FcsRI has been shown to involve activation of PKC [31]. Consistent with these results,we find that histamine as well as leukotriene release in response to anti-IgE antibodies is inhibited by treatment with staurosporine regardless of whether the cells have been exposed to IL-3 or not (Fig. 2).

+

C5a

The experiments above indicate that PKC activation is neither involved in IL-3 signaling in human basophils nor required for the formation of leukotrienes. To address this issue further we examined whether PKC activation by phorbol ester can mimick the IL-3. Previous studies by other groups have shown that phorbol ester can induce basophil degranulation, a response independent of extracellular calcium in contrast to the calcium requirement of receptor-dependent secretagogues [31, 321. Consistent with previous data, Fig. 3 shows that phorbol ester induces histamine release, and that this release is further enhanced by subsequent C5a stimulation. However, PMA itself does not induce the formation of LTC4. In contrast to IL-3, PMA does not prime basophils for C5a-induced LTC4 formation. To the contrary, the IL-3-dependent LTC4 synthesis in response to C5a is abolished by treatment of basophils with PMA at concentrations above 10 ng/ml (Fig. 3a and b). Thus, activation of PKC by phorbol ester appears to suppress rather than to mimick the IL-3 effect. The interpretation that phorbol ester inhibits IL-3 priming for LTC4 synthesis through activation of PKC is supported by the fact that the inhibitory effect of PMA can be reversed by staurosporine (Fig. 3b).

'

z

n

? IL-3 ? control

1 40--

' 2 min.

Figure 1. Intracellular calcium mobilization in human basophils. [Caz+],changes are expressed as changes in percent saturation of the fura-2-probe. The resting level before stimulation corresponded to 51 ? 0.6% (mean k SEM, n = 8) fura-2 saturation. Basophils were enriched from peripheral blood to 85% purity by Percoll density centrifugation and negative selection with magnetic beads as described in Sect. 2.3. Measurements were performed with 0.5 x lo6 cells. Top, comparison of the C5a (5 x ~ O - * M ) induced [Ca2+], response in basophils pretreated for 8 min with 50 ng/ml IL-3 (left) or buffer control (right) No [Ca2+],response was observed with 50nglml IL-3 alone or buffer control (bottom).

"

contr. C5a

IL-3

+

C5a

1

algE IL-3

+

algE

Figure 2. IgE-dependent and IgE-independent mediator release in basophils: effect of the PKC inhibitor staurosporine. Mononuclear cells containing basophils were sequentially stimulated with M) or or without IL-3 (10 nglml) for 10 min, followed by C5a the IgE-cross-linking mAb LE27 (100 nglml) for another 20 min. White columns show controls (0.1 vol% DMSO), black columns show the mediator release observed in cells treated with 100 nM staurosporine 3 min before addition of IL-3 or control buffer.

Eur. J. Immunol. 1992. 22: 2907-2913

M. Krieger,V v. Tscharner and C. A. Dahinden

2910

B

A 100 I

80 60 40

20 0

:i, 10

0 5

0

.1

control

C5a

IL-3

1

+

PMA

!. 5 i

C5a

40 20

0 n

control

C5a

Figure 3. The effect of PKC activation by PMA upon mediator release by basophils. (A) PMA at the indicated concentrationswas added 3 rnin before adding IL-3 (10 ng/ml), followed 10 rnin later by CSa M). (B) Cells were exposed M), with (IL-3 + to buffer control or C5a CSa) or without (C5a) preincubation with IL-3 (10 ng/ml) for 10 min. Black columns: cells pretreated for 3 min with PMA (20ng/ml) before the addition of IL-3 or control buffer; hatched columns: cells pretreated for 3 rnin with staurosporine (100 nM), followed by 3 min with PMA (20 ng/ml); white columns: controls without drugs.

