Enhancement of interleukin-3-dependent mast cell proliferation by suppression of c-jun expression.

We have previously shown that protein kinase C (PKC) depletion is associated with an increase in the proliferation of interleukin 3 (IL-3)-induced mast cells. Here we show that the AP-1 components c-Jun and c-Fos are induced by IL-3. While c-Jun's induction by IL-3 is totally dependent on PKC, c-Fos induction by IL-3 is only attenuated by PKC depletion. AP-1 binding activity was also induced by IL-3 but this induction was PKC independent. These results indicated a possible involvement of c-Jun in the inhibition of IL-3-induced growth regulation. A support for this assumption came from experiments in which an increase in thymidine incorporation into mast cells was noted when c-jun antisense oligomers were administered to IL-3-treated cells. Since the only known effect of direct inhibition of c-Jun on proliferation rates in several cellular systems was a reduction of proliferation, we verified that our c-jun antisense oligomer could also inhibit proliferation rates in fibroblasts where such a repression was previously reported. Thus c-Jun has an inhibitory effect on IL-3 induction of mast cells proliferation that is distinct from its role in several other cellular environments. These observations reveal the involvement of AP-1 and its components in IL-3-induced signal transduction and the importance of the mast cell environment in determining their specific cellular function.

Increased numbers of mast cells have been found in a variety of pathological situations, such as in the nasal mucosa of seasonal rhinitic patients, in acutely inflamed lesions of atopic dermatitis, in the conjunctiva in vernal conjunctivitis, in the skin in systemic mastocytosis, and in the lung associated with fibrotic disease (1). Furthermore, the abnormal proliferation of these cells may give rise to malignant systemic mastocytosis with few cases of true mast cell leukemia (2). Therefore, the intracellular events regulating mast cell proliferation are of considerable interest for the understanding and ultimate management of mast cell-associated disease states.
Interleukin-3, is a T-cell-derived growth factor, which has been extensively characterized in the mouse (3-7) and human (8) and is known to stimulate the proliferation and differentiation of a broad spectrum of hemopoietic cells, including pluripotential stem cells, mature megakaryocytes, macrophages, and mast cells. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solelv to indicate this fact.
$ To whom correspondence should be addressed: Institute of Biochemistry, Hebrew University-Hadassah Medical School, P. 0. Box 12272, Jerusalem 91120, Israel.
In the past few years we have extensively investigated the signal transduction pathways induced by this lymphokine in murine mast cells. Recently we produced evidence that PKCldepleted mast cells expressed accelerated rates of DNA synthesis in response to IL-3, revealing an inhibitory role played by PKC in IL-3-mediated mast cell proliferation (9). PKC is one of the major regulatory enzymes involved in the control of a wide variety of physiological processes including tumor promotion, membrane receptor function, differentiation, and proliferation (10). PKC has been found to play different regulatory roles, depending on the cell type and the stimulant used (11)(12)(13)(14).
A common response to extracellular signals involves changes in the spectrum and rates of gene expression, regulated by modulation of the activity of transcription factors such as AP-1 via activation of PKC (15)(16)(17)(18)(19). The DNA binding complex, AP-1, is a dimer that can be composed of two members of the jun family or by a combination of a jun family member and a fos family member. Furthermore, both fos and jun can dimerize with non-family members such as Fos-interacting protein (20) or CAMP response element-binding protein (21). The different AP-1 dimers containing JudJun or Fos/Jun can bind DNA with different affinities (22). The differential control of expression of each of the AP-1 components in different cell types or as a result of different exogenous stimuli can lead to specific cellular responses. These responses can also be influenced by other cellular factors in the cellular environment where they take place (23). Therefore the role ofAP-1 in IL-3 signal transduction can be different from its role in other cellular systems. Regulation ofAP-1 activity can occur at several levels including expression and degradation of its components and regulation of its DNA binding activity by phosphorylation (24)(25)(26). Thus we studied the influences of IL-3 on the mRNA steady state level and protein synthesis rates of AP-1 potential components c-Fos and c-Jun and AP-1 DNA binding activity in mast cells.
