Liver Biology and Pathobiology---Pathobiology
S-adenosylmethionine and its metabolite induce apoptosis in HepG2 cells: Role of protein phosphatase 1 and Bcl-xS
Article first published online: 30 JUN 2004
Copyright © 2004 American Association for the Study of Liver Diseases
Volume 40, Issue 1, pages 221–231, July 2004
How to Cite
Yang, H., Sadda, M. R., Li, M., Zeng, Y., Chen, L., Bae, W., Ou, X., Runnegar, M. T., Mato, J. M. and Lu, S. C. (2004), S-adenosylmethionine and its metabolite induce apoptosis in HepG2 cells: Role of protein phosphatase 1 and Bcl-xS. Hepatology, 40: 221–231. doi: 10.1002/hep.20274
- Issue published online: 30 JUN 2004
- Article first published online: 30 JUN 2004
- Manuscript Accepted: 21 MAR 2004
- Manuscript Received: 21 NOV 2003
- National Institutes of Health. Grant Numbers: DK51719, AA12677, AT1576, DK56040
- Plan Nacional de. Grant Number: I+D 2002-00168
- Cell Culture Core of the University of Southern California Research Center for Liver Diseases. Grant Number: DK48522
- Postdoctoral Fellowship of the Training Program in Alcoholic Liver and Pancreatic Diseases. Grant Number: T32 AA07578
S-adenosylmethionine (SAMe) and its metabolite 5′-methylthioadenosine (MTA) are proapoptotic in HepG2 cells. In microarray studies, we found SAMe treatment induced Bcl-x expression. Bcl-x is alternatively spliced to produce two distinct mRNAs and proteins, Bcl-xL and Bcl-xS. Bcl-xL is antiapoptotic, while Bcl-xS is proapoptotic. In this study we showed that SAMe and MTA selectively induced Bcl-xS in a time- and dose-dependent manner in HepG2 cells. There are three transcription start sites in the human Bcl-x gene which yield only Bcl-xL in control HepG2 cells. SAMe and MTA treatment did not affect promoter usage, but while one promoter yielded only Bcl-xL, the other two yielded both Bcl-xL and Bcl-xS, with Bcl-xS as the predominant messenger RNA (mRNA) species. Trichostatin A, 3-deaza-adenosine, cycloleucine, and okadaic acid had no effect on Bcl-xS induction by SAMe or MTA. Calyculin A and tautomycin, on the other hand, blocked SAMe and MTA-mediated Bcl-xS induction and apoptosis in a dose-dependent manner. SAMe and MTA increased protein phosphatase 1 (PP1) catalytic subunit mRNA and protein levels and dephosphorylation of serine–arginine proteins, with the latter blocked by calyculin A. The effects of SAMe and MTA on Bcl-xS, PP1 expression, and apoptosis were also seen in 293 cells, but not in primary hepatocytes. Induction of Bcl-xS by ceramide in HepG2 cells also resulted in apoptosis. In conclusion, we have uncovered a highly novel action of SAMe and MTA, namely the ability to affect the cellular phosphorylation state and alternative splicing of genes, in this case resulting in the induction of Bcl-xS leading to apoptosis. Supplementary material for this article can be found on the HEPATOLOGY website (http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html). (HEPATOLOGY 2004;40:221–231.)
