Induction of Stem Cell Factor/c-Kit/Slug Signal Transduction in Multidrug-resistant Malignant Mesothelioma Cells*

Malignant mesothelioma (MM) is strongly resistant to conventional chemotherapy by unclear mechanisms. We and others have previously reported that cytokine- and growth factor-mediated signal transduction is involved in the growth and progression of MM. Here, we identified a pathway that involves stem cell factor (SCF)/c-Kit/ Slug in mediating multidrug resistance of MM cells. When we compared gene expression profiles between five MM cells and their multidrug-resistant (MM DX) sublines, we found that MM DX cells expressed both SCF and c-Kit and had higher mRNA levels of Slug. Knockdown of c-Kit or Slug expression with their respective small interfering RNA sensitized MM DX cells to the induction of apoptosis by different chemotherapeutic agents, including doxorubicin, paclitaxel, and vincristine. Transfection of c-Kit in parental MM cells in the presence of SCF up-regulated Slug and increased resistance to the chemotherapeutic agents. Moreover, MM cells expressing Slug showed a similar increased resistance to the chemotherapeutic agents. These results indicate that induction of Slug by autocrine production of SCF and c-Kit activation plays a key role in conferring a broad

Malignant mesothelioma (MM) 1 is an aggressive tumor caused primarily by asbestos exposure. MM is characterized by overexpression of several cytokines, growth factors, and tyrosine kinase receptors (1). Some of these are proto-oncogenes and key regulators for proliferation, differentiation, and motil-ity of MM cells (2). Whether cytokines and growth factors may control other mechanisms of MM progression remains to be determined.
MM is poorly responsive to chemotherapy, and median patient survival is 8 -18 months (3). MM cells exhibit resistance in vitro and in vivo to many anti-cancer agents, including doxorubicin and cisplatin, which are nevertheless widely used to treat MM (4). In most cases, this is the consequence of overexpression of the ATP-binding cassette (ABC) transporters, including MDR1 (ABCB1), MRP1 (ABCC1), and MRP2 (5). However, ABC transporters cannot completely explain the widespread MDR phenotype of MM cells (6,7), suggesting that other factors may be involved.
The tyrosine kinase receptor c-Kit and its ligand stem cell factor SCF are essential to the maturation of hemopoietic and primordial germ cell precursors and melanocytes during embryonic development (8,9). In addition, aberrant expression of c-Kit and/or SCF has been reported in several tumors, such as acute myeloid leukemia (10,11), small cell lung cancer (12), gynecological tumors (13), and breast carcinomas (14). Thus, constitutive activation of c-Kit is also involved in tumor progression.
Multiple cellular functions are affected by c-Kit-dependent signals including cell survival, proliferation, adhesion, and differentiation (15). SCF augments the derivation of melanocytes from embryonic stem cells (16). Moreover, c-Kit activation was found to suppress apoptosis of normal murine melanocyte precursors (17), soft tissue sarcomas of neuroectodermal origin (18), and normal and malignant human hemopoietic cells (19). This antiapoptotic activity of SCF/c-Kit signals is mediated, at least in part, by the zinc finger transcription factor Slug (20), which functions as transcriptional repressor (21). However, we can find very little or no documentation regarding the expression of SCF/c-Kit system or Slug in human MM tissue.
In this study, we set out to gain insights into the control of MDR in MM cells. We determined whether cytokines, growth factors, or their receptors can activate and consequently protect MM cells against chemotherapeutic injury, and, if so, we dissected the molecular mechanisms underlying this cytoprotection. We identified the SCF/c-Kit signaling pathway as a component of the MDR program induced by chemotherapy and provide evidence that this signal acts in the MDR phenotype, in part, by modulating the activity of Slug.

EXPERIMENTAL PROCEDURES
Antibodies and Reagents-All of the primary antibodies used, excluding anti-FLAG monoclonal antibody (by Sigma) and monoclonal anti-MDR1 antibody (Calbiochem), were obtained from R&D Systems Inc. (Minneapolis, MN). The antineoplastic agents doxorubicin (adriamycin), paclitaxel (taxol), and vincristine as well as verapamil were purchased from Sigma. Recombinant human SCF was obtained from * This work was supported by grants from the Italian Ministry of Research and from Associazione Italiana per la Ricerca contro il Cancro (to A. P.). 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 solely to indicate this fact.
