Molecular mechanisms of ceramide-mediated telomerase inhibition in the A549 human lung adenocarcinoma cell line.

This study was aimed at identifying the molecular mechanisms by which ceramide inhibits telomerase activity in the A549 human lung adenocarcinoma cell line. C(6)-ceramide (20 microm) caused a significant reduction of telomerase activity at 24 h as detected using the telomeric repeat amplification protocol, and this inhibition correlated with decreased telomerase reverse transcriptase (hTERT) protein. Semi-quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) and Northern blot analyses showed that C(6)-ceramide significantly decreased hTERT mRNA in a time-dependent manner. Electrophoretic mobility shift and supershift assays demonstrated that the binding activity of c-Myc transcription factor to the E-box sequence on the hTERT promoter was inhibited in response to C(6)-ceramide at 24 h. These results were also confirmed by transient transfections of A549 cells with pGL3-Basic plasmid constructs containing the functional hTERT promoter and its E-box deleted sequences cloned upstream of a luciferase reporter gene. Further analysis using RT-PCR and Western blotting showed that c-Myc protein but not its mRNA levels were decreased in response to C(6)-ceramide at 24 h. The effects of ceramide on the c-Myc protein were shown to be due to a reduction in half-life via increased ubiquitination. Similar results were obtained by increased endogenous ceramide levels in response to nontoxic concentrations of daunorubicin, resulting in the inhibition of telomerase and c-Myc activities. Furthermore, the elevation of endogenous ceramide by overexpression of bacterial sphingomyelinase after transient transfections also induced the inhibition of telomerase activity with concomitant decreased hTERT and c-Myc protein levels. Taken together, these results show for the first time that both exogenous and endogenous ceramides mediate the modulation of telomerase activity via decreased hTERT promoter activity caused by rapid proteolysis of the ubiquitin-conjugated c-Myc transcription factor.

Without their telomeric caps, human chromosomes undergo chromosome fusion or degradation (1)(2)(3)(4). Those telomeric ends comprise long tandem repeats of hexanucleotide 5Ј-TTAGGG-3Ј, and since conventional DNA polymerases cannot replicate the 5Ј-end of linear DNA, telomeres shorten as a function of each cell division in normal human somatic cells. This telomeric shortening is believed to control cellular senescence (5)(6)(7). Telomeres are primarily controlled by telomerase, which is composed of a catalytic protein subunit, telomerase reverse transcriptase (hTERT), 1 a stably associated RNA moiety (hTR), and telomerase-associated protein (TEP1) (8 -13). Moreover, it has been shown that the disruption of telomere maintenance by inhibiting telomerase limits proliferation of human cancer cells due to excessive telomere erosion (14,15). It has been shown also that telomerase is not active in most somatic tissues, whereas it is activated in the majority of cancer-derived cell lines and malignant tumors, suggesting that telomerase plays an important role in cell immortalization and tumorigenesis (16,17).
It has been shown that telomerase activity is present in the majority of lung cancers and not detectable in normal lung tissues (18 -21). Moreover, it has been demonstrated that the presence of telomerase activity in tumors of non-small cell lung cancer patients correlates with a high cell proliferation rate and an advanced pathologic stage (22). Therefore, it has been suggested that telomerase activity is one of the most important prognostic factors in lung cancer patients and that telomerase can be an important target to develop novel therapeutic strategies for the treatment of lung cancers.
The regulation of telomerase is well studied. The overexpression of hTERT mRNA parallels telomerase activity in nearly all tumor cells whereas it is repressed in most normal human somatic cells (16,17). However, hTR and hTEP1 are expressed in normal tissues (11,23), and overexpression of hTERT in telomerase negative cells is sufficient for enzyme activity, suggesting that the activation of hTERT transcription may potentially be the dominant rate-limiting step in telomerase regulation (12). Recently, the promoter region of hTERT has been cloned and sequenced (24 -27), and it has been shown to contain recognition sequences for several transcription factors including SP1, AP2, and c-Myc (24 -27). It has also been shown that c-Myc activates hTERT transcription by directly interacting with its promoter, leading to increased promoter activity (28 -31). Moreover, recent studies have demonstrated that c-Myc and SP1 cooperate to induce hTERT promoter activity in several human cancer cell lines (32). Also, other cellular factors such as p53, Rb, protein phosphatase 2A, and Akt have been implicated in the regulation of telomerase in various cancer cells (33)(34)(35). However, the molecular mechanisms of telomerase regulation in human lung cancers remain unknown.