S-adenosylmethionine (SAMe), which is generated as the first product of methionine metabolism in mammals, is the principal biological methyl donor present in all cells.1, 2 Although present in all tissues, SAMe is mainly synthesized and consumed in the liver, where it serves as a methyl donor in numerous metabolic reactions and as a precursor for glutathione and polyamines.1, 2 In the biosynthesis of polyamines, 5′-methylthioadenosine (MTA) is generated as an end product.3 SAMe synthesis is impaired in experimental models of liver injury and patients with cirrhosis.2 SAMe administration protects against experimental liver injury and improves survival in patients with less severe alcoholic cirrhosis.2, 4 Remarkably, SAMe has also been shown to prevent the development of liver tumors induced by a variety of hepatocarcinogens, which was accompanied by an increase of apoptotic bodies.5–7 Interestingly, the chemopreventive property of SAMe can be mimicked by MTA.5
In addition to being a methyl donor, accumulating evidence shows that SAMe modulates liver growth, response to apoptotic signals, and is critical in maintaining a normal liver phenotype.2, 8–12 In normal hepatocytes, SAMe inhibits the mitogenic response to hepatocyte growth factor.8 In liver cancer cell lines HepG2 and Huh-7, SAMe inhibits their growth directly,9 which can be partly explained by its proapoptotic effect in these cells.10 The proapoptotic effect of SAMe is mimicked by MTA and occurs in HepG2 and Huh-7 cells but not primary hepatocytes.10 In the methionine adenosyltransferase 1A knockout mice, chronic hepatic SAMe deficiency results in predisposition to liver injury and development of hepatocellular carcinoma.11, 12 Taken together, these studies illustrate the complexity of SAMe's biological action independent of being a methyl donor, particularly because MTA is not a methyl donor and can inhibit methyltransferases.13
SAMe and MTA-induced apoptosis of HepG2 and Huh-7 cells is accompanied by the release of cytochrome c, implicating the involvement of mitochondria in apoptosis.10 Bcl-2 family members are known to modulate apoptosis at the mitochondrial level.14 In preliminary studies using microarray analysis (Atlas Human 1.2 Array, Clontech, Palo Alto, CA), we found Bcl-x to be up-regulated in HepG2 cells in response to SAMe treatment. Bcl-x is alternatively spliced to produce two major distinct messenger RNAs (mRNAs) and variant proteins, Bcl-xL and Bcl-xS, that have antagonistic functions.14–16 The larger Bcl-xL is antiapoptotic, while the shorter Bcl-xS is proapoptotic.15 The aims of the current study were to determine if SAMe and MTA modulate the expression of Bcl-x variants, elucidate the mechanism, and explore their relevance to apoptosis.
Materials and Methods
SAMe, in the stable form of sulfate-p-toluensulfonate salt, was obtained from Knoll Farmaceutici (Milan, Italy). MTA, cycloleucine, trichostatin A, 3-deaza-adenosine, calyculin A, and ceramide-D-e-C6 ceramide were obtained from Sigma (St. Louis, MO). Okadaic acid and tautomycin were obtained from Calbiochem (La Jolla, CA). mAb104 (phospho-serine–arginine [SR] protein antibody) hybridoma cells were purchased from the American Type Culture Collection (Rockville, MD). All antibodies used in this study were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), except for anti-Bcl-xS antibodies, which were obtained from EMD Biosciences (San Diego, CA). 32P-dCTP (3,000 Ci/mmol) was purchased from New England Nuclear (DuPont, Boston, MA). All other reagents were of analytical grade and were obtained from commercial sources.
Cell Culture and SAMe/MTA Treatment.
Primary hepatocytes and HepG2 and 293 cells were obtained from the Cell Culture Core of the University of Southern California Liver Disease Research Center. Primary mouse hepatocytes were isolated from 2-month-old male C57B6 mice and cultured in DME/F12 medium supplemented with methionine (1 mmol/L), 10% fetal bovine serum, insulin (1 μg/mL), and hydrocortisone (50 nmol/L). Medium was changed 3 hours after plating to remove dead, unattached cells. HepG2 and 293 cells were grown according to instructions provided by the American Type Culture Collection. Prior to treatment with SAMe or MTA, medium was changed to 1% fetal bovine serum overnight. Cells were then treated with SAMe (0.125 to 1 mmol/L), MTA (0.125 to 1 mmol/L dissolved in dimethyl sulfoxide, final concentration 0.5%) or ceramide (5–80 μmol/L) for up to 48 hours. Controls for SAMe and MTA contained vehicle alone. Because of the instability of SAMe and MTA, medium was changed every 6 hours.
To assess the effect of trichostatin A (3 μmol/L, an inhibitor of histone deacetylase), cycloleucine (20 mmol/L, an inhibitor of MTA), 3-deaza-adenosine (10 μmol/L, an inhibitor of methylation), calyculin A (1.25–10 nmol/L, an inhibitor of serine–threonine protein phosphatase 1 [PP1] and 2A [PP2A]), tautomycin (100–800 nmol/L, a selective PP1 inhibitor at these concentrations), and okadaic acid (2.5 nmol/L, at this concentration a specific PP2A inhibitor) on SAMe- and MTA-mediated alteration of Bcl-x gene expression, HepG2 and 293 cells were pretreated with these agents for 2 hours, followed by cotreatment with SAMe or MTA for another 24 hours.