Sigma. The pBabePuro plasmid was a gift from Dr. G. J. Clark (NCI, National Institutes of Health, Bethesda, MD). All other reagents were obtained from Sigma unless otherwise specified.
Cell Lines-We used a panel of five human MM cell lines (22,23). MM1, MM3, MM4, and MM5 cells were derived from previously untreated patients and identified morphologically and by extensive phenotypic analysis as previously described (24). They were then expanded and utilized for the experiments. H-Meso cell lines (MM2) have been characterized previously (25). MM DX cells were selected from their parental cells by stepwise increases in doxorubicin concentrations from 0.0001 to 0.3 M, as detailed elsewhere (26). MM DX cells were then maintained in the presence of 2.5 M doxorubicin. All cells were maintained in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 1% L-glutamine, 1% penicillin/streptomycin (complete medium) (all from HyClone, Rome, Italy) at 37°C and 5% CO 2 .
Proliferation Assays-Chemotherapy sensitivity was assessed by MTT assay as previously described (22). The cells were pulse-exposed to different concentrations of doxorubicin, vincristine, or paclitaxel for 24 h. The cells were then cultured for an additional 36 h in culture medium with 10% fetal bovine serum. After this time, the remaining cells were stained with MTT (Promega, Madison, WI) using the manufacturer's instructions. Absorbances were normalized, and LC 90 values were calculated by nonlinear regression. Normalized data and nonlinear regression curves were plotted graphically as a percentage of viable cells. Alternatively, cells were incubated with chemotherapeutic drugs for 24 h. Cells were then collected after different times, stained with trypan blue, and counted.
Apoptosis-Cells (5 ϫ 10 5 ) were plated 24 h before treatment. Cells were then exposed to the indicated chemotherapeutic agents for 24 h. At the end of treatment, cells were washed twice with phosphate buffer with calcium and magnesium (PBS), and apoptosis was determined using the cell death detection ELISA kit (Roche Applied Science) according to the manufacturer's instruction.
Reverse Transcription and Real Time PCR-For real time PCR, total RNA was extracted with the RNeasy minikit (Qiagen, Milan, Italy) and reverse-transcribed, and the cDNA was quantified using the TaqMan system (Applied Biosystems, Milan, Italy). Values for each mRNA were normalized to the relative quantity of GAPDH mRNA in each sample. The PCRs were carried out in a Chromo4 sequence detector (MJ Research, Waltham, MA). The primer and probe sets used for the amplification of the investigated genes were supplied by Applied Biosystems. Details of sequences and thermal cycle conditions are available upon request.
To analyze the expression of Slug and Snail in cell lines, reverse transcription (RT) was performed as previously described (20). Thermocycling parameters for PCR and the sequences of the specific primers were as follows: Slug, 30 cycles at 94°C for 1 min, 56°C for 1 min, and 72°C for 2 min, sense primer 5Ј-GCCTCCAAAAAGCCAAACTA-3Ј, antisense primer 5Ј-CACAGTGATGGGGCTGTATG-3Ј; Snail, 30 cycles at 95°C for 2 min, 60°C for 2 min, and 72°C for 2 min, sense primer 5Ј-CAGCTGGCCAGGCTCTCGGT-3Ј, antisense primer 5Ј-GC-GAGGGCCTCCGGAGCA-3Ј. Amplification of ␤-actin RNA served as a control to assess the quality of each RNA sample.
Immunoblotting, Immunoprecipitation, and ELISA-Proteins (50 -75 g/lane) were separated on 7.5-10% SDS-polyacrylamide gels and transferred to nitrocellulose. Transfers were blocked for 2 h at room temperature with 5% nonfat milk in TBS, 0.1% Tween 20 and then incubated overnight at 4°C in the primary antibody diluted (1:1000) in 5% bovine serum albumin in TBS, 0.05% Tween 20. The transfers were rinsed with TBS, 0.05% Tween 20 and incubated for 1 h at room temperature in horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse (Bio-Rad) diluted 1:5000 in 5% nonfat milk in TBS, 0.1% Tween 20. The immunoblots were developed with the Super Signal reagent (Pierce). Protein concentration was determined by standard Bradford protein assay (Bio-Rad). All Western blots were reprobed with human ␤-actin antibody (sc-1616; 1:1000) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) to confirm equivalent loading of protein samples. Human SCF levels in cell lysates and culture supernatants were determined using an ELISA kit (R&D Systems) according to the manufacturer's instructions.