In the past decade, sphingolipids have been shown to play an important role in signal transduction, mediating stress responses, and regulation of growth (36,37). Previous studies have shown that ceramide is involved in mediating important cellular activities such as induction of cell differentiation, growth arrest, senescence, and apoptosis in some human cancer cells (38). Ceramide has been shown to regulate the activity of various biochemical and molecular targets, including serine/ threonine phosphatases of the protein phosphatase 2A and PP1 families designated as CAPP, KSR, PKC-, Rb, JNK, phospholipase D, Akt, and effector caspases, and to down-regulate c-Myc (38). Recently, we have shown that endogenous ceramide is involved in mediating the specific inhibition of telomerase activity in response to ceramide, and that this inhibition is independent of apoptotic cell death in A549 human lung adenocarcinoma cells (39). However, the mechanisms of ceramideinduced telomerase inhibition in A549 cells are unknown. Therefore, in this study we attempted to identify the molecular mechanisms triggered by ceramide that are involved in the regulation of telomerase activity.

EXPERIMENTAL PROCEDURES
Cell Lines and Culture Conditions-The A549 human lung carcinoma cells were obtained from Dr. Alice Boylan (Medical University of South Carolina, Charleston, SC). Cells were maintained in growth medium containing 10% fetal calf serum and 100 ng/ml each of penicillin and streptomycin (Life Technologies Inc., Grand Island, NY) at 37°C in 5% CO 2 . Cell permeable and biologically active short chain ceramide (C 6 -ceramide) was obtained from the Synthetic Lipid Core at the Department of Biochemistry and Molecular Biology, Medical University of South Carolina.
Determination of Telomerase Activity-Telomerase activity in cell extracts were measured by the PCR-based telomere repeat amplification protocol (TRAP) using TRAPeze kit (Intergen, Gaithersburg, MD) which includes a 36-base pair internal control to allow quantification of activity as described by the manufacturer. Briefly, the cells, grown in 6-well plates, were washed in phosphate-buffered saline and homogenized in 1 ϫ CHAPS lysing buffer for 30 min on ice. Then 50 -100 ng of proteins from each cell extract were analyzed in the TRAP reaction. The cell extracts were added directly to the TRAP reaction mixture containing dNTPs, TS primer (6 ϫ 10 5 cpm), reverse primer mixture, and Taq polymerase. Then, the extended telomerase products were amplified by two-step PCR (94°C for 30 s, 60°C for 30 s) for 27 cycles. The telomerase activity in each sample was quantitated by measuring the ratio of the 36-base pair internal standard to the extended telomerase products as described by the manufacturer using ChemiImager (Alpha Innotech Corp., San Leandro, CA).
Isolation of RNA, RT-PCR, and Northern Blot Analysis-The mRNA levels of hTERT and hTR were analyzed by RT-PCR and Northern blotting as described previously (40,41). One g of total RNA, isolated using a RNA isolation kit (Qiagen), was used in reverse transcription reactions as described by the manufacturer. The resulting total cDNA was then used in the PCR to measure the mRNA levels of hTERT and hTR using primers and conditions as described (40). The mRNA levels ␤-actin and rRNA were used as internal controls (40). Linear amplification cycles were determined separately for each gene as described elsewhere (40). For Northern blotting, 4 -5 g of total RNA separated on 0.8% denaturing agarose gels was transferred to a nylon membrane and then was probed with the hTERT-specific 32 P-labeled PCR product as described previously (41), except that Express Hybridization Buffer (CLONTECH) was used for blotting.
Electrophoretic Mobility Shift Assay (EMSA)-Nuclear extracts (8 -10 g of protein) isolated as described previously (42) from cells grown in the presence or absence of various concentrations of C 6ceramide were preincubated in 15 l of binding buffer (42) containing 1.5 g of poly(dI⅐dC) (Amersham Pharmacia Biotech, Piscataway, NJ) with or without oligonucleotides used as competitors at 25°C for 15 min. Then, 3-5 ng of 5Ј-end labeled DNA probes (60,000 cpm) were added to the reaction and incubated at 25°C for 15 min. End labeling of DNA fragments with [␥-32 P]ATP (Amersham Pharmacia Biotech) was performed using T4 DNA kinase (Promega). Competitor double stranded oligonucleotides (Santa Cruz) were used at 10 -100-fold molar excess. The reaction mixtures were separated on 5% native polyacrylamide gels and visualized by autoradiography (43,44). The 5Ј-3Ј sequences of double stranded oligomers, which were synthesized and annealed by Integrated DNA Technologies, Inc. (Coralville, IA), contained the hTERT-E-box and its mutated form: hTERT-E-box, GGGC-TAGCGCGCTCCCCACGTGGCGGAGGGAAAGCTTCC; hTERT-Mut-E, GGGCTAGCGCGCTCCCTTTGTGGCGGAGGGAAAGCTTCC.
The hTERT Reporter Plasmids-The pGL3-Basic plasmids containing hTERT promoter fragments (26) spanning Ϫ279 to ϩ5 which is the minimal functional core promoter designated as pBTdel-279, the fragment Ϫ149 to ϩ5 which lacks the functional c-Myc recognition sequence designated as pBTdel-149, or the fragment Ϫ211 to ϩ40 which contains two c-Myc binding sequences designated as p2XEB were all kindly provided by Dr. J. C. Barrett (NIEHS, National Institutes of Health, Research Triangle Park, NC).