In order to understand the putative importance of PKC in transducing the IL-3 stimulus, we performed the same experiments in IL-3-induced PKC-depleted cells. We found that PKC depletion only attenuates most of the IL-3 induced c-fos mRNA accumulation and yet this treatment abrogates c-jun's mRNA accumulation. As PKC depletion leads to a large increase in mast cell proliferation, a n involvement of c-Jun in mast cell growth inhibition can be proposed. In order to clarify c-Jun's role in mast cell growth regulation we applied c-jun antisense oligodeoxynucleotides to IL-3 treated or non-treated cells. As envisaged, this treatment preferentially enhanced the IL-3induced mast cell proliferation.

MATERIALS AND METHODS
Cell Culture-IL-3-dependent mouse fetal liver-derived mast cells (MC-9) (27) were obtained from the American Type Culture Collection.
Bioassay F'H]Thymidine Incorporation-Staurosporine-treated and untreated MC-9 cells were washed three times with RPMI 1640 to remove traces of conditioned medium. They were then cultured in 200 pl of enriched medium in round-bottomed 96-well microtiter plates (Nunc, Denmark) a t a density of 50 x lo4 cells/well. IL-3 was added according to the requirements of each experiment. The cultures were incubated for 48 h a t 37 "C in 5% CO,. The cells were labeled with 1 pCi of ["Hlthymidine (ARC Inc., St. Louis, MO) for the last 4 h at 37 "C and transferred onto glass fiber filter paper (Titerek, United Kingdom) where they were water lysed and washed in an automated cell harvest unit (Brandel, Rockville, MD). The incorporation of radioactivity into cell DNA was then quantified by a scintillation counter. Cell number was determined using a Neubauer counter.
Chronic Phorbol Ester lkeatment of MC-9 Cells-Cells were incubated in enriched medium containing 20% WEHI-3 conditioned medium a t 37 "C in 5% CO,. PMA was added at a concentration of 20 ng/ml for a period of 72 h. After 72 h incubation, cells were pelleted and the supernatants were removed. Cells were lysed and processed as mentioned below.
Preparation of Cytosolic and Particulate Membrane Fractions-Five million MC-9 cells grown in enriched medium containing 20% WEHI-3 conditioned medium were treated with either 10 nM staurosporine for 1.5 h or 20 ng/ml PMA for 72 h prior to processing. Cells were removed into ice and then washed once by centrifugation at 500 x g in 4 ml of cold serum-free RPMI 1640 containing 1 M sucrose. The sucrose was added to the medium in order to pellet the viable cells at the end of each treatment (28). The supernatant of each sample was removed, and the cells in the pellet were lysed in 50 pl of ice-cold double-distilled water, by 2 cycles of freezing and thawing in liquid nitrogen, and immediately reconstituted with 450 pl of ice-cold extraction buffer (20 mM Tris-HC1, 50 p~ p-mercaptoethanol, 100 pg/ml aprotinin) (Sigma). Cytosol fractions were prepared by centrifugation at 80,000 x g for 60 min at 4 "C. The supernatants were collected for determination of cytosolic PKC activity. Membrane pellets were homogenized by 70 up and down strokes on ice in the course of 60 min with tight fitting homogenizers in 500 pl of ice-cold homogenization buffer (extraction buffer plus 0.1% Triton), followed by centrifugation a t 80,000 x g for 60 min. After checking the PKC activity in the original Triton-treated particulate fraction, we found that the Triton supernatant collected contained all the PKC activity associated with the membrane fraction.
PKC activity in the membrane fractions was then determined. One hundred percent cell lysis was confirmed by testing release of lactate dehydrogenase into the supernatants (29).