For analysis of c-Kit receptor autophosphorylation, cells were rapidly lysed in Nonidet P-40 lysis buffer (1% Nonidet P-40, 20 mM Tris (pH 8.0), 137 mM NaCl, 1 M NaF, 0.5 mM EDTA, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, 0.15 units/ml aprotinin, 20 M leupeptin, and 1 mM sodium vanadate) on ice for 20 min. Cell lysates were cleared by centrifugation for 20 min at 14,000 ϫ g, and protein content was quantitated as described above. One mg of total protein was immunoprecipitated with anti-c-Kit antibody and detected by Western blotting with the anti-phosphotyrosine antibody (Santa Cruz Biotechnology).
Fluorescence-activated Cell Sorting Analysis-Cells (1 ϫ 10 5 ) were washed three times in PBS containing 0.5% bovine serum albumin (PBS buffer) and then incubated for 30 min at 4°C with appropriate amounts of anti-c-Kit monoclonal antibody (R&D Systems). Cells were washed twice with PBS and then incubated with fluorescein isothiocyanate-conjugated (Fab) 2 goat anti-mouse IgG. After three washes, samples were analyzed on a FACScan (BD Biosciences). A minimum of 10,000 events/sample was analyzed. As a negative control, an isotypematched monoclonal antibody with irrelevant specificity was used.
Plasmids-The full-length c-Kit cDNA was cloned into the pcDNA3 plasmid vector and transfected into MM5 cells by LipofectAMINE reagent (Invitrogen) according to the manufacturer's instructions. Stable transfected cells were selected in 400 g/ml Geneticin and further subcloned as previously described (27). The expression of c-Kit protein in the single-cell clones (G5 and G12) was characterized by Western blotting and fluorescence-activated cell sorting analysis.
Full-length cDNA for human Slug (GenBank TM accession number BC014890) and Snail (GenBank TM accession number NM005985) were amplified from RNA of MM cell lines by reverse transcription-PCR, and a COOH-terminal FLAG epitope tag was added by PCR as previously described (28). Constructs were subcloned into the pBabePuro plasmid. The identities of all plasmid inserts and vector boundary regions were confirmed by sequence analysis. Details of the construct are available upon request. Transient transfection with Slug and Snail constructs and culturing of transduced cells were performed by standard procedures.
To generate Slug d-siRNA, full-length cDNA of human Slug with T7 promoters at both ends were made by PCR and subjected to in vitro transcription to produce large double-stranded RNA as previously described (29). Then Slug d-siRNA was produced using an r-Dicer kit (Stratagene, La Jolla, CA) according to the manufacturer's instructions. For transfection, cells (10 5 cells/well) were plated in 6-well plates, grown in antibiotic-free medium overnight, and then transfected with specific siRNA using Oligofectamine reagent (Invitrogen) according to the manufacturer's instructions. At 24 or 48 h after transfection, cells were trypsinized, replated, and incubated overnight before treatment.
Northern Blotting-Total RNA was isolated from cells using TRIzol reagent (Invitrogen, Milan, Italy). Electrophoretic separation and membrane transfer of RNA were carried out by standard methods. For use as probes, Slug, Snail, and GAPDH cDNA fragments were labeled with [ 32 P]dCTP (Amersham Biosciences) by random priming with the Re-diPrime II kit (Amersham Biosciences). Prehybridization and hybridization were carried out in Rapid-Hyb buffer (Amersham Biosciences), a The cells were pulse-exposed to increasing concentrations of doxorubicin (from 0 to 2.5 M), paclitaxel (from 0 to 0.5 M), and vincristine (from 0 to 0.5 M) for 24 h. The cells were then cultured for an additional 36 h in culture medium supplemented with 10% fetal bovine serum. Drug concentration that was cytotoxic for 90% of treated cells were determined with an MTT assay as described under "Experimental Pro- the membrane was washed, and the blot was exposed to BioMax MS film (Eastman Kodak Co.). Statistical Analysis-All values were expressed as mean Ϯ S.D. Comparison of results between different groups was performed by oneway analysis of variance and paired t test using StatView 5.0 (NET Engineering, Pavia, Italy).