Transient Transfections and Luciferase Reporter Assay-The cells were co-transfected with 2.5-3 g/well of hTERT reporter plasmids constructed as described above and pSV-␤-galactosidase control plasmid (Promega) by a cationic liposome-mediated transfection method using the Effectine transfection kit (Qiagen) for 6 -8 h as described by the manufacturer. After the transient transfectants were recovered in fresh media for 18 h, luciferase and ␤-galactosidase activities were measured using Promega's luciferase and ␤-galactosidase enzyme assay systems as described by the manufacturer. The light intensity of the luciferase reactions measured in the lysates of the transient transfectants were normalized to their ␤-galactosidase activity, used as an internal control (43,44).
Inhibition of Protein Synthesis Using Cycloheximide-The effects of protein synthesis inhibition using cycloheximide (CHX) on c-Myc expression in human lung cancer cell lines were investigated as described previously (40). The optimal CHX concentrations that inhibit 95% of the protein synthesis in each cell line (grown in 24-well plates) were determined by measuring the total incorporation of [ 35 S]methionine (4 Ci/ ml) into trichloroacetic acid-precipitable macromolecules by scintillation counting (40). The cells then were pretreated with CHX at its optimum concentrations for various time points in the absence or presence of C 6 -ceramide before immunoprecipitation or Western blotting as described below.
Western Blotting-The protein levels of c-Myc and hTERT were detected by Western blot analysis. In short, total proteins (20 -50 g/lane) were separated by 5-15% SDS-PAGE (Bio-Rad), blotted onto an Immobilon membrane, and c-Myc, Max, Mad1, and hTERT proteins were detected using 1 g/ml of rabbit polyclonal anti-c-Myc, Max, and Mad1 (Santa Cruz Biotechnology), and anti-hTERT (Abcam, Cambridge, UK) antibodies, and peroxidase-conjugated secondary anti-rabbit antibody (1:2500). The proteins were visualized using the ECL protein detection kit (Amersham Pharmacia Biotech) as described by the manufacturer. Equal loading were confirmed by Coomassie Blue staining of SDS-PAGE strips cut from gels containing 50 g of protein/lane of each sample prior to blotting.
Immunoprecipitations-Subconfluent cells (70 -80% confluent) were metabolically labeled with [ 35 S]methionine in methionine-free growth medium at 37°C in 5% CO 2 for 2-4 h. Following phosphate-buffered saline washes, the cells were resuspended in 1 ϫ CHAPS or RIPA buffer and frozen at Ϫ80°C (45). Equal trichloroacetic acid counts of the lysates were used for immunoprecipitations as described previously (45). After the lysates were precleared with 20 -50 l of protein A-Sepharose (Life Technologies) at 4°C for 16 h, they were incubated at 4°C for 2-4 h with 20 g of the rabbit polyclonal antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) which recognize hTERT or Mad1. The immunoprecipitation of c-Myc was performed using the agaroseconjugated mouse monoclonal anti-c-Myc antibody (Santa Cruz). The immunocomplexes were then precipitated using protein A-Sepharose for 16 h at 4°C. The Sepharose beads were washed with the CHAPS buffer to remove nonspecific binding. Then the immune complexes were analyzed by SDS-PAGE followed by autoradiography.
Overexpression of b-SMase-The full-length b-SMase cDNA fragment was cloned into the pEGFPN1 mammalian expression vector system (Invitrogen) upstream of the green fluorescent protein (GFP) sequence as described (46). Transient transfections of A549 cells were performed using the effectene transfection reagent as described by the manufacturer (Qiagen).

Analysis of the Half-life and Ubiquitination of c-Myc Protein-
The half-life of c-Myc protein was analyzed by immunoprecipitation, Western blot, or pulse-chase analysis with [ 35 S]methionine labeling following CHX treatments as described previously (47,48) using anti-c-Myc antibodies (Santa Cruz). The half-life of the proteins was measured by densitometry of x-ray films. To analyze the ubiquitin conjugates, c-Myc protein was immunoprecipitated with the protein A-agarose-conjugated rabbit polyclonal c-Myc antibody followed by Western blotting using goat polyclonal anti-ubiquitin or c-Myc antibodies after CHX treatments. The involvement of proteasome 26S in the degradation of ubiquitin-conjugated c-Myc was examined by using lactacystin (Calbiochem) which is a specific inhibitor of proteasome, and E-64 (transepoxysuccinyl-L-leucylamido-(4-guanidino)-butane (Roche Molecular Biochemicals), which is a cysteine protease inhibitor, used as a control, prior to CHX treatments in the experiments described above.