Cell-free PKC Assuy-PKC activity was determined essentially as previously described (30). Reaction mixtures contained 10 mM MgCI,, 1 mM CaCl,, 100 pg of histone (type 111-SI, with or without 20 pg of phosphatidylserine, and 5 pg of 1,2-diacylglycerol (Sigma) and 10 p~ ["PIATP (Amersham Carp.) in a final volume of 250 p1 of 20 mM Tris-HC1. Before addition to the reaction mixtures, phosphatidylserine and 1,2-diacylglycerol were prepared by evaporation under N, and sonication in 20 mM Tris-HCI, pH 7.5. Enzyme reactions were started by the addition of 50 pl of either cytosol or membrane fractions to the reaction mixtures, and the samples were incubated for 10 min at 30 "C.
Reactions were stopped by pipetting 100 pl from each sample onto a square (2.5 x 2.5 cm) filter paper (Whatman 3M chr, Whatman) which was then immediately washed with agitation in 200 ml of ice-cold 10% trichloroacetic acid for 10 min. This was followed by three 20-min agitated washes in 10% trichloroacetic acid a t room temperature and a fourth wash overnight. The papers were then soaked in 95% ethanol for 5 min followed by soaking in ether for an additional 5 min, and then allowed to air dry before counting.
Zl-3-mediated Actiuation-MC-9 cells were washed three times with RPMI 1640 and then resuspended in RPMI 1640 alone or containing 20 ng/ml PMAor 10 nM staurosporine, depending on each experiment. FCS was removed from the medium due to the finding that a serum-response element exists on the c-fos gene which renders c-fos inducible by serum (31). In the case of c-jun, serum can induce c-jun through the AP-1 binding site (32). This removal allowed us to assay the cells response specifically to each lymphokine. Cells were incubated for 3 h a t 37 "C in 5% CO, prior to stimulation with 100 unitsiml synthetic IL-3 (kindly provided by Ian-Clark Lewis, Biomedical Research Centre, Canada) for various periods of time and then processed for cytosolic RNAextraction.
Isolation ofCytosolic RNA (33)"Ten million cells per treatment were pelleted and lysed in lysis buffer, containing heparin, spermidine, and Nonidet P-40 (Sigma), centrifuged immediately to remove the nuclei, and processed by phenol and ch1oroform:isoamyl alcohol extraction. The RNA was precipitated overnight a t -20 "C in ethanol and sodium chloride, then washed in 70% ethanol, and the concentration determined in a spectrophotometer. Northern Blot Analysis (33)"For Northern blot analysis, samples of 10 pg of RNA derived from MC-9 cells were concentrated and denatured in a mixture of formaldehyde (BDH, United Kingdom) and formamide (Fluka) for 10 min at 65 "C. The samples were separated on an agarose/ MOPS gel (Sigma). The gel was blotted onto a Hybond-N filter ( h e rsham) and then the filter was exposed to UV for 4 min. Hybridization was carried out a t 42 "C for 24 h in a solution containing 6 x SSC, 50% formamide, 1% SDS, 5 x Denhardt's solution, 10 mM EDTA, 50 mM Tris, pH 8, and 20 ng of heat-denatured 32P-labeled DNAprobes. After washing once in 2 x SSC, 0.1% SDS at room temperature and once for 25 min at 56 "C, the papers were exposed at -70 "C to film (Kodak Curix RP2) for up to 2 days. The relative intensities of the bands on the autoradiograms were quantified by a light scanning densitometer.
The DNA Probes-A pc-fos specific probe was used for the analysis of its specific mRNAlevel on Northern blots. A 3.2-kilobase EcoRI-BamHI fragment was isolated from A c-fos (mouse)-2 clone (34). The c-jun probe was a 400-base pair oligonucleotide fragment derived from the 3' region of the mouse c-jun gene generated by the PCR method. The primers were (5' primer) 5'-ATG CCC TCAACG CCT CGT TCC TCC AGT-3' = residue 951, and (3' primer) 5'-CTG CTA CTG AGG CCA CCG CGG GAG CCA-3' = residue 1370. This probe was hybridized to a specific mRNA a t a size of 2.6 kilobases. A 0.97-kilobase fragment bearing the rpL32 processed gene, 4A, joined to the 5' and 3' flanks of p3Awas used as a control for loading of equivalent total RNA (35).