Characterization of the Multidrug-resistant Malignant Mesothelioma Cells (MM DX Cells)-Five
MM DX sublines were selected from their parental MM cell lines by exposure to stepwise increases in doxorubicin concentrations. Their drug sensitivity profile to doxorubicin, paclitaxel, and vincristine are shown in Table I. The concentration of drug lethal for 90% of treated cells (LC 90 ) was calculated from dose-response curves for each drug tested using the MTT assay. MM DX cells with an LC 90 value Ն3-fold greater than that of parental cells were considered resistant.
All of the five MM DX sublines expressed an elevated amount of MDR1 protein (also known as P-glycoprotein) (Fig.  1A). To investigate to what extent P-glycoprotein expression determines MDR phenotype, we treated MM DX sublines with verapamil, a substance that antagonizes P-glycoprotein activ-ity through competitive block of calcium channels. Contemporary application of doxorubicin and 5 M verapamil, a concentration known not to have a cytotoxic effect, did not significantly modify the LC 90 value of MM1 DX or MM2 DX sublines as determined by trypan blue exclusion (Fig. 1, B and C) and MTT assay (data not shown). These results indicated that MM DX cells acquired an MDR phenotype, which was independent by MDR1 expression.
Cytokines, Growth Factors, and Their Receptor Expression in MM DX and MM Cells-Differential cytokine and growth factor mRNA expression between MM DX and MM cells were determined by real time PCR analysis. Four of five MM DX sublines showed a 2-4-fold increase in the expression of SCF ( Fig. 2A). By contrast, MM DX and parental cells expressed equal amounts of VEGF, TGF-␤, insulin-like growth factor-1, PDGF, and FGF-2 ( Fig. 2A) as well as TGF-␣ and hepatocyte growth factor (data not shown). Expression of SCF protein was confirmed by Western blotting analysis. Membrane-bound SCF showed a marked increase in the MDR sublines compared with the parental cells, whereas soluble SCF was not detectable in all cell types (Fig. 2B, inset). Accordingly, increased amounts of SCF were detected by ELISA in the cell lysates (Fig. 2B), but The expression of tyrosine kinase receptors in both MM DX and MM cells was then evaluated by real time PCR, and most of the MDR sublines expressed higher mRNA levels of the SCF receptor c-Kit than parental cells (Fig. 3A). VEGF receptor-1 (Flt-1) and PDGF receptor-␣ mRNA increased in three of five MM DX cells, whereas VEGF receptor-2 (KDR), FGF receptor-1 and PDGF receptor-␤ decreased in four of five MDR sublines compared with parental cells (Fig. 3A). In addition, MM DX cells did not modify mRNA levels of EGF receptor (c-erbB2) and hepatocyte growth factor receptor (MET) compared with parental cells (data not shown). Consistently, c-Kit protein expression was higher in four of five MM DX sublines with respect to parental cells with the exception of MM3 DX cells (Fig. 3B).
Down-regulation of c-Kit Expression or Function by Small Interfering RNA and Neutralizing Anti-c-Kit Antibody Restore Sensitivity to Anticancer Agents in MM DX Sublines-The coexpression of SCF and its receptor c-Kit in some of the MM DX cells suggested that this pathway could confer MDR to anticancer compounds by an autocrine/paracrine mechanism. To address this hypothesis, we knocked down c-Kit with small interfering RNA (c-Kit siRNA) duplexes. Exposure of MM1 DX or MM2 DX cells to c-Kit siRNA was associated with downregulation of c-Kit expression (Fig. 4A). There was no detectable down-regulation of c-Kit in MM1 DX or MM2 DX sublines exposed to a control siRNA (CsiRNA) (Fig. 4A). Neither c-Kit siRNA nor CsiRNA inhibited MDR1 protein expression in MM1 DX (Fig. 4B) as well as in MM2 DX sublines (data not shown), suggesting no functional association between MDR1 and SCF/ c-Kit signals.