Northern Blot Analysis and Reverse-Transcriptase Polymerase Chain Reaction.
Total RNA was isolated and Northern hybridization analysis was performed as previously described.17, 18 The complementary DNA (cDNA) probe for Bcl-x corresponds to nucleotides 93–670 of the human Bcl-x sequence.15 The cDNA probes for PP1 catalytic and regulatory subunits correspond to nucleotides 24–395 of the human PP1 catalytic subunit of the gamma isoform and nucleotides 51–585 of the human PP1 regulatory subunit, respectively.19 To ensure equal loading of RNA samples and transfer in each of the lanes, membranes were rehybridized with a 32P-labeled β-actin cDNA probe and autoradiography and densitometry were used to quantitate relative RNA as previously described.18 Results of Northern blot analysis were normalized to β-actin.
To distinguish between Bcl-xL and Bcl-xS, reverse-transcriptase polymerase chain reaction (RT-PCR) was performed using a forward primer that corresponded to +305 to +327 relative to ATG translational start site of Bcl-x (5′-GAGGCAGGCGACGAGTTTGAAC-3′) and a reverse primer that corresponded to +744 to +766 relative to ATG translational start site of Bcl-x (5′-TGGGAGGGTAGAGTGGATGGTC-3′) as previously described.20
Transcription Start Sites of the Human Bcl-x Gene.
Transcription start sites were determined using the GeneRacer kit (Invitrogen, Carlsbad, CA). The first cDNA strand was synthesized with Moloney murine leukemia virus reverse transcriptase (Invitrogen) and oligo(dT)17 as the reverse complementary primer. To determine transcriptional start sites, primer ATG-reverse (5′-CTAAACTGACTCCAGCTGTATCCTTTCTGGGTTTCGA-3′), corresponding to nucleotides +80 to +41 from ATG, was used as the reverse complementary primer. GeneRacer 5′ primer (5′-CGACTGGAGCACGAGGACACTGA-3′) was used as the forward primer. The cDNA pool (2 μL), 1.25 units of Thermus aquaticus polymerase (Invitrogen), and amplification primers (20 pmol each) in 50 μL of PCR mixture (1X polymerase buffer, 2 mmol/L MgCl2, 200 mmol/L each dNTP) were denatured at 95°C for 3 minutes, followed by 35 cycles of amplification by using a step program (95°C for 30 seconds, 69°C for 1 minute 30 seconds, and a final extension at 72°C for 10 minutes). The initial PCR product was reamplified with the nested oligonucleotides: ATG-nested rev.2 (5′-AGGAGAGAAAGTCAACCACCAGCTCCCGGTTCGTCTGAGA-3′), corresponding to nucleotides +40 to +3 from ATG, was used as the reverse complementary primer while GeneRacer 5′ nested primer (5′-GGACATGACATGGACTGAAGGAGTA-3′) was used as the forward primer. PCR was performed for 30 cycles (95°C for 30 seconds, 72°C for 1 minute 30 seconds, and a final extension at 72°C for 10 minutes). PCR products were purified by electrophoresis in 2% agarose gel, and the different bands were extracted from agarose and purified with a clean-up kit (cDNA clean-up system, Promega, Madison, WI). Sequencing analysis was performed by the Sequencing Facility, Norris Cancer Center, Keck School of Medicine, University of Southern California.
Determination of mRNAs Generated From the Human Bcl-x Gene.