The DNA fragments were labeled with ["2PldCTP ( h e r s h a m ) using the random primed labeling technique (36) up to a specific activity of 3 x loR c p d p g . Labeled probes were used a t a final concentration of 5 x 10" cpndml hybridization mixture.
Immunoprecipitation-The procedure which was previously described for immunoprecipitation (37) was slightly modified a s follows. Three to 10 million MC-9 cells were incubated for 1.5 h at 37 "C in 250 pl of FCS and growth factor-free Dulbecco's modified Eagle's medium (Beit Haemek, Israel) containing [35Slmethionine (Amersham), followed by the addition of 40 units/lOfi cells of IL-3 for either 45 or 60 min. The cells were lysed by addition of 500 p1 of cold lysis buffer (0.01 M Tris-HC1, pH 7.4, 15% deoxycholate, 1% Triton X-100, 0.1% SDS, 0.15 M NaCI, and 0.25 mM phenylmethylsulfonyl fluoride), homogenized, and centrifuged in a microcentrifuge a t 4 "C for 30 min. The supernatants were collected and affinity purified sheep anti-c-fos (Cambridge Research Biochemicals Wilmington, DE) was added. After overnight incubation a t 4 "C, the mixtures were incubated for 2 h at room temperature with rabbit antisheep antiserum (Bio-Makor, Israel). Then, the mixtures were incubated with agitation for 3 h at room temperature with 10 mg of protein A-Sepharose beads (Sigma). A similar procedure was used to determine t h e c J u n , using rabbit anti-c-Jun antibody (Oncogene Science Inc.). The immunocomplexes derived from the treatment of the lysates with each antibody were dissociated by boiling for 10 min in the presence of 0.5% SDS and then diluted 5-fold with SDS-free buffer. In order to normalize samples before reimmunoprecipitation, 2 pl of each mixture were removed and counted by scintillation. Equal counts from each mixture were then reprecipitated with the same respective antibody. This technique facilitates the identification of the appropriate protein by eliminating most of the background. After being boiled in Laemmli sample buffer, precipitates were resolved on discontinuous 10% acrylamide-bisacrylamide SDS slab gels, along with 14C-methylated molecular weight markers (Sigma). Gels were dried and exposed to Kodak X-Omat AR film at -70 "C for various periods of time. Quantitation of the autoradiographic bands was performed by densitometric analysis.
Gel Shift Assay-Nuclear extracts were prepared from lo7 cells and used in the gel-shift assay as previously described (33). Synthetic double-stranded TRE and KB oligonucleotides were used for this assay. The following TRE sequence was synthesized: 5'-AGC TTA AAA AAG CAT GAG TCA GAC ACC TG-3' and the following KB sequence: 5'-CCT CTC GGA AAG TCC CCT CTG AAG CT-3'. The nature of the bound proteins were identified by using rabbit polyclonal antibodies against c-Jun (kindly provided by R. Bravo, Squibb).
Antisense and Sense Oligodeozynucleotides-Oligodeoxynucleotides were synthesized by Biotechnology General (Rehovot, Israel) and were purified by reverse phase high-performance liquid chromatography. The oligodeoxynucleotide products contain exclusively phosphorothioate internucleotidic linkages. The antisense and sense oligodeoxynucleotides employed were antisense c-jun (5'-GAA GGT CGT TTC CAT-3') and c -j u n -b Introduction ofAntisense and Sense 0li~~deox.vnucleotide.s into MC-9 Cells and into 37'3 Fibroblasts-Murine 3T3 fibroblasts were grown in Dulbecco's modified Eagle's medium with 0.5% fetal calf serum for 72 h with 5 p,I sense and antisense oligonucleotides. Tritiated thymidine was added for 7 h and then trypsin was added to the plates until all of the cells were detached from the plate and then transferred into microtiter plate wells and harvested as described.