MM1 DX or MM2 DX sublines and their parental cells exposed to CsiRNA or c-Kit siRNA were then analyzed for nucleosome formation, a marker of apoptosis (30). MM1 DX, MM2 DX, or parental cells exposed to c-Kit siRNA exhibited no significant increase in nucleosome formation compared with controls (Fig. 4C). By contrast, MM1 DX and MM2 DX cells exposed to c-Kit siRNA responded to doxorubicin, paclitaxel, or vincristine with an increase in apoptosis compared with that obtained with cells exposed to CsiRNA (Fig. 4C). Similar results were obtained by adding neutralizing anti-c-Kit antibody (5 g/ml), whereas an antibody with irrelevant specificity had no effect (data not shown). These findings indicate that blocking the autocrine activity of SCF or knocking down c-Kit expression increased sensitivity to genotoxic agents in the MDR sublines.
Activation of the c-Kit-dependent Pathway by SCF Confers MDR in MM Cells-To further determine the role of SCF/c-Kit signaling in mediating the resistance of MM cells, MM5 cells lacking endogenous c-Kit (see Fig. 3) were engineered to stably express a wild-type, full-length c-Kit receptor. We used representative clones G5 and G12 in our studies, which had high c-Kit protein levels as well as MM5 DX sublines (Fig. 5A). G12 or G5 cells were treated with SCF (100 ng/ml) for 10 min, and autophosphorylation of c-Kit was examined by immunoprecipitation with anti-c-Kit antibody followed by Western blot with anti-phosphotyrosine antibody. In the presence of SCF, autophosphorylation of c-Kit was detected in G12 cells, indicating c-Kit activation (Fig. 5B). Notably, MM5 DX cells also showed c-Kit autophosphorylation in the absence of SCF (Fig. 5B), suggesting that the autocrine loop of SCF is constitutively active in MM DX sublines.
G5 or G12 cells were next cultured in the presence or absence of SCF, and their resistance to doxorubicin, paclitaxel, and vincristine was determined. Unlike control vector-transfected MM5 cells, G12 cells in the presence of SCF showed universally increased resistance to all three chemotherapeutic agents (Fig.  5C). Thus, the presence of both c-Kit and SCF in MM cells promotes MDR to cytotoxic drugs. Interestingly, c-Kit overexpression also increased resistance to doxorubicin in G12 cells in the absence of exogenous SCF by 30%, consistent with an autocrine role of MM5-derived SCF (see Fig. 2B). This effect, however, was not statistically significant. ern blot analysis. Expression of Slug, rather than that of Snail, was strongly correlated with the MDR phenotype of MM cells (Fig. 6A). To assess whether SCF affected Slug expression in MM cells, we determined Slug and Snail mRNA levels in c-Kittransfected cells in the presence of SCF using RT-PCR. G12 cells specifically expressed Slug upon SCF stimulation and in a time-dependent manner (Fig. 6B). By contrast, Snail was expressed at similar levels in SCF-unstimulated and -stimulated G12 cells (data not shown). In concert with these findings, knockdown of c-Kit expression in MM1 DX or MM2 DX cells by c-Kit siRNA was associated with decreased Slug expression, as compared with cells transfected with CsiRNA (Fig. 6C). We observed comparable results when we used the neutralizing anti-c-Kit antibody in place of c-Kit siRNA (results not shown), ruling out unspecific effects of siRNA.

Modulation of Slug Expression and Function by SCF/c-Kit Pathway in MM DX
To address whether Slug functionally contributes to the MDR phenotype of MM cells, we used the recombinant human Dicer (r-Dicer) to generate random pools of siRNA (d-siRNA) from double-stranded RNA corresponding to the entire coding regions of Slug (29). The Slug d-siRNA was transfected into MM1 DX sublines, and the expression of Slug and Snail was assessed by RT-PCR. The Slug d-siRNA decreased Slug by 80% (Fig. 6D). Expression of Snail or ␤-actin was unaffected, and d-siRNA derived from an irrelevant laminin cDNA had no effect on either Slug or Snail expression (Fig. 6D). As observed above with c-Kit siRNA, transfection of Slug d-siRNA sensitized MM1 DX or G12 cells treated with SCF to apoptosis induced by all three drugs tested (Fig. 6E). As control, Slug d-siRNA did not increase apoptosis in MM1 parental cells or SCF untreated G12 clones (data not shown). The results indicate that Slug as well as SCF/c-Kit can mediate the broad chemoresistance observed in MM DX sublines.