IL-3

+

C5a

aF&I

IL-3

+

aF&l

Figure 4. Effect of pertussis toxin on mediator release by human hasophils. Cells were triggered by the FceRI-cross-linking mAb 29C6 (100 ng/ml) or CSa M). Where indicated, cells were preincubatcd with IL-3 (10 nglml) for 10 min. Buffer control (white columns), pertussis toxin at 1 nM (hatched coumns) or S n M (dotted columns), or the fl oligomer of pertussis toxin at 5 nM (black columns) was addcd YO rnin before the experiment.

3.4 Role of pertussis toxin-sensitive G proteins

With the exception of IgE receptor stimulation, agonists for basophil degranulation are, as far as they have been examined, coupled to pertussis toxin-sensitive G proteins [6, 331. As expected, C5a-induced histamine release is indeed suppressed by preincubation with pertussis toxin but not by the b oligomer of pertussis toxin which cannot ribosylate G proteins. A similar inhibition by pertussis

toxin is seen in IL-3-pretreated basophils which fail to release both histamine and LTC4 in response to C5a.These results indicate that IL-3 priming does not render the C5a-induced signaling pathway pertussis toxin resistant. To examine whether the IL-3 signal itself may use a pertussis toxin-sensitive pathway,we used anti-IgE receptor antibodies as a second signal. IgE-dependent mediator release is not suppressed by pertussis toxin, as compared to the oligomer control [33], (Fig. 4). Since IgE receptor crosslinking by itself leads to leukotriene synthesis, we can also conclude that the regulation of LTC4 synthesis is not affected directly by pertussis toxin. Finally, the IL-3 signal does not seem to involve pertussis toxin-sensitive G proteins since the enhancement of IgE-dependent mediator release by IL-3 treatment is preserved.

3.5 Tyrosine kinase inhibitors block IL-3-dependent LTC4 synthesis The recently cloned receptors for IL-3, IL-5 and GM-CSF belong to an emerging superfamily of cytokine receptors [14, 16-18]. The intracellular domain of all these receptors do not show any consensus sequences known to be involved in signal transduction processes. For IL-3, IL-5 and GMCSF receptors, a common fl chain has been characterized very recently which is required for high-affinity binding and probably signaling [15,16].This fl chain belongs to the same cytokine receptor superfamily and again shows no intracellular consensus sequence of known functions, in particular no tyrosine kinase motif. Accordingly, the signaling pathways utilized by this receptor family has remained elusive. In growth factor-deprived, factor-dependent cell lines, rechallenge with interleukins induces tyrosine phosphorylation of certain intracellular proteins [34-371 indicating that the cytokine receptors may be linked to certain intracellular tyrosine kinase(s). It was, therefore, interesting to test whether the IL-3 induced rapid alteration of the cellular function in non-dividing, terminally differentiated basophils also involves tyrosine phosphorylation.

IL-3-dependent leukotriene synthesis

Eur. J. Immunol. 1992. 22: 2907-2913

293 1

B

A

' looT" 7 40 20

20

0

20/l

30T

I

'Ot

W

15t

Figure 5. Inhibition of IL-3-dependent lipid mediator synthesis by tyrosine kinase inhibitors. (A) Cells were treated for 90 min with different concentrations of the synthetic tyrosine kinase inhibitor RG-50864before stimulating cells with IL-3 (10 ng/ml) and 10 rnin later M). DMSO controls (0.1 ~ 0 1 % ) with C5a do not affect either release (not shown). (B) the Streptomyces sp. - derived antibiotic herbimycin A was added at the indicated concentrations 100 min before adding buffer control (white M C5a (hatched columns), or columns) or 90 min before IL-3 (10 ng/ml), followed by C5a (lo-* M) 10 min later (black columns).

' I

Tyrphostin RG-50864 (kM)

8 40 Herbirnycin A (FM)

Fig. 5 shows that indeed IL-3-dependent LTC4 synthesis is suppressed in a dose-dependent way by tyrosine kinase inhibitors. It is important to note that two different, structurally unrelated, inhibitors which probably suppress tyrosine kinase activity by different mechanisms [38-40] are both effective and that the complex cellular response leading to C5a-induced basophil degranulation is still intact in the presence of the drugs (Fig. 5). These observations suggest that LTC4 formation in this experimental model is regulated by tyrosine phosphorylation events and that the drugs are not toxic and not interfering with cellular function in a nonspecific manner.