Race-ready cDNAs were prepared as described above. For PCR amplification of the first round, GeneRacer 5′ primer described above was used as forward primer. The oligonucleotide L/S reverse primer I (5′-TGGGAGGGTAGAGTGGATGGTC-3′) was used as the reverse primer. The race-ready cDNA pool (2 μL), 1.25 units of Thermus aquaticus polymerase and amplification primers (20 pmol each) in 50 μL of PCR mixture (1X polymerase buffer, 2 mmol/L MgCl2, 200 mmol/L each dNTP) were denatured at 94°C for 1 minute, followed by 35 cycles of amplification using a step program (94°C for 30 seconds, 69°C for 1 minute 30 seconds, and a final extension at 72°C for 10 minutes). The initial PCR product was reamplified with the nested oligonucleotides as forward primers: A (5′-GAAGCTCTTTCTCTCCCTTCA-3′), B1 (5′-CCTGTGCGCCTTCTGGG-3′), and B2 (5′-CTCGATCCGGGCGATG-3′), which correspond to nucleotides −134 to −114, −590 to −573, and −650 to −634 relative to ATG translational start site, respectively. L/S reverse primer II (5′-GCTGCATTGTTCCCATAGAGTTCC-3′) was used as reverse primer. PCR was performed for 30 cycles (94°C for 30 seconds, 58°C for 20 seconds, 72°C for 1 minute, and a final extension at 72°C for 10 minutes). PCR products were subjected to electrophoresis on a 1.5% agarose gel and photographed.
Western Blot Analysis.
Total cell lysates and nuclear protein were extracted from primary hepatocytes and HepG2 and 293 cells after various treatments and subjected to Western blot analysis as previously described.21 Membranes were probed with antiphosphorylated SR proteins (mAb104, 1:10 dilution of medium), antitotal SR proteins, anti-Bcl-xL/S (for 293 cells), anti-Bcl-xS or anti-Bcl-xL (for HepG2 cells and mouse hepatocytes), and anti-PP1c (catalytic) antibodies. To ensure equal loading, membranes were stripped and reprobed with antiactin antibodies. A horseradish peroxidase-conjugated secondary antibody was used. Blots were developed using enhanced chemoluminescence.
Measurement of Apoptosis.
To quantitate apoptosis, HepG2 and 293 cells were grown on 35-mm or 22-mm dishes. Following treatment, cells were fixed with 4% paraformaldehyde at 4°C for 1 hour and stained with 8 μg/mL HOECHST 33258 dye at 37°C for 30 minutes. Cells with bright, fragmented, condensed nuclei were identified as apoptotic cells using the Nikon Eclipse TE300 fluorescent microscope (Melville, NY). Three to five random fields (at a magnification of ×300) totaling approximately 500 cells were counted.
SAMe levels were measured in HepG2 cells treated with cycloleucine (20 mmol/L, 24 hours) or vehicle control as we previously described.9
Effect of SAMe and MTA on Bcl-x Expression in HepG2 Cells.
In preliminary studies using microarray analysis (Atlas Human 1.2 Array, Clontech), we found Bcl-x to be up-regulated in HepG2 cells in response to SAMe treatment (0.5 mmol/L) for 24 hours (data not shown). We first verified this effect using Northern blot analysis. SAMe treatment resulted in a near doubling (180%) of the Bcl-x mRNA level (Fig. 1A). Because the two major splice variants of Bcl-x differ by only 63 amino acids,15 RT-PCR was performed using specific primers. SAMe treatment resulted in a dose-dependent induction of only Bcl-xS (see Fig. 1A). Because the proapoptotic effect of SAMe can be mimicked by MTA,10 we also examined the effect of MTA on Bcl-x gene expression. MTA also exerted a dose-dependent induction of only Bcl-xS (Fig. 1B). Note that MTA is effective at much lower doses compared with SAMe. The induction of Bcl-xS by SAMe and MTA requires a minimum of 18 hours (Fig. 1C). Supplementary Figs. 1A and 1B confirm that both SAMe and MTA treatment resulted in an increase in Bcl-xS protein levels in a dose- and time-dependent manner. However, they have no significant effect on Bcl-xL protein levels in HepG2 cells (Supplementary Fig. 1C) or Bcl-xS protein levels in primary mouse hepatocytes (Supplementary Fig. 1D).
Effect of SAMe and MTA on Bcl-x Promoter Usage and Isoform Expression.