MC-9 cells were washed three times with RPMI 1640 and then resuspended in enriched medium containing only 0.5% FCS. These cells were then cultured in a flat-bottomed 96-well microtiter plate a t a density of 10.000 cellslwell, in 100 p1 of the above medium containing 5 p~ of either sense or antisense to c-jun. After 24 h incubation a t 37 "C, 0.1 unit of IL-3lwell was added for another overnight incubation. The cells were labeled with radioactive thymidine and processed as described above.
RNA Isolation and cDNA Synthesis-The efficiency of the antisense treatment was determined by measurement of c-jun mRNA using the PCR method (33).
MC-9 cells were distributed a t a density of 5 million cellsll.5 ml of enriched medium containing 0.5% FCS in the presence of 20 1.111 ofeither antisense or sense to c-jun. The cells were cultured in a flat-bottomed 6-well Costar plate overnight a t 37 "C, after which they were stimulated for 15 min with 100 unitslml IL-3. RNA was then isolated from the cells by using the guanidinium method (33), essentially, cells were lysed with 5 M guanidinium isothiocyanate, 0.5% sarcosyl, 25 mhl sodium citrate, and 100 mM P-mercaptoethanol. The aqueous phase was layered on 5 11 CsCl and pelleted by centrifugation a t 38,000 rpm a t 18 "C for 18 h using a TST 41.14 rotor. The RNA pellet was resuspended in 300 pl of distilled water and finally ethanol precipitated. The RNA concentration was determined for each treatment by spectrophotometry and checked again for accuracy by running 1 pg of RNA from each treatment on a 1.5% agarose gel containing ethidium bromide. Then 1 pg of RNA from each treatment was incubated at 70°C for 2 min with a mixture of oligo(dT),* primer and then removed to ice. This mixture was then incubated a t 37 "C for 60 min with 400 units of Moloney murine leukemia virus reverse transcriptase and the following reagents: 1:s (v:v) 5 x reaction buffer, 0.5 mht dNTPs, and 1 unit of RNasin in a final volume of 20 pl. The resulting cDNA was diluted 1:5 with distilled water. Two cDNA doses were chosen for the following PCR procedure. 2.5 units of Taq DNA polymerase were added to a final volume of 50 pl containing 3 p~ 5' c-jun sequence specific primer and 3 pal 3' c-jun sequence primer as detailed above, with 0.25 mu dNTPs, 25 mA< Tris-HCI, pH 9.3, 2 my MgCI,, 50 mu KCI, and 0.018 (w/v) gelatin. The mixtures were overlaid with mineral oil to prevent evaporation and then were amplified by PCR (Thermocycler programmable heating block, Perkin Elmer Cetus). Samples were amplified up to 35 cycles according to the following program: 94 "C for 1 min, 54 "C for 1 min, and 72 "C for 2 min. Ten 1. 11 of each mixture was electrophoresed on a 1.8%. agarose gel. The cDNA separated on an agarose gel was transferred to Hybond-N filter and hybridized with a :lSP-labeled probe for c-jun. Linearity of PCR amplification was demonstrated by varying the concentration of cDNA.

RESULTS
We confirmed in a clone of mouse fetal liver-derived mast cells, MC-9 (27), that IL-3 caused an increase in DNA synthesis similar to that previously found in PKC-depleted mouse bone marrow-derived mast cells ( Ref. 12; Fig. 1). The effectiveness of

. Kinetics of c-fos and c-jun accumulation in IL-3-stimu-
lymphokines and serum for 3 h were exposed for various periods of time to 100 unitslml IL-3. After stimulation, cytoplasmic RNA was extracted and then loaded on an agarose/MOPS gel (10 pgflanc), electrophoresed, and blotted onto Hybond-N filter. Hybridization of the filter with the :"P-labeled probes of c-fos and L32 was performed and then the filter was stripped and rehybridized with a radioactive c-jun probe. One representative experiment out of four is shown.