RESULTS
C 6 -ceramide Inhibits Telomerase Activity-Our previous results showed that endogenous ceramide mediates the inhibition of telomerase activity in response to daunorubicin, and that this inhibition is independent of cell toxicity or apoptotic cell death (39). Consistent with these results, Fig. 1A shows that telomerase activity was slightly decreased after 3 h treatment with exogenously applied C 6 -ceramide (20 M), whereas treatment of the cells with C 6 -ceramide for 6 and 24 h resulted in a significant inhibition of telomerase activity by about 50 and 85%, respectively. Moreover, C 6 -ceramide had no effect on telomerase activity in vitro (data not shown), demonstrating that ceramide is indirectly involved in the inhibition of telomerase.  (45). As shown in Fig. 1B, this antibody successfully recognized the 132-kDa hTERT protein using our protocol. These results also showed that hTERT protein level was significantly decreased in response to C 6ceramide treatment at 20 M for 24 h compared to that of untreated A549 cells (Fig. 1B, lanes 2 and 1, respectively). C 6 -ceramide Decreases hTERT mRNA Expression-To examine the effects of ceramide on the transcription of hTERT, A549 cells were treated in the absence or presence of 20 M C 6ceramide for various time points (0 -24 h), and the mRNA levels of hTERT, hTR, rRNA, and ␤-actin were measured by semiquantitative RT-PCR. The linearity of amplification cycles in each PCR was established using 32 P-labeled primers (data not shown) as described under "Experimental Procedures." As seen in Fig. 1C, hTERT mRNA level was significantly decreased in response to C 6 -ceramide. This down-regulation of hTERT mRNA by C 6 -ceramide was time dependent: a slight decrease in hTERT mRNA was observed after 3 and 6 h C 6ceramide treatment (about 5 and 14%, respectively), whereas its mRNA level was significantly reduced at 24 h in response to C 6 -ceramide (about 84%) compared with that of controls (Fig.  1C, second panel from the top, lanes 1-4). On the other hand, C 6 -ceramide had no effect on the mRNA levels of hTR (Fig. 1C, third panel from the top). The mRNA levels of ␤-actin and rRNA were used as internal controls, and their levels were  1-4, respectively) in 5% CO 2 at 37°C, and telomerase activity in each sample was detected by the TRAP assay as described under "Experimental Procedures." The 36-base pair internal standard was used as a control in the TRAP assay. B, hTERT protein levels were detected in the absence or presence of C 6 -ceramide at 24 h (lanes 1 and 2, respectively) by immunoprecipitation as described under "Experimental Procedures." C, mRNA levels of rRNA, hTERT, hTR, and ␤-actin in response to C 6 -ceramide at 0, 3, 6, and 24 h (lanes 1-4, respectively) were measured by the semi-quantitative RT-PCR as described under "Experimental Procedures". D, Northern blot analysis was performed to measure the mRNA levels of hTERT in response to C 6 -ceramide at 0, 3, 6, and 24 h (lanes 2-5, respectively) using 4 g of total RNA separated in 0.8% denaturing agarose gels and then blotted on a nylon membrane (upper band) before hybridizations with the 32 P-labeled hTERT-specific probe (lower panel) as described under "Experimental Procedures." In D, lane 1 contains total RNA from Wi-38 cells used as negative controls. D, upper panel, shows the undegraded total RNA samples as detected by the ethidium bromide staining of intact 28 S and 18 S rRNA bands. The figures shown are representative of at least two independent experiments. similar in each sample (Fig. 1C). Similar results, which showed that hTERT mRNA expression was inhibited by about 25 and 90% after 6 and 24 h C 6 -ceramide treatment, were obtained by Northern blot analysis (Fig. 1D, lanes 2-5). The RNA of Wi-38 lung fibroblast cell line which do not express hTERT was used as a negative control in Northern blotting (Fig. 1D, lane 1). C 6 -ceramide Reduces the Binding of c-Myc/Max Transcription Factors to the E-box Region of the hTERT Promoter-In an attempt to determine whether the inhibition of hTERT mRNA expression in A549 cells by ceramide was associated with decreased DNA binding activity of c-Myc/Max transcription factor, we performed EMSA using a double-stranded oligonucleotide containing the E-box motif (CACGTG) on the hTERT promoter sequence spanning Ϫ173 to Ϫ152 region as a probe as described under "Experimental Procedures." The cells were treated with 20 M C 6 -ceramide for 24 h, nuclear extracts were isolated, and 5 g of the extracts were used in EMSAs. As seen in Fig. 2A, our results show that the DNA binding activity of c-Myc in A549 nuclear extracts (lane 1) was greatly inhibited by C 6 -ceramide treatment (lane 4). The specificity of c-Myc binding to the E-box region of the hTERT promoter was confirmed by the complete competition of the c-Myc⅐DNA complex with the presence of cold oligomer containing hTERT E-box region but not with a double-stranded oligonucleotide containing the mutated E-box sequence used as a nonspecific competitor ( Fig. 2A, lanes 2 and 3, respectively). Moreover, as Fig. 2A shows, the presence of c-Myc in the protein-DNA complex was confirmed with the supershift of the DNA/protein band (lane 5) in response to the incubation of A459 nuclear extracts with the rabbit polyclonal c-Myc antibody (Santa Cruz Biotechnology) prior to the addition of the probe in EMSA. The use of rabbit IgG in these experiments did not result in the supershifted c-Myc band (data not shown). To determine whether the inhibition of c-Myc DNA binding activity by C 6 -ceramide is specific and not due to protein degradation during the isolation of nuclear extracts, we examined the DNA binding activity of NF-B by EMSA as described (44). Our results show that NF-B-DNA interaction was slightly increased in C 6 -ceramide-treated A549 compared with untreated A549 cells (Fig. 2B,  lanes 1 and 4, respectively). Taken together, these results show that C 6 -ceramide treatment results in inhibiting the interaction between E-box region of the hTERT promoter and c-Myc/ Max transcription factor in these cells.