The Bcl-x gene in mice and humans has complex structures with multiple transcription start sites.16, 23 We next examined whether or not SAMe or MTA treatment may have altered the Bcl-x promoter usage in HepG2 cells. The structure of the human Bcl-x gene is shown in Fig. 2. There are multiple transcription start sites in the human Bcl-x gene. Promoters 1, 2, and 3 are for transcription start sites at −134 bp, −590 bp, and −650 bp relative to the translational start site, respectively. All three promoters are active in control HepG2 cells and yield only Bcl-xL mRNA. SAMe and MTA treatment did not affect promoter usage, but while promoter 1 yielded only Bcl-xL, the other two promoters yielded both Bcl-xL and Bcl-xS, with Bcl-xS as the predominant mRNA species. With promoter 3 in particular, SAMe and MTA treatment caused a reciprocal change in Bcl-xL and Bcl-xS, with the former reduced while the latter induced (see Fig. 2).
Effect of Trichostatin A, 3-Deaza-adenosine, Cycloleucine, Calyculin A, Tautomycin, and Okadaic Acid on SAMe- and MTA-Mediated Induction of Bcl-xS.
We next examined whether or not SAMe- and MTA-mediated induction of Bcl-xS can be modulated by inhibitors of histone deacetylation, methylation, or protein phosphatases. The chosen dosage of these agents was based on previous studies.24, 25 Trichostatin A, an inhibitor of histone deacetylation,25 and 3-deaza-adenosine, an inhibitor of methylation,24 had no effect on basal Bcl-x expression or Bcl-xS induction by SAMe or MTA (Supplementary Fig. 2). Cycloleucine is an inhibitor of methionine adenosyltransferase, which would prevent the conversion of MTA to SAMe through the methionine salvation pathway.24, 26 Cycloleucine treatment resulted in a 80% fall of cellular SAMe level (data not shown) but had no effect on basal Bcl-x expression or the induction of Bcl-xS caused by MTA (see Supplementary Fig. 2).
Serine–threonine protein phosphastase inhibitors were examined next, because PP1 has been shown to regulate alternative splicing.20, 27 Calyculin A, an inhibitor of both PP1 and PP2A, inhibited the SAMe- and MTA-mediated induction of Bcl-xS in a dose-dependent manner (Figs. 3A and 3B). However, okadaic acid, an inhibitor of PP2A as used in these experiments,28 had no effect (Fig. 3C). Tautomycin, a selective inhibitor of PP1 at nmol/L to μmol/L concentrations,29 also blocked the SAMe- and MTA-mediated Bcl-xS induction dose-dependently (Fig. 3D). These results are consistent with the notion that the induction of Bcl-xS by SAMe and MTA requires PP1 activity.
Effect of SAMe and MTA on SR Protein Phosphorylation.
SR proteins belong to a family of arginine–serine-rich domain containing proteins that are required for constitutive and alternative pre-mRNA processing.20, 30–32 Although phosphorylation of SR proteins is required for spliceosome assembly, dephosphorylation is critical to the splicing reaction.31–33 Because SR proteins are substrates of PP1, and PP1 activity is required for the induction of Bcl-xS by SAMe and MTA, we next examined the effect of these agents on SR protein phosphorylation. Figure 4 shows that SAMe and MTA treatment led to dephosphorylation of SR proteins, which was blocked by calyculin A but not okadaic acid. The degree of dephosphorylation varied, with SRp75, SRp55, and SRp30 exhibiting 35%–50% less phosphorylation, while SRp40 and SRp20 exhibited 65%–80% less phosphorylation after either SAMe or MTA treatment (Fig. 4A). Note that these treatments did not alter the total amount of SR proteins (see Fig. 4A). In nuclear protein extracts, a similar magnitude of loss of SR protein phosphorylation occurred after SAMe or MTA treatment (Fig. 4B). SR protein phosphorylation fell by 43%–45% 12 hours after SAMe or MTA treatment and 70%–77% after 24 hours (Fig. 4C).
Effect of SAMe and MTA on PP1.