10 nxl staurosporine (38) in inhibiting PKC activity was confirmed in MC-9 cells, by executing PKC activity assays prior to these experiments (data not shown). Thus the similarity between the effects of IL-3 on a primary culture of mast cells and on this IL-3-dependent mast cell clone enabled us to continue our study using MC-9 cells. Due to the large number of MC-9 cells that could be obtained, they were much more convenient for work, a s compared to cells derived from the primary cultures.
It has been suggested that the process of cell proliferation involves expression of nuclear proto-oncogenes c-fos and c-jun (23). To study whether these proto-oncogenes are expressed after IL-3 stimulation of mast cells, we measured their mRNA levels in IL-3-stimulated cells by Northern blot analysis. The kinetics of the lymphokine effect were determined in MC-9 cells which were exposed to 100 unitdm1 IL-3 for different time intervals. As shown in Fig. 2, IL-3 caused an increase in both c-fos and c-jun mRNA accumulation in a similar time-dependent manner. Maximal levels of mRNA for c-fos and c-jun were observed after 15 min of exposure to IL-3,9.5 2 1.6-fold ( n = 4) and 3.7 2 1.2-fold ( n = 4), respectively. No significant changes in the mRNA level of the gene encoding the ribosomal protein L-32 (35) (which was used as control here and in the following experiments), were detected in response to lymphokine stimulation.
Since depletion of PKC allows for increased growth in response to IL-3 (9), we assessed the role of PKC in the regulation of c-fos and c-jun expression in response to this lymphokine. Two approaches were employed to remove PKC activity from the cells, the first by chronically treating the cells with 20 ng/ml PMA for 72 h, which totally depletes the cells of PKC activity (9), the second approach was by using 10 n M staurosporine, a PKC inhibitor (38). As mentioned above, PKC activity assays were performed in order to confirm the effectiveness of these approaches in MC-9 cells (data not shown). Following 72 h of chronic PMA treatment, 100 unitdm1 IL-3 were added for 15 min (chosen as an adequate time point for comparison with this lymphokines effects on PKC-containing cells), and then the cells were assayed for mRNAlevels of each proto-oncogene. The different response of c-fos and c-jun to IL-3 stimulation a t the mRNA level, and the apparent PKC intervention in IL-3stimulated cells, prompted us to investigate a t which leveVs PKC regulates the expression of these proto-oncogenes, i.e. mRNA accumulation and/or protein and/or AP-1 binding activity.
Therefore, we investigated IL-3 regulation of the translation of these proto-oncogenes, and the role played by PKC in this regard. The c-Fos and c-Jun proteins were identified using a re-immunoprecipitation technique (37), with polyclonal antibodies against c-Fos and c-Jun. As mentioned, this technique facilitates the identification of the appropriate protein by eliminating most of the background. However, regardless of such treatment with both antibodies, a nonspecific band of approximately 46 kDa appeared. We were able to reconfirm the identification of c-Fos by either preincubating or not anti-c-Fos with the nonradioactive c-Fos peptide, prior to the addition of the antibody to the lysate. It was found that the molecular masses of c-Fos and c-Jun in MC-9 cells are approximately 59 and 44 kDa, respectively. Fig. 5, ZA and ZZA, show c-Fos and c-Jun to be elevated by 12.5 2 2.5-fold (n = 2) and 3.5 2 0.5-fold (n = 2), respectively, above control in IL-3-triggered cells. In staurosporine-treated IL-3-triggered cells (Fig. 5, ZB and ZZB), new synthesis of c-Fos was increased 3.  It is known that co-expression of c-Jun and c-Fos results in more efficient activation of AP-1 responsive genes because of the increased stability of the heterodimer (22). Therefore, to complete the evolving picture of IL-3 induced mRNA accumulation and translation of c-fos and c-jun, and PKCs involvement in these processes, we investigated its effects on AP-1 binding antisrnsr oligonucleotide to c-jun mRNA MC-9 cells were grown overnight in lymphokine-free medium containing 0.1% FCS and either sense or antisense phosphorothioate oligodeoxynucleotides to c-jun mRNA. Then 0.1 units of IL-3 was added to each well for further overnight incubation. Results are expressed in countslmin and are the mean 2 S.E. of quadruplicate samples. One representative experiment out of three. 