4 Discussion Stimulation of human basophils by cross-linking of the high affinity IgE receptor leads to the release of preformed mediators by degranulation and de novo synthesis of the lipid meditor LTC4. Despite the fact that recent studies have proposed tyrosine phosphorylation as an early event in signal tansduction [41-431 and that interesting analogies to T cell activation by the T cell receptor have been found [4446], the complicated series of intracellular signal transduction events finally leading to mediator release in human basophils are difficult to dissect. Using the IgEindependent agonist CSa, it can be demonstrated that degranulation and de rzovo synthesis of lipid meditors are differently regulated, since leukotriene synthesis in contrast to histamine release is observed only when C5a is given as a second signal shortly after an appropriate hematopoietic growth factor, such as IL-3 [lo].The hypothesis that IL-3 and CSa must act in concert through different signal transduction pathways is supported by the fact that only sequential stimulation by IL-3 and C5a leads to lipid mediator formation,while simultaneous addition of the two peptides or stimulation in reversed order does not [47]. Indeed, CSa induces a transient rise in [Ca2+],in human basophils, typical for a G protein-coupled chemotactic factor receptor as previously reported in neutrophils and

eosinophils [48, 491. No change in [Ca2+]i, however, is observed after IL-3 addition. Thus, IL-3 receptor activation in mature human basophils is not coupled to a phosphatidylinositol4,5-bisphosphate (PIPz)-specific phospholipase C-activating G protein or to a phosphorylation-sensitive phospholipase G y, as are the IgE receptor [43] and certain growth factor receptors in other cell types [50]. Interestingly, IL-3 does not alter the C5a-induced [Ca2+]ichanges, at least within the precision of measurements obtained with the relatively diluted cell suspensions of this rare leukocyte type, indicating that the priming effect of IL-3 cannot be explained simply by alterations in the regulation of [Ca2+]i. Serinelthreonine phosphorylation is an important pathway of intracellular signaling events for a large variety of cellular functions. Indeed, the PKC inhibitor staurosporine inhibits IgE-dependent histamine release by basophils [31], and as shown here, LTC4 synthesis.This inhibitory effect could not be overcome by IL-3 priming. Furthermore, activation of PKC by phorbol ester results in histamine release consistent with previous studies [31,32]. However, it does not directly stimulate lipid mediator formation. The fact that staurosporine does not inhibit but rather enhances basophil degranulation induced by C5a indicates that histamine release can occur in the absence of PKC activation, although, by analogy to studies in other cell types such as neutrophils, this G protein-coupled receptor presumably also activates PKC in basophils. Thus, PKC activation through G protein-coupled receptors in basophils may rather have an inhibitory effect on histamine release. The IL-3-dependent LTC4 synthesis induced by C5a was clearly independent of a staurosporine-sensitive protein kinase C. Thus, activation of the metabolic machinery for LTC4 production [phospholipase(s) and 5-lipoxygenase] does not require PKC activation, in contrast to what has been suspected from earlier studies demonstrating an enhancement by phorbol ester of the calcium ionophoreinduced leukotriene synthesis in neutrophils [5]. The data also show that the IL-3 signal modifying qualitatively the