Because the induction of Bcl-xS by SAMe and MTA can be blocked by inhibiting PP1 (but not PP2A), we next examined the effect of SAMe and MTA on PP1 itself. Figure 5A shows that the steady state protein levels of PP1 catalytic subunit increased after both SAMe (170% of control) and MTA (280% of control) treatment in HepG2 cells. The mechanism for this increase is pretranslational, because a comparable increase in the steady state mRNA levels of PP1 catalytic subunit also occurred after SAMe (203% of control) and MTA treatment (220% of control) (Fig. 5B). On the other hand, the regulatory subunit of PP1 is unaffected. The induction in PP1c protein levels occurred after 12 hours of SAMe or MTA treatment in HepG2 cells (Fig. 5C) but not after 24 hours of treatment in primary hepatocytes (Fig. 5D). The increase in PP1 catalytic subunit, along with dephosphorylation of SR proteins, which are specific PP1 substrates, support the notion that SAMe and MTA treatment up-regulate PP1 expression and PP1 activity.
The Role of Bcl-xS and PP1 on SAMe- and MTA-Induced Apoptosis.
We next examined the role of Bcl-xS and PP1 on SAMe- and MTA-induced apoptosis. Figure 6A shows that SAMe and MTA induced apoptosis in a dose-dependent manner. Calyculin A also blocked this process in a dose-dependent manner (Fig. 6B). Ceramide was shown to induce Bcl-xS expression via increased PP1 activity in A549 lung adenocarcinoma cells.20 Ceramide also induced the expression of Bcl-xS in HepG2 cells and—importantly—apoptosis in a dose-dependent manner (Fig. 6C). This is consistent with the notion that increases in PP1 activity and Bcl-xS expression result in apoptosis.
Effect of SAMe and MTA in 293 Cells.
To examine whether or not the effects of SAMe and MTA are limited to liver cancer cell lines, 293 cells were treated with varying concentrations of SAMe and MTA. Supplementary Fig. 3A shows that similar to HepG2 cells, both SAMe and MTA also induced the expression of Bcl-xS in a dose- and time-dependent manner. Supplementary Fig. 3B shows that the steady state mRNA levels of PP1 catalytic subunit are also increased following these treatments and Supplementary Figs. 3C and 3D show that both SAMe and MTA induced apoptosis in this cell line in a dose-dependent manner, which was blocked by calyculin A. Thus the effect of SAMe and MTA on Bcl-xS and apoptosis is more generalized than simply liver cancer cells.
Since its original discovery over 50 years ago,34 the biological actions of SAMe have been largely attributed to its role as a methyl donor and a precursor for polyamines and glutathione.2 Recent evidence suggests that in the liver, SAMe has critical functions in modulating growth and apoptotic responses—some of which are independent of methylation, because they can be mimicked by MTA.8–10 MTA is a product of SAMe metabolism in the polyamine pathway.1 Exogenous SAMe can also undergo nonenzymatic hydrolysis in vivo into MTA and homoserine.35 In contrast to SAMe, MTA does not contribute to glutathione synthesis, is not a methyl donor, and inhibits methyltransferases.13 These observations suggest that many of the biological effects of SAMe could be mediated in part through its conversion to MTA. The present study extends these observations and provides evidence for a highly novel SAMe and MTA biological action, namely modulation of alternative splicing through changes in SR protein phosphorylation.
A key finding of the present study is that SAMe and MTA selectively up-regulated Bcl-xS. Bcl-x is a Bcl-2–related gene that encodes several alternatively spliced mRNAs.23 Bcl-xL and Bcl-xS are the two major transcripts in humans that exert antagonistic actions. Progestins were shown to prevent apoptosis in a rat endometrial cell line by increasing the ratio of Bcl-xL to Bcl-xS.36 On the other hand, amphetamines were shown to induce apoptosis in neocortical neurons by decreasing the ratio of Bcl-xL to Bcl-xS.37 Regarding a critical ratio for apoptosis, Minn et al. demonstrated that only one molecule of Bcl-xS per four molecules of Bcl-xL was necessary to overcome the Bcl-xL survival mechanism.38 Although many authors have reported alteration in the Bcl-xL to Bcl-xS ratio as it relates to apoptosis, few have examined the molecular mechanisms. Recently, Chalfant et al. demonstrated the ability of ceramide to differentially induce Bcl-xS through activation of alternative splicing.20 Another major determinant of Bcl-x isoform expression is promoter usage, which has been demonstrated in the mouse gene.23 Our study explored all of the potential mechanisms that are known to influence the expression of Bcl-xS.