3T3 fibroblasts were grown for 72 h with 0.5% fetal calf serum with the same oligonucleotides used in MC-9 cells utilizing the same concentrations. Results are expressed in countdmin and are the mean f S.E. of quadruplicate samples. One representative experiment out of two. Cell extracts were prepared from staurosporine-treated and untreated MC-9 cells that were exposed or not for 45 min to IL-3. Twenty micrograms from each extract were incubated with I:12PldATP and 1:"PldCTP radiolabeled TRE consensus sequence oligonucleotide in the absence or presence of 100-fold excess of unlabeled oligonucleotide, corresponding to the radiolabeled TRE probe (TRE compc't. or the KR oligonucleotide (KB conpet.) (nonspecific), and with polyclonal antibodies against c-Jun (anti-c-Jrol), before analysis by nondenaturing PAGE and autoradiography. One representative experiment out of three is shown.
activity. Fig. 6 shows that IL-3 induced AP-1 binding to the TRE oligonucleotide, while PKC inhibition, surprisingly, had no effect on these results. The nature of the bound proteins was determined, as mentioned under "Materials and Methods," by using a polyclonal antibody against c-Jun. As can be seen, this antibody bound to the AP-1 complex and thus caused a shift in the mobility of AP-1.TRE complex on the gel, it also caused a decrease in AP-1 activity, possibly due to interference with the binding of c-Jun to the TRE.
The dependence of c-jun mRNA accumulation and translation upon PKC activity, yet the apparent independence of IL-3 induced AP-1 activity, suggested no connection between IL-3induced AP-1 activity and the increase in IL-3-stimulated proliferation previously found in PKC-inhibited mast cells, yet indicated a possible connection between c-Jun expression and proliferation. This aspect was investigated using antisense treatment targeted against c-jun mRNA. By applying antisense oligos against c-jun mRNA in MC-9 cells we found an enhancement in DNA synthesis of IL-3-treated cells as compared to IL-3-stimulated cells treated with the sense oligomer (Table I). Possible toxicity of the oligodeoxynucleotides or other possible side effects specific to these antisense oligos were checked by using these same oligos in a different cell system. 3T3 fibroblasts, known to respond to antisense treatment against c-jun mRNA with reduced proliferation rates (39) were chosen. The same oligomers reduced fibroblast proliferation rates, as anticipated (Table I). Furthermore, the effectiveness of these treatments on c-jun mRNA expression was determined using the PCR method. Fig. 7 shows that the amount of c-jun mRNA available for reverse transcriptase, and then amplification by PCR in antisense-treated cells was much smaller than that available in sense-treated cells. Equal quantities of RNA were determined first by spectrophotometry and second by running equal amounts of RNA on a 1.56 agarose gel and staining with ethidium bromide. Linearity between c-jun template concentration and PCR amplification product was established by using several dilutions of cDNA at the starting of the PCR procedure.

DISCUSSION
The intracellular events regulating mast cell proliferation are of considerable interest taking into account the variety of pathological situations in which a n increased number of these cells are found. Several exogenous triggers can lead to increases in mast cell proliferation. An important cytokine that can cause this effect is IL-3. The effect of this cytokine is initially relayed by increases in the activity of several enzymes including some displaying PKC activity (40). We have previously shown that PKC is a regulator of IL-3-mediated mast cell growth (9). Further cellular events in the signal transduction of growth factors usually includes specific activation of transcription factors that enhance tissue-specific and nonspecific genes. One of the transcription factors that is induced by many exogenous stimuli is the AP-1.DNA binding complex (23). The activation of transcription by AP-1 can be influenced by many factors, such as: ( a ) its interaction with other transcription factors like the glucocorticoid receptor (41); ( b ) by its binding to c-jun in IL-3-mediated Pr the cytoplasmatic inhibitor IP-1 (42); ( c ) by specific phosphorylation (24); and ( d ) by the nature of its specific components (22,43). Furthermore, the interaction of its components with different transcription factors can create transcription complexes different than the AP-1 and lead to different cellular responses (43). For example, we have recently observed in IgEantigen-stimulated mast cells an increased binding of c-Fos to fos-interacting protein (44).