To extend these data, constructs expressing full-length, FLAG epitope-tagged Slug and Snail were generated (Fig. 7A), and the effects of these proteins in mediating the resistance of MM cells were assessed. When compared with the effects of the empty expression vector, Slug expression reduced the sensitivity of MM cells to apoptosis induced by all three chemotherapeutic agents, whereas Snail had no effect (Fig. 7B). Taken together, these results indicate that the effects of SCF/c-Kit signal transduction on MDR phenotype of MM cells was contingent on Slug activity. MDR have been intensively studied, since experimental models can be easily generated by in vitro selection with cytotoxic agents, and different types of MDR have been described. One way is to pump drugs out of cells by increasing the activity of efflux pumps, such as ATP-dependent transporters (31). In cases in which drug accumulation is unchanged, activation of detoxifying proteins, such as cytochrome P450 mixed function oxidases, can promote drug resistance (32). Cells can also activate mechanisms that repair drug-induced DNA damage (31). Again, endogenous mitogens, such as oncogene activation, can suppress cell death through NF-B signaling (33) and other prosurvival pathways, such as the phosphatidylinositol 3-kinase/AKT signaling cascade (34). Finally, disruptions in apoptotic signaling pathways (e.g. p53 or ceramide) allow cells to become resistant to drug-induced cell death (35).
Selection of cancer cells in culture with natural product anticancer drugs, such as paclitaxel, doxorubicin, or vinblastine, frequently results in MDR that is due to expression of the ABC transporter MDR1 (5). However, host and tumor genetic alterations, epigenetic changes, and tumor environment all seem to contribute to the complex story of cancer drug resistance (36). Therefore, in any population of cancer cells that is exposed to chemotherapy, more than one mechanism of MDR can be present. In this study, we have investigated whether the MDR phenotype of MM cells is also mediated by environmental factors, such as cytokines and growth factors, that are overexpressed in MM (1, 2). We identified the tyrosine kinase receptor c-Kit and its ligand SCF as genes up-regulated in MDR sublines of MM cells. Moreover, induction of SCF and c-Kit is not simply associated with an MDR state, but these molecules actively contribute to MDR. Indeed, knocking down c-Kit expression increases sensitivity to chemotherapeutic agents in MDR sublines, and forced expression of SCF/c-Kit signal is sufficient to lead to MDR in parental cells.
The biological events controlled by the SCF/c-Kit signaling pathway are first implicated in the generation and migration of hematopoietic stem cells (8,9). Moreover, mutations resulting in constitutive activation of c-Kit have been described in acute myeloid leukemia (10,11), small cell lung cancer (12), gynecological tumors (13), breast carcinomas (14), and colonic tumors derived from interstitial cells of Cajal, which are SCF-dependent (37,38). Thus, activation of the c-Kit-dependent pathway is also involved in malignant transformation of both leukemia and solid tumors.
Recently, it was observed that a SCF/c-Kit-dependent pathway protects erythroid precursor cells from chemotherapeutic agents (19) and that exposure of cultured human keratinocytes and melanocytes to UVB light up-regulates SCF and c-Kit, promoting cell survival (17). These findings are congruent with our results and suggest that some cell types may acquire the ability to induce SCF/c-Kit signals as a protective mechanism against DNA-damaging agents. However, the stimuli responsible for the constitutive expression of SCF and c-Kit in MDR sublines remain unclear. Recent evidence indicates that drug exposure may induce not only resistance but also invasiveness in cancer cells (39). We observed that MM cells and MM DX sublines secreted similar levels of matrix metalloproteinases 2 and 9. 2 Moreover, we observed higher protein levels of membrane-bound SCF in MDR sublines than parental cells, whereas soluble SCF was not detectable in all cell types or in conditioned media (see Fig.  2B). Thus, it is unlikely that proteases are involved in SCFinduced MDR.
Although c-Kit expression was found in a number of MM cell lines (40), little is known about the expression of SCF and c-Kit in primary MM tissue. One study evaluating c-Kit expression in formalin-fixed paraffin-embedded MM tissue identified c-Kit in 7 of 33 samples examined (41). Another study observed c-Kit in a subset of 37 cases of archived MM with one type of anti-Kit antibody, but only in a single MM case with a second anti-Kit antibody (42). Since fresh tissue was not available for examination by complementary methods or optimized histochemical procedures, c-Kit expression in additional samples may have been beyond the limits of detection. In addition, no information was presented regarding the clinical staging of the MM tissue. Nevertheless, detection of c-Kit expression in at least some of the archival MM tissue studied supports an involvement for c-Kit-dependent pathways in MM biology.