C 6 -ceramide Decreases hTERT Promoter Activity in Situ-In order to confirm the EMSA results in situ, we performed transient transfections of A549 cells with luciferase-plasmid constructs containing the proximal promoter region of the hTERT promoter (pBTdel-279) and its E-box deleted form (pBTdel-149) in the absence or presence of 20 M C 6 -ceramide for 24 h. As seen in Fig. 2C, hTERT promoter activity was 78-fold increased in cells transfected with the pGL3-hTERT promoter vector compared with controls. However, in the absence of E-box region, its promoter activity was reduced about 3-fold, indicating the role of c-Myc binding in increased hTERT promoter activity in A549 cells. Moreover, when the pGL3-hTERT-transfected cells were treated with C 6 -ceramide at 20 M for 24 h, hTERT promoter activity was reduced about 4-fold compared with that of cells grown in the absence of ceramide. Moreover, no difference in the activity of the E-box-deleted hTERT promoter was observed in the presence or absence of C 6 -ceramide in these cells. Therefore, these results show clearly that the presence of E-box motif, which is recognized by c-Myc/Max, is necessary for decreased hTERT promoter activity in response to ceramide. The presence of a second E-box motif on the hTERT promoter fragment, spanning the ϩ22 to ϩ27 region, in p2XEB plasmid had no significant effect on the hTERT promoter activity with or without ceramide (Fig. 2C).  1 and 4, respectively) were detected by EMSA as described above using 32 P-labeled oligonucleotides containing the NF-B recognition sequence. Lanes 2 and 3 contain EMSA results obtained in the presence of cold NF-B and c-Myc oligomers used as specific and nonspecific competitors, respectively. C, to determine the effects of C 6 -ceramide on hTERT promoter activity in situ, A549 cells were transiently transfected with pGL3-Basic plasmid constructs containing the core hTERT promoter (pGL3-pBT-del 279), the E-box-deleted hTERT promoter (pGL3-pBT-del-149), and hTERT core promoter containing a second E-box motif downstream of the major transcription initiation site at position ϩ22 to ϩ27 (p2XEB), upstream of luciferase sequence. The cells were co-transfected with the ␤-galactosidase plasmid used to normalize the luciferase levels in the assay performed as described under "Experimental Procedures." Standard deviations of luciferase activity assays are shown in parentheses. The results shown are representative of at least two independent experiments. c-Myc protein (p64), which represents the major translational product, is produced from an ATG start codon in exon 2 (49). The results show that C 6 -ceramide had no effect on the mRNA levels of c-myc (Fig. 3A), whereas it caused a significant decrease in p64, but not p67, c-Myc protein levels (about 3.2-fold) compared with untreated controls (Fig. 3B, lanes 2 and 1,  respectively). There was no significant change in the levels of Max proteins in these samples (Fig. 3C).
In order to determine whether the decreased c-Myc protein levels in response to C 6 -ceramide treatment is due to decreased protein stability, the half-life of c-Myc (p64) was measured as described under "Experimental Procedures." First, the optimum concentration of cycloheximide that inhibited Ͼ90% of protein synthesis was determined by measuring the incorporation of [ 35 S]methionine into trichloroacetic acid-precipitable macromolecules in the presence of various concentrations of CHX as described under "Experimental Procedures," and found to be 500 g/ml (data not shown). Then, the half-life of c-Myc protein in the absence or presence of C 6 -ceramide at various time points following CHX treatment was analyzed by Western blotting using the anti-c-Myc antibody as described under "Experimental Procedures." The results showed that the half-life of c-Myc was about 54 (Ϯ9) min in A549 cells (Fig. 4A, lanes 1-5), whereas it was reduced to about 28 (Ϯ4) min in the presence of 20 M C 6 -ceramide (Fig. 4A, lanes 6 -9). Similar results were obtained also with pulse-chase studies in which the half-life of c-Myc was determined as 48 and 25 min in the absence or presence of C 6 -ceramide, respectively, by immunoprecipitations following metabolic labeling of the cells with [ 35 S]methionine and CHX treatment at various time points (Fig. 4B, lanes  1-7). Moreover, to analyze whether ceramide-induced c-Myc degradation is specific, we examined the effects of ceramide on the half-life of Mad1 protein by immunoprecipitations followed by Western blotting as described above. Our results showed that ceramide had no detectable effect on the half-life of Mad1, which was around 15 min in the absence or presence of 20 M C 6 -ceramide (Fig. 4C, lanes 1-10).