Both SAMe and MTA selectively up-regulated Bcl-xS in a time- and dose-dependent fashion with the effective dose for MTA lower than that for SAMe, suggesting that the effect of SAMe may be mediated in part by MTA. Although MTA is a metabolite of SAMe, it can also be converted back to SAMe via methionine.26 However, blocking this conversion by cycloleucine did not prevent the induction of Bcl-xS by MTA, supporting the notion that MTA exerts its action directly. Inhibitors of histone deacetylase and DNA methylation also had no effect on basal Bcl-x expression or the induction of Bcl-xS by SAMe and MTA, thereby excluding their involvement.
Similar to the mouse gene, we found that there are multiple transcription start sites in the human Bcl-x gene. Three promoters are identified in control HepG2 cells and yield only Bcl-xL mRNA. SAMe and MTA treatment did not affect promoter usage; however, while promoter 1 yielded only Bcl-xL, the other two promoters yielded both Bcl-xL and Bcl-xS, with Bcl-xS as the predominant mRNA species. Thus alternative splicing appears to selectively involve promoters 2 and 3.
We next investigated the role of SR proteins and protein phosphatases in alternative splicing of Bcl-x. Pre-mRNA splicing is a critical step in gene expression for metazoans.32 Although small nuclear ribonuclear particles (snRNPs) are important determinants for splice site recognition and catalysis, non-snRNP protein splicing factors are essential for establishing and stabilizing interactions necessary for splicing.32, 33 SR proteins are a family of highly conserved non-snRNP factors vital to constitutive and regulated splicing.31–33 The function of SR proteins is regulated by phosphorylation. Although SR protein phosphorylation is required for spliceosome formation, dephosphorylation is required for progression of the splicing reaction.33 SR proteins are substrates of PP1, and PP1 has been shown to regulate alternative splicing.20 We found that calyculin A, an inhibitor of both PP1 and PP2A, and tautomycin, a selective inhibitor of PP1, were able to block the induction of Bcl-xS by SAMe and MTA in a dose-dependent manner, but okadaic acid, an inhibitor of PP2A, was not. This result is consistent with a role of PP1 in mediating the effect of SAMe and MTA on induction of Bcl-xS. We also found that both SAMe and MTA treatment of HepG2 cells led to dephosphorylation of SR proteins, which was prevented by calyculin A but not okadaic acid. These results suggest that dephosphorylation of SR proteins occurred as a result of increased PP1 activity.
We next examined whether SAMe and MTA treatment affected the amount of PP1. We found that the steady state protein and mRNA levels of PP1 catalytic subunit increased after SAMe and MTA treatment. Consistent with our findings, it has been shown by Wang et al. that treatment of HL-60 (human promyelotic leukemia) cells with a proapoptotic stimulus (cytosine arabinoside) resulted in a rapid increase in PP1 protein (twofold) and activity (30%).39 Increases in PP1 preceded caspase activation. Inhibition of PP1 activity in these cells prevented caspase-mediated apoptosis. Conversely, electroporation of the cells with constitutively active PP1 protein resulted in apoptosis.
Increased PP1 activity can result in multiple consequences; SR protein dephosphorylation is one of them. This in turn can activate alternate splicing of multiple genes; Bcl-x is one of them. When the dephosphorylation is blocked by calyculin A, apoptosis is prevented. Our findings allow us to conclude that the trigger for the apoptotic cascade in HepG2 cells by SAMe or MTA is increased PP1 protein and activity, because its inhibition prevents all the downstream steps. PP1 in turn dephosphorylates the SR proteins, resulting in the formation of the alternate variant Bcl-xS and apoptosis. This proposed sequence of events is supported by the fact that induction of PP1 and SR protein dephosphorylation preceded Bcl-xS induction. Bcl-xS induction is one of many downstream targets of PP1, more work will be necessary to identify other targets that can also contribute to apoptosis.