2912

M. Krieger,V. v. Tscharner and C. A. Dahinden

basophil response to C5a does not require PKC activation. Consistent with this conclusion is the observation that activation of PKC by phorbol ester does not mimic the IL-3 effect on basophils. To the contrary, serinekhreonine phosphorylation through PKC totally abolished IL-3dependent LTC4 production triggered by C5a without preventing the G protein-coupled degranulation response. The fact that this effect can be reversed by staurosporine shows that PMA inhibits IL-3 priming by activation of PKC rather than by a nonspecific phenomenon and again indicates that IL-3 signaling pathways and LTC4 production in human basophils are independent of PKC activation. Consistent with previous studies in other cell types (reviewed in [51]), and consistent with the primary structure of the recently cloned human C5a receptor homologous to other G protein-coupled chemotactic factor receptors [6], degranulation induced by C5a was pertussis toxin sensitive. Similarly, CSa-induced LTC4 production in IL-3-primed basophils was also inhibited by pertussis toxin, and therefore, such experiments did not allow to decide whether IL-3 receptor activation also involves pertussis toxin-sensitive G proteins. Since basophil activation by IgE is also enhanced by IL-3 preincubation, albeit in a quantitative rather than qualitative fashion, experiments were performed using anti-lgE receptor antibodies as a trigger.These experiments showed that basophil mediator release was enhanced by IL-3 in both control and pertussis toxin-treated cells. Thus, only the CSa but not the IL-3 signal seems pertussis toxin sensitive.

Eur. J. Immunol. 1992. 22: 2907-2913

and of the tyrosine kinase inhibitors on the other hand are qualitatively distinct, (c) two different tyrosine inhibitors presumably acting through entirely different mechanisms [38-40] show identical results, (d) only rapidly occurring cellular functions during short-term incubations are studied, and (e) two different cell functions are analyzed (LTC4 production and histamine release), which were clearly influcenced differently by these drugs, making nonspecific or toxic drug effects unlikely. However, the data presented here do not allow to decide whether IL-3 by itself activates tyrosine kinase(s) or whether IL-3 couples the C5a response to a tyrosine kinase pathway by a yet-unknown mechanism. Although direct biochemical analysis of intracellular metabolic events in mature basophils is very difficult due to the low cell number of this rare leukocyte type, we attempted to demonstrate tyrosine phosphorylation in human basophils in response to IL-3 or IL-3 and C5a by immunoblotting using monoclonal and polyclonal antiphosphotyrosine antibodies with different detection systems. No inducible phosphotyrosine proteins could be detected in basophil extracts in contrast to strongly positive controls of extracts of growth factor dependent cell lines. Nevertheless, such negative results do not speak against an induction of tyrosine phosphorylation, since immunoblotting may be of insufficient sensitivity for the detection of phosphorylation of functionally important proteins, particularly in this terminally differentiated, non-dividing cell type. Thus, further studies are needed to investigate whether a regulatory protein for synthesis of lipid mediators is activated by tyrosine phosphorylation in basophils.

Tyrosine kinases have recently received increased attention a \ important regulators of gene expression and cell growth [S?]. Recent studies in FcERI-bearing rodent cell lines have also indicated an early involvement of tyrosine kinase activation in IgE receptor-dependent cell stimulation [41-431. Our study shows that IL-3-dependent LTC4 synthesis can be blocked in a dose-dependent way by two structurally unrelated inhibitors of tyrosine kinase activity. Thus, tyrosine kinases must be considered not only as regulators of late cell functions such as cell growth, but also as regulators of rapidly occuring cellular functions such as the formation of inflammatory mediators. In particular, tyrosine kinase activation may represent an important and necessary step for the regulation of leukotriene generation. The hypothesis that IL-3 primes basophils by activation of (a) tyrosine kinase(s) is consistent with the demonstration of tyrosine phosphorylation in IL-3-dependent cell lines after rechallenge of growth factor-starved cultures [34-371. Furthermore, nerve growth factor fi primes human basophils for CSa-induced LTC4 synthesis similar to IL-3 1271. The high-affinity binding component of the nerve growth factor (NGF) receptor has recently been defined as the product of the proto-oncogene trk 1531which belongs to the tyrosine kinase receptor family. The identical biological effect of IL-3 and NGF on human basophils implicates that tyrosine kinase(s) activated by these distinct receptors may share functionally important substrates.