C 6 -ceramide Induces the Rapid Degradation of c-Myc Protein by the Ubiquitin/Proteasome Pathway-To determine whether the reduced half-life of c-Myc in response to C 6 -ceramide was due to the activation of the ubiquitin/proteasome pathway, we examined the ubiquitination of c-Myc protein by immunoprecipitation followed by Western blotting using agarose-conjugated mouse monoclonal anti-c-Myc and goat polyclonal antiubiquitin antibodies, respectively, as described under "Experimental Procedures." As seen in Fig. 5A, the reduced c-Myc protein levels at 60 min in the absence of ceramide and at 30 -60 min in the presence of ceramide (upper panel) correlated with its increased ubiquitination at these time points (lower panel) compared with untreated controls (lanes 1-5). These results demonstrate that the decreased c-Myc protein level in response to ceramide is associated with its increased ubiquitination which most likely induces its rapid proteolysis. Indeed, pretreatment of cells with 5 M lactacystin, an inhibitor of 26 S proteasome, for 2 h before CHX addition, prevented the rapid degradation of c-Myc in response to C 6 -ceramide at 30 min (Fig. 5B, lanes 1-3), confirming the previously published reports that 26 S proteasome is involved in the rapid degradation of highly ubiquinated c-Myc (47,48) in these cells. Moreover, we also examined the effects of E-64, a cysteine protease inhibitor, on ceramide-induced c-Myc degradation, and found that pretreatment of cells with E-64 did not have a significant effect on the degradation of c-Myc in response to 20 M C 6ceramide at 30 min (Fig. 5C, lanes 1-3). These results further demonstrate that ceramide mediates the rapid degradation of c-Myc by ubiquitin/proteasome pathway, and that this process is not merely due to the activation of a nonspecific proteasedependent degradation.
Daunorubicin, Which Increased Endogenous Ceramide Levels, Inhibits Telomerase, and c-Myc Activities-In order to determine whether known inducers of endogenous ceramide mediate telomerase inhibition through c-Myc inactivation, A549 cells were treated in the absence or presence of 1 M DNR for 6 h and its effects on ceramide generation, and telomerase and c-Myc activities were determined using the TRAP and EMSA as described above. The results show that DNR caused a significant elevation of endogenous ceramide levels at 6 h as determined by the diacylglycerol kinase assay (Fig. 6A) which correlated with almost complete inhibition of telomerase activity (Fig. 6B, lanes 1-3) detected by the TRAP assay. Moreover, EMSA results demonstrated that treatment of cells with DNR (1 M for 6 h) almost completely abolished DNA binding activity of c-Myc⅐Max complex (Fig. 6C, lanes 7 and 8). In Fig. 4C, the specificity of c-Myc/Max binding in gel-shift assays were confirmed by competition of protein-DNA complex by cold oligomers containing the commercial E-box (Santa Cruz) and hTERT-specific E-box sequences (lanes 2 and 4, respectively) but not with their mutated forms or cold SP1 oligomer (lanes 3 ,  5, and 6, respectively). This inactivation of c-Myc⅐Max complex in response to DNR was due to decreased nuclear c-Myc protein levels as shown in Fig. 6D by Western blot analysis as described above.  3. The role of C 6 -ceramide on c-Myc expression. A, the mRNA expression levels of ␤-actin and c-myc in response to C 6 -ceramide at 0, 3, 6, and 24 h (lanes 2-5 and 6 -9, respectively) were measured by semi-quantitative RT-PCR as described under "Experimental Procedures." The c-Myc (B) and Max (C) protein levels in the absence or presence of 20 M C 6 -ceramide at 24 h (lanes 1 and 2, respectively) were measured by Western blotting using rabbit polyclonal anti-c-Myc and Max antibodies as described under "Experimental Procedures."

b-SMase Overexpression Decreases the Levels of c-Myc and
hTERT Proteins and Telomerase Activity-The ability of endogenously generated ceramide to induce c-Myc degradation was more specifically determined in A549 cells by transient transfections using an expression vector containing the full-length b-SMase cDNA cloned upstream of the GFP, which results in the hydrolysis of sphingomyelin and elevation of endogenous ceramide. Overexpression of b-SMase, as detected by Western blotting and fluorescence microscopy (Fig. 7A, lanes 1-2, and  panels 3-4, respectively), caused about 40% increased endogenous ceramide levels (data not shown), which correlated with about 85% inhibition of telomerase activity compared with control transfectants (Fig. 7B, lanes 1-2). Importantly, the overexpression of b-SMase resulted in a significant decrease in hTERT and c-Myc protein levels (Fig. 7, C and D).  1-5, respectively) or presence of C 6 -ceramide (lanes 6 -9, respectively) by Western blot analysis. B, the half-life of c-Myc protein was determined by pulse-chase labeling with [ 35 S]methionine and then treating the cells with CHX at 0, 30, and 60 min in the absence (lanes 1-3, respectively) or at 0, 15, 30, and 60 min in the presence of C 6 -ceramide (lanes 4 -7, respectively) as described under "Experimental Procedures." C, the half-life of Mad1 protein in the absence or presence of C 6 -ceramide was determined after CHX treatments at time points shown by immunoprecipitations followed by Western blot analysis as described above. The results shown are representative of two independent experiments.  2 and 3, respectively) as described above. C, the c-Myc protein levels after 30 min CHX treatment in the absence (lane 1) or presence of C 6 -ceramide without or with 5 M E-64, a cysteine protease inhibitor, were detected by immunoprecipitation (lanes 2 and 3, respectively) as described above.