Our results show that the MDR conferred to MM cells by the SCF/Kit signaling pathway is mediated by a specific member of the Snail family of zinc finger transcription factors, namely Slug. The Snail transcription factors are implicated in epithelial-mesenchymal transitions during mammalian development as the generation and migration of mesoderm and neural crest cells (21). Recent findings show that Slug is expressed in t (17,19) leukemic cells (43), in rhabdomyosarcoma cells (44), and in breast cancer (28). Moreover, Slug mediates the oncogenic effect of the E2A-HLF fusion protein in leukemia (43) and is the molecular target of c-Kit-dependent pathway that promotes the radioresistance of bone marrow cells (20). Thus, Slug may be a common component providing chemoresistance to tumor cells and, therefore, might constitute an attractive target in the treatment of MM. However, the precise mechanisms FIG. 6. SCF/c-Kit signaling up-regulates Slug in MM DX cells, and Slug inhibition increases MM DX apoptosis induced by chemotherapeutic drugs. A, total RNA was collected from the indicated cells, separated on an agarose-formaldehyde gel, and subjected to Northern blotting. The same membrane was hybridized sequentially with probes to Slug, Snail, and GAPDH. The GAPDH blot confirms roughly equivalent loading of RNA samples. B, G12 cells that constitutively express c-Kit were incubated for the indicated periods of time with SCF (100 ng/ml). After incubation, Slug, Snail, and actin expression were determined by semiquantitative RT-PCR. C, MM1 DX and MM2 DX cells were transiently transfected with CsiRNA or c-Kit siRNA as described in Fig. 4. After 48 h, expression of Slug was analyzed by RT-PCR. D, MM1 DX cells were transfected with two different pools of d-siRNA, one complementary to Slug and another to laminin C. After 48 h, total RNA was extracted and subjected to RT-PCR with specific primers for Slug and Snail. Expression of actin was used as an internal control. E, MM5 DX sublines and G12 clones, treated with SCF as described in Fig. 5, were transfected with the Slug d-siRNA pool. After 24 h, cells were pulse-exposed for 24 h to the chemotherapeutic agents followed by an additional 16-h drug-free culture period. Apoptosis was then determined by a cell death detection ELISA kit. *, p Յ 0.01 versus mock, ANOVA, n ϭ 4. FIG. 7. Slug, but not Snail, increased resistance to apoptosis induced by chemotherapeutic drugs. A, empty vector, Snail, and Slug constructs were transfected into MM1 cells, lysates were prepared, and expression of FLAG epitope-tagged Slug and Snail proteins was determined by immunoblotting. The anti-FLAG signal indicates the expression of Snail and Slug proteins, and the anti-actin signal confirms equivalent loading of the lanes. B, empty expression vector or a vector encoding Snail or Slug was transfected in MM1 cells. After 48 h, cells were pulse-exposed for additional 24 h to the chemotherapeutic agents as described in the legend to Fig. 5. Apoptosis was then determined by a cell death detection ELISA kit. *, p Յ 0.01 versus control, ANOVA, n ϭ 4.
of Slug effects on multidrug resistance remain to be clarified, although they were not the consequence of MDR1 expression, since slug inhibition by c-Kit siRNA did not alter MDR1 expression and verapamil did not restore the sensitivity to pharmacological treatment in MM DX sublines. Slug and other members of the Snail family bind to specific target genes and function as transcriptional repressors (21). We were unable to detect any significant influence of Slug on pro-or antiapoptotic genes of the Bcl-2 family members, such as bax and bcl-2 (results not shown), which are frequently involved in the tumor resistance to apoptosis (31). Interestingly, Slug represses Ecadherin expression in breast cancer (28), and the loss of Ecadherin function occurred during tumor progression (45). The role of E-cadherin within this context is being investigated.
In summary, we demonstrated for the first time that SCF, c-Kit, and Slug are up-regulated in MDR sublines. Autocrine production of SCF by tumor cells activates its tyrosine kinase receptor c-Kit and then induces slug gene expression that mediates resistance of the cells to chemotherapy. This novel function of SCF/c-Kit/Slug pathway in cancer biology may provide a rationale to use a combination of conventional chemotherapeutic drugs with a new generation of SCF/c-Kit/Slug signal transduction inhibitors for the treatment of MM.