Most ligands interacting with the cytokine receptor superfamily seem to use similar signal transduction mechanisms since the factor dependence of cell lines can be changed by transfection of the appropriate receptors 1541. Nevertheless, the signal transduction mechanisms for these interleukins and hematopoietic growth factors are still poorly defined although most known signaling mechanisms have been postulated to be involved, such as tyrosine phosphorylation for IL-2, IL-3 and GM-CSF 134-37, 551, serinekreonine phosphorylation for IL-2 and IL-3 [36,56,57], IP3 and [Ca2+],increase for IL-4 and IL-6 [SS,591, p21ras activation for IL-2, IL-3 and GM-CSF [60], hydrolysis of an inositol containing glycolipid for IL-2 1611, pertussis toxinsensitive G proteins for IL-1 and IL-3 162,631and activation of tyrosine phosphatase for IL-4 [64]. Some of these conflicting results may be due the predominant use of different cell lines which facilitate biochemical studies. The response of cell lines may, however, not always reflect that of untransformed, normal human cells. Our study indicates that at least with regards to the rapid functional alteration of human basophils to IL-3, some of the proposed signaling mechanisms are rather unlikely, namely: activation of PKC, pertussis toxin-sensitive G proteins, activation of phospholipase C PLC y or other PIPz-specific PLC generating diacylglycerol and IP3, and intracellular [Ca2+],increase.

However, the specificity of different inhibitors towards separate protein kinases is not absolute [38, 391, and nonspecific drug effects must be considered. Nevertheless, the interpretation of our data is facilitated by the facts that (a) the drugs act at relatively low concentrations, (b) the effects of the PKC inhibitor staurosporine on the one hand

Studies from several laboratories suggest that priming of effector cells is important in inflammation and allergic reactions. Interference at the level of effector cell priming may provide a novel approach of pharmacological interventions. Our study indicates an opposing role of serinekreonine and tyrosine phosphorylation for IL-3 priming of basophils. Therefore, inhibitors of tyrosine kinases or

Eur. J. Immunol. 1992. 22: 2907-2913 activators of PKC may have a potential as anti-inflammatory therapeutic agents. We thank Drs. J. Hakimi, R. Chizzonite, E: Ueharaand M. Schreier for generous gift of reugents; 7: Brunner, S. C. Bischoff and A . L . de Weck for critical reading of the manuscript and f? Winkler and J. Zingg for expert technical assistance.

Received June 26. 1992.