Although SAMe and MTA induced PP1c and Bcl-xS expression in HepG2 cells, they have no effect in primary mouse hepatocytes. This suggests that a lack of PP1 and Bcl-xS induction may contribute to the differential response of liver cancer cell lines and primary hepatocytes to SAMe- and MTA-induced apoptosis.10
Finally, SAMe and MTA also induce apoptosis and Bcl-xS expression in the 293 cell line, which is originally derived from human embryonal kidney (ATCC). In this cell line, SAMe and MTA also induced the expression of PP1 catalytic subunit and apoptosis in a dose-dependent manner, and—most importantly—apoptosis was also blocked by calyculin A. These findings suggest that SAMe and MTA's effect on phosphorylation state; apoptosis and alternative splicing may be more generalized.
In conclusion, we have uncovered a highly novel biological action of SAMe and MTA. These agents selectively induce Bcl-xS by activating SR protein–mediated alternative splicing. This occurs through the activation of PP1, which leads to dephosphorylation of SR proteins. The selective up-regulation of Bcl-xS likely contributes to the proapoptotic effect of SAMe and MTA in HepG2 and 293 cells. Modulation of the state of protein phosphorylation is an action that has not been previously shown for either SAMe or MTA. This ability further widens the impact that these agents have on cellular function.
- 3Polyamine metabolism and its importance in neoplastic growth and a target for chemotherapy. Cancer Res 1988; 15: 759–774..
- 28Assay and purification of protein serine/threonine phosphatases. In: HardieDG, ed. Protein Phosphorylation: A Practical Approach. London: Oxford University Press, 1993: 97–230..
Supplementary material for this article can be found on the H EPATOLOGY website ( http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html ).
|suppmat_fig1.tif||1773K||Effect of SAMe and MTA on the steady state protein amount of Bcl-xS and Bcl-xL. In A) and B), HepG2 cells were treated with either SAMe or MTA (0 to 2.5mM for 24 hrs, or 1mM for 0 to 48 hrs). In C) and D), HepG2 cells (C) or mouse hepatocytes (D) were treated with either SAMe or MTA (1mM for 24 hrs). Total cell lysates (50µg) were subjected to Western blot analysis using anti- Bcl-xS or anti-Bcl-xL antibodies as described in Methods. The same membranes were stripped and probed with antibodies against actin to ensure equal protein loading. The right panels show densitometric changes expressed as % of control. *p<0.05 vs. respective controls.|
|suppmat_fig2.tif||466K||Effect of trichostatin A (Tricho), cycloleucine (CL), and 3-deaza-adenosine (3-dea) on SAMe (A) and MTA (B)-mediated induction of Bcl-xS. Total RNAs from HepG2 cells pretreated with trichostatin A (3µM, an inhibitor of histone deacetylase), cycloleucine (20mM, an inhibitor of methionine adenosyltransferase) or 3-deaza-adenosine (10µM, an inhibitor of methylation) for 2 hrs followed by co-treatment with SAMe (1mM) or MTA (1mM) for another 24 hrs were subjected to RT-PCR to evaluate the effect on Bcl-x gene expression as described in Methods. Representative RT-PCRs are shown.|
|suppmat_fig3.tif||1476K||Effect of SAMe and MTA on Bcl-xS expression (part A), PP1 catalytic subunit expression (part B) and apoptosis (parts C and D) in 293 cells. In part A, 293 cells were treated with SAMe or MTA (0 to 5mM for 24 hrs, or 1mM for 0 to 48 hrs) and Western blot analysis was performed using anti-Bcl-xL/S antibodies. In part B, PP1 catalytic subunit Northern blot analysis was performed after 24 hr treatment with either MTA or SAMe (1mM). The same membrane was rehybridized with the cDNA probe for b-actin to ensure equal loading. In part C, 293 cells were treated with SAMe or MTA (0 to 5mM) for 24 hrs and apoptosis was quantitated as described in Methods. In part D, 293 cells were pretreated with calyculin A (0 to10nM) for 2 hrs followed by SAMe or MTA (1mM) for 24 hrs. Apoptosis was quantitated as described in Methods. Note tha calyculin A prevented apoptosis in a dose-dependent manner as in HepG2 cells.|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.