FIG. 6. The effects of DNR on telomerase and c-Myc activities.
A, the effects of 1 M DNR at 0, 3, and 6 h (lanes 1-3) on the total endogenous ceramide levels were measured by the diacylglycerol kinase assay as described under "Experimental Procedures." The amounts of ceramide were normalized to the inorganic phosphate levels determined as described under "Experimental Procedures." B, the effects of DNR on telomerase activity using 1 M DNR at 0, 3, and 6 h (lanes 1-3) were determined by the TRAP assay as described above. C, the role of DNR on the DNA binding activity of c-Myc transcription factor at 0 and 6 h (lanes 7 and 8, respectively) by the EMSA as described above. Lanes 1-6 in Fig. 5C show EMSA results obtained using nuclear extracts incubated in the presence of 32 P-labeled E-box oligomer without or with cold E-box, mutated E-box, hTERT-E-box region, mutated hTERT-Ebox region, and SP-1 oligomers used as competitors, respectively. D, the levels of c-Myc protein in the absence or presence of 1 M DNR at 6 h were measured by Western blotting (lanes 1 and 2, respectively) as described above. The results shown are representative of two independent experiments.

DISCUSSION
This study shows for the first time that telomerase regulation by ceramide involves decreased hTERT promoter activity via c-Myc inactivation. The results provide novel data demonstrating that c-Myc inactivation by ceramide is due to its reduced half-life through rapid proteolysis of highly ubiquitinated c-Myc protein in A549 cells.
It has been well established that telomerase is not active in most somatic tissues, whereas it is activated in majority of the lung cancer-derived cell lines and malignant tumors, suggesting that telomerase plays an important role in cell immortalization and tumorigenesis (16,17). Previous studies and the results of this study show that telomerase inhibition correlates with decreased hTERT mRNA levels in most of the human cancer cell lines and tissue samples (12,23). The regulation of hTERT promoter has been established as one of the main mechanisms for the control of hTERT mRNA levels (24 -27), and c-Myc has been shown to directly bind the hTERT promoter resulting in its activation (28 -31). The down-regulation of hTERT promoter activity by repression of c-Myc was demonstrated in previous studies (31). The ability of c-Myc to function as a transcription factor depends on its dimerization with another protein Max, and this interaction is mediated by HL-HZip domains of the two proteins and enables the Myc/Max dimer to recognize the CACGTG or related DNA sequences known as E-box motifs (50,51). In addition to c-Myc, other proteins have been shown to form dimers with Max, and these other bHLHZip proteins include Mad1, Mxi1 (Mad2), Mad3, and Mad 4, collectively referred to as Mad family, and Mnt Rox protein (52)(53)(54). In contrast to Myc/Max, these Mad⅐Max complexes act as repressors of transcription for promoters containing E-box motifs. Interestingly, it has been shown previously that the activation of Mad1 by 12-O-tetradecanoylphorbol-13acetate treatment decreased hTERT promoter-driven reporter gene activity in U937 cells (55). Similar results on repression of hTERT promoter by Mad1 in various immortal and mortal cells was also reported recently (56,57). However, in our studies, the repression of hTERT promoter was independent of changes in the Mad⅐Max complexes, but rather was dependent on the decreased c-Myc protein levels in response to ceramide in A549 cells.