5 References 1 Samuelsson, B., DahlCn, S.-E., Lindgren, J.-A., Rouzer, C. A. and Serhan, C. N., Science 1987. 237: 1171. 2 Lewis, R. A . , Austen, K. F. and Soberman, R. J., N. Engl. J. Med. 1990. 323: 645. 3 Borgeat, l? and Samuelsson, B., J. Biol. Chem. 1979.254: 7865. 4 Owen,W. F. Jr., Soberman, R. J.,Yoshimoto,T., Sheffer, A. L., Lewis, R. A . and Austen, K. F., J. Immunol. 1987. 138: 532. 5 Liles,W. C., Meier, K. E. and Henderson,W. R., J. Immunol. 1987. 138: 3396. 6 Gerard, N. P. and Gerard. C., Nature 1991. 349: 614. 7 Clancy, R. M., Dahinden, C. A. and Hugli, T. E., Proc. Natl. Acad. Sci. USA 1983. 80: 7200. 8 Dahinden. C. A., Zingg, J., Maly, F. E. and de Weck, A. L., J. Exp. Med. 1988. 167: 1281. 9 DiPersio, J. F., Billing, P., Williams, R . and Gasson, J. C., J. Immunol. 1988. 140: 4315. 10 Kurimoto,Y., de Weck, A. L. and Dahinden, C. A., J. Exp. Med. 1989. 170: 467. 11 Bischoff, S. C.. Brunner,T., de Weck, A. L. and Dahinden, C. A , , J. Exp. Med. 1990. 172: 1577. 12 Takafuji, S., Bischoff, S. C., de Weck, A. L. and Dahinden, C. A , , J. Immunol. 1991. 147: 3855. 13 Kurimoto,Y., de Weck, A. L. and Dahinden, C. A., Eur. J. Immunol. 1991. 21: 361. 14 Cosman, D., Lyman, S. D., Idzerda, R. L., Beckmann, M. l?, Park, L. S., Goodwin, R. G. and March, C. J., Trends Biochem. Sci. 1990. 15: 265. 15 Nicola, N. A. and Metcalf, D., Cell 1991. 67: 1. 16 Kitamura,T., Sato, N., Arai, K.-I. and Miyajima, A., Cell 1991. 66: 1165. 17 Tavernier, J., Devos, R., Cornelis, S., Tuypens, T., Van der Heyden, J.. Friers,W. and Plaetinck, G., Cell 1991. 66: 1175. 18 Murata,Y., Takaki, S., Migita, M., Kikuchi,Y., Tominaga, A. and Takatsu, K., J. Exp. Med. 1992. 175: 341. 19 Hugli, T. E., Gerard, C., Kawahara, M., Scheetz, M. E., 11, Barton, R., Briggs, S., Koppel, G. and Russell, S., Mol. Cell. Biochem. 1981. 41: 59. 20 Hong, C. S., Stadler, B. M.,Walti, M. and de Weck, A. L., J. Immunol. Methods 1986. 95: 195. 21 Riske, F., Hakimi, J., Mallamaci, M., Griffin, M., Pilson, B., Tobkes, N., Lin, €? , Danho,W., Kochan J. and Chizzonite, R., J. Biol. Chem. 1991. 266: 11245. 22 Yaish, P., Gazit, A , , Gilon, C., Levitzki, A., Science 1988.242: 933. 23 Lyall, R. M., Zilberstein, A., Gazit, A , , Gilon, C., Levitzki, A. and Schlessinger, J., J. Biol. Chem. 1989. 264: 14503. 24 Uehara,Y, Hori, M., Takeuchi, T. and Umezawa, H., Jpn. J. Cancer Res. 1985. 76: 672. 25 Brunner,T., de Weck, A. L. and Dahinden, C. A., J. Immunol. 1991. 147: 237. 26 Siraganian, R. P., Anal. Biochem. 1974. 57: 383. 27 Bischoff, S. C. and Dahinden, C. A., Blood 1992. 79: 2662. 28 Moser, B., Schumacher, C., von Tscharner,V , Clark-Lewis, I. and Baggiolini, M., J. Biol. Chem. 1991. 266: 10666. 29 Sorensen, L. S., Mullarkey, M. F., Bean, M. A , , Mochizuki, D. Y., Chi, E.Y. and Henderson, W. R., Int. Arch. Allergy Appl. Immunol. 1988. 86: 267.