In order to begin the study of the potential importance of such a complex system in IL-3-induced mast cell proliferation we first checked IL-3 influence on c-fos and c-jun steady state mRNA and newly synthesized protein levels and on AP-1 DNA binding activity. In accordance with the recent findings of Horie and Broxmaeyer (45) we noted increased c-fos mRNA levels after IL-3 induction. We also observed IL-3 induced increases in all other parameters checked, including AP-1 DNA binding activity. Next, since we have previously shown that PKC is a regulator of IL-3-mediated mast cell growth, we tried to analyze the importance of PKC for IL-3 induction ofAP-l. We found that in PKC-depleted cells the IL-3 induced increases in the accumulation of c-jun mRNA and newly synthesized protein were abrogated while that of c-fos was only weakened (Figs. [3][4][5]. Yet IL-3-induced AP-1 DNA binding activity was not repressed in comparison to PKC-containing cells. The fact that AP-1 binding is not repressed in these cells can be explained by several ways including a different composition of the AP-1 complex in these cells and differential AP-1 DNA binding inhibition in PKC-depleted cells. IL-3 induces increases in c-Fos, c-Jun, and also cellular proliferation. Thus one might suggest a role for c-Jun in the increase of mast cell proliferation, that complies with most direct observations on the role played by c-Jun in the regulation of proliferation of other cell types (39,46,47). Contrary to those observations, the abrogation of the IL-3induced increase of c-Jun in PKC-depleted cells is associated with increased mast cell proliferation rates, suggesting that c-Jun is a growth inhibitor in these cells. Therefore, two contradicting clues for the role of c-Jun in IL-3-induced mast cell proliferation were obtained. In order to try and resolve this problem we decided to specifically inhibit c-Jun expression by the addition of sense and antisense oligodeoxynucleotides targeted against c-jun mRNA to the cells. Initially we determined the effectiveness of this treatment on c-jun mRNA levels. Fig. 7 shows that the amount of c-jun mRNA as detected by PCR in antisense-treated cells was much smaller than that in sensetreated cells. We then showed that c-Jun depletion increases specifically IL-3-induced proliferation and hardly has any effect on uninduced mast cells (Table I). This result is in contrast with all previous studies on the involvement of c-Jun in cellular proliferation that utilized direct repression of c-Jun by intracellular injection of c-Jun antibodies (37) or by c-jun specific antisense inhibition (39,46,47). In order to exclude the possibility that our results are due to special properties of our specific antisense oligomers, we checked their effect on 3T3 murine fibroblasts. As expected, the c-jun antisense oligomer induced a reduction in [3H]thymidine uptake in these cells. Therefore, c-Jun has a growth inhibitory role in IL-3-treated mast cells in contrast to its role in several other cellular systems. An attractive possibility to explain c-juns role, especially in the light of the relatively undiminished IL-3 induction of AP-1 binding activity in PKC depleted MC-9 cells, is that c-Jun is involved in cellular growth inhibition not through the AP-1 complex but through its binding to other transcription factors. These factors can be a part of the CAMP response element-binding protein family (43,21) or perhaps tissue specific ones. We are currently .oliferation of Mast Cells 8503 on AP-1 component regulation in mast cells is a complex one, fitting in several aspects to the situation in other cellular systems but distinct in other ways. We believe that further research on the control and function ofAP-1 components in this cellular environment will lead to a better understanding of mast cell growth regulation in particular and hopefully contribute to a better understanding of cellular proliferation in general.