Ceramide has been shown to down-modulate c-Myc mRNA levels in various cancer cell lines (37,42,59). In our experiments, however, c-Myc inactivation by ceramide was shown to be at the protein level, but not at the mRNA level in A549 cells. Further analysis of post-transcriptional/translational mechanisms involved in the reduced c-Myc protein levels showed that the half-life of c-Myc was reduced significantly in response to ceramide. There are data which demonstrate that c-Myc has a very short half-life, about 20 to 120 min in various cancer cell lines, and that the proteolysis of c-Myc is mediated by the ubiquitin/proteasome pathway in vivo (47). In addition, it has been shown previously that expression of the high risk papillomavirus oncoprotein in the NGP human neuroblastoma cell line results in significant shortening of the half-life of c-Myc protein with subsequent decrease in its protein level (48). It has also been shown that phytosphingosine, a proposed sphingolipid second messenger in Saccharomyces cerevisiae, mediates the activation of ubiquitin/proteasome pathway in response to heat stress (60). In accordance with these results, our results show, for the first time, that ceramide mediates the activation of ubiquitin/proteasome pathway, not a general protease-dependent degradation, which in turn involves in rapid proteolysis of c-Myc protein, resulting in its decreased levels in A549 cells. These results imply that the activation of ubiquitin/proteasome pathway, which is important in regulating the function of a number of cellular factors, by sphingolipids is not limited to yeast cells but is conserved also in human cancer cells. Moreover, it is interesting that the half-life of another short-lived protein Mad1 (61) is not effected by ceramide treatment, which suggest that the reduced half-life of c-Myc in response to ceramide is not a constitutive process, but is specific and regulated.
In this study, we showed that the E-box motif that is recognized by c-Myc (CACGTG, spanning a region between Ϫ187 to Ϫ182) plays an important role for the activity of hTERT promoter in A549 cells. However, although the absence of this E-box motif reduced hTERT promoter activity significantly, the promoter activity in cells transfected with pBT-del 149 plasmid missing the E-box region was still higher than that of controls, indicating that some other factors are also involved, either directly or indirectly, in the activation of hTERT promoter in these cells. These observations are consistent with current data which suggest that some other factors are also important for the activation of hTERT promoter in various cancer cells. Indeed, a novel transcription factor MZF-2 that negatively regulates hTERT promoter was recently identified (62). Interestingly, ceramide had no effect on the activity of hTERT promoter lacking the Myc-responsive E-box. In addition, it has been recently reported that DNA methylation and/or histone deacetylation might be involved in the negative regulation of hTERT promoter, indicating that the hTERT is regulated by multiple mechanisms in various human immortal and mortal cell lines (63). The involvement of these factors, however, in ceramide-mediated telomerase inhibition needs to be determined.
Also, our results show that the 50% inhibition of telomerase in response to ceramide was observed at 6 h treatment at which hTERT mRNA levels were about 25% less than controls. These results suggest that some other mechanisms may also be involved in the early inhibition of telomerase activity in A549 cells in response to ceramide. In this context, the phosphorylation status of telomerase is an important posttranslational mechanism in regulating its function. It has A549 cells were transiently transfected with plasmids containing the full-length b-SMase cDNA upstream of GFP and then the overexpression of b-SMase/GFP was detected by Western blotting (Fig. 5A, lanes 1 and 2) and fluorescence microscopy (Fig. 5A, panels 3 and 4). Then, the effects of its overexpression on telomerase activity (B), and the expression levels of hTERT and c-Myc proteins (C and D, respectively) compared with control transfectants (lanes 2 and 1, respectively) were determined by the TRAP and Western blot analyses as described under "Experimental Procedures." been shown that the serine/threonine protein phosphatase 2A is involved in negative regulation of telomerase by dephosphorylating both hTERT and hTEP1, whereas these proteins are rephosphorylated by PKC-␣, resulting in telomerase activation (33,34). Moreover, there is evidence that Akt kinase is also involved in inducing telomerase activity, probably by phosphorylating the serine residue at position 824 in human melanoma cells (35). Since ceramide is known to activate protein phosphatase 2A, it is possible that the treatment of cells with ceramide might affect the phosphorylation status of hTERT protein at earlier time points. This is subject on ongoing studies in our laboratory.
There is also evidence that, after translation, the assembly and maintenance of a functionally active telomerase enzyme might be controlled by direct protein-protein interactions. One of these proteins, TEP1, has been shown to serve as a scaffold for organizing the assembly of telomerase holoenzyme (13). In addition, recombinant p53 tumor suppressor protein has been shown to interact with TEP1 and inhibit telomerase activity in vitro, suggesting that telomerase might be a downstream target of p53 (64). Additional studies have shown that increased telomerase activity correlates with mutated p53 status in various human cancer cell lines (65,66), however, in some human cancer cells the overexpression of wild-type p53 by transfections did not have any effect on telomerase activity (58). These studies also indicate that the control of telomerase activity by p53 might be tissue specific. Therefore, it is becoming evident that telomerase activity is regulated by complex mechanisms involving various different cellular factors in mammalian cells by transcriptional and post-translational mechanisms. Thus, identification of ceramide-mediated mechanisms involved in regulating telomerase activity is necessary for cancer research and therapy.
In summary, this study provides data showing for the first time that the regulation of telomerase by ceramide involves complex mechanisms including the activation of ubiquitin/ proteasome pathway that mediates the rapid degradation of c-Myc transcription factor which in turn results in decreased hTERT promoter activity and reduced hTERT gene transcription in A549 cells. We believe that these results strongly support the anti-proliferative function of ceramide, and its role in cellular senescence, and suggest that ceramide is a potential upstream candidate for the regulation of telomerase in A549 cells.