IL-3-dependent leukotriene synthesis

2913

30 Grant, J. A., Lett-Brown, M. A., Warner, J. A., Plaut, M., Lichtenstein, L. M., Haak-Frendscho, M. and Kaplan, A. P., Fed. Proc. 1986. 45: 2653. 31 Warner, J. A. and MacGlashan, D. W. Jr., J. Immunol. 1990. 145: 1897. 32 Schleimer, R. P., Gillespie, E. and Lichtenstein, L. M., J. Immunol. 1981. 126: 570. 33 Warner, J. R., Yancey, K. B. and MacGlashan, D. W., Jr., J. Immunol. 1987. 139: 161. 34 Sorensen, l? H. B., Mui, A. L.-F., Murthy, S. C. and Krystal, G., Blood 1989. 73: 406. 35 Sorensen, l?, Mui, A. L.-F. and Krystal, G., J. Biol. Chem. 1989. 264: 19253. 36 Linnekin, D. and Farrar, W. L., Biochem. J. 1990. 271: 317. 37 Kanakura, Y., Druker, B., Cannistra, S. A., Furukawa, Y., Torimoto,Y. and Griffin, J. D., Blood 1990. 76: 706. 38 Levitzki, A., Gazit, A., Osherov, N., Posner, I. and Gilon, C., Methods Enzymol. 1991. 201: 347. 39 Uehara,Y. and Fukuzawa, H., Methods Enzymol. 1991. 201: 370. 40 Uehara, Y., Fukuzawa, H., Murakami, Y. and Mizuno, S., Biochem. Biophys. Rex Commun. 1989. 163: 803. 41 Benhamou, M., Gutkind, J. S., Robbins, K. C. and Siraganian, R. P., Proc. Natl. Acad. Sci. USA 1990. 87: 5327. 42 Yu, K.-T., Lyall, R., Jariwala, N., Zilberstein, A. and Heimovich, J., J. Biol. Chem. 1991. 266: 22564. 43 Park, D. J., Min, H. K. and Rhee, S. G., J. Biol. Chem. 1991. 266: 24237. 44 Orloff, D. G., Ra, C., Frank, S. J., Klausner, R. D. and Kinet, J. l?, Nature 1990. 347: 189. 45 Irving, B. A. and Weiss, A., Cell 1991. 64: 891. 46 Klausner, R. D. and Samelson, L. E . , Cell 1991. 64: 875. 47 Dahinden, C. A , , Bischoff, S. C., Brunner, T., Krieger, M., Takafuji, S. and de Weck, A. L., Int. Arch. Allergy Appl. Immunol. 1991. 94: 161. 48 vonTscharner,V ,Prod’hom, B., Baggiolini, M. and Reuter, H., Nature 1986. 324: 369. 49 Kernen, l?, Wymann, M. P., von Tscharner,V, Deranleau, D. A. ,Tai, P., Spry, J., Dahinden, C. A. and Baggiolini, M., J. Clin. Invest. 1991. 87: 2012. 50 Margolis, B., Rhee, S. G., Felder, S., Mervic, M., Lyall, R., Levitzki, A, Ullrich, A., Zilberstein, A. and Schlessinger, J., Cell 1989. 57: 1101. 51 Simon, M. I., Strathmann, M. P. and Gautam, N., Science 1991. 252: 802. 52 Cantley, L. C., Auger, K. R., Carpenter, C., Duckworth, B., Graziani, A., Kapeller, R. and Solthoff, S., Cell 1991. 64: 281. 53 Klein, R., Jing, S, Nanduri,V , O’Rourke, E. and Barbacid, M., Cell 1991. 65: 189. 54 Doi,T., Hatakeyama, M., Minamoto, S., Kono, T., Mori, H. and Taniguchi, T., Eur. J. Immunol. 1989. 19: 2375. 55 Sugamura, K.,Takeshita,T., Asao, H., Kumaki, S., Ohbo, K., Ohtani, K. and Nakamura, M., Lymphokine Rex 1990.9: 539. 56 Evans, S. W. and Farrar, W. L., J. Cell. Biochem. 1987. 34: 47. 57 Mufson, R. A., Szabo, J. and Eckert, D., J. Immunol. 1992. 148: 1129. 58 Ashida,T., Kubo, K., Kawabata, I., Katagiri, M., Ogimoto, M. and Yakura, H., Cell Immunol. 1990. 126: 233. 59 Fukunaga, M., Fujiwara,Y., Fujibayashi, M., Ochi, S.,Yokoyama, K., Ando, A., Hirano,T., Ueda, N. and Kamada,T., FEBS Lett. 1991. 285: 265. 60 Satoh, T., Nakafuku, M., Miyajima, A. and Kaziro,Y., Proc. Natl. Acad. Sci. USA 1991. 88: 3314. 61 Eardley, D. D. and Koshland, M. E., Science 1991.251: 78. 62 Rollins, l?, Witham, S., Ray, K., Thompson, N., Sadler, H., Smithers, N., Grenfell, S. and Solari, R., Cytokine 1991.3: 42. 63 Kelvin, D. J., Shreeve, M., McAuley, C., McLeod, D. L., Simard, G. and Connolly, J. A., J. Cell. Physiol. 1989.138: 273. 64 Mire-Sluis, A. R. and Thorpe, R., J. Biol. Chem. 1991. 266: 18113.