Advertisement

Sodium Butyrate Induces Transcription from the Gαi2Gene Promoter through Multiple Sp1 Sites in the Promoter and by Activating the MEK-ERK Signal Transduction Pathway*

  • Jianqi Yang
    Affiliations
    Department of Biochemistry, Meharry Medical College, Nashville, Tennessee 37208-3599 and the
    Search for articles by this author
  • Yumiko Kawai
    Affiliations
    Department of Biochemistry, Meharry Medical College, Nashville, Tennessee 37208-3599 and the
    Search for articles by this author
  • Richard W. Hanson
    Affiliations
    Department of Biochemistry, Meharry Medical College, Nashville, Tennessee 37208-3599 and the
    Search for articles by this author
  • Ifeanyi J. Arinze
    Correspondence
    To whom correspondence should be addressed: Dept. of Biochemistry, Meharry Medical College, 1005 David B. Todd Jr. Blvd., Nashville, TN 37208-3599. Tel.: 615-327-6586;
    Affiliations
    Department of Biochemistry, Meharry Medical College, Nashville, Tennessee 37208-3599 and the
    Search for articles by this author
  • Author Footnotes
    * This work was supported by National Science Foundation Grant MCB-9905070 (to I. J. A.) and by National Institutes of Health Grant DK 25541 (to R. W. H.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:July 13, 2001DOI:https://doi.org/10.1074/jbc.M102821200
      Sodium butyrate, an erythroid differentiation inducer and a histone deacetylase inhibitor, increases Gαi2 levels in differentiating K562 cells. Here we show that sodium butyrate induces Gαi2 gene transcription via sequences at −50/−36 and −92/−85 in the Gαi2 gene promoter. Both sequences contain core sequence motif for Sp1 binding; electrophoretic mobility shift as well as supershift assays confirmed binding to Sp1. Transcription from the Gαi2 gene promoter was also activated by two other histone deacetylase inhibitors, trichostatin A andHelminthsporium carbonium toxin (HC toxin), which also induce erythroblastic differentiation in K562 cells. However, hydroxyurea, a potent erythroid differentiation inducer in these cells, did not activate transcription from this gene promoter, indicating that promoter activation is inducer-specific. Mutations within the Sp1 sites at −50/−36 and −92/−85 in the Gαi2 gene promoter substantially decreased transcriptional activation by sodium butyrate, trichostatin A, or HC toxin. Transfection with constitutively activated ERKs indicated that this promoter can be activated through the MEK-ERK signal transduction pathway. Inhibition of the MEK-ERK pathway with U0126 or reduction in the expression of endogenous ERK with an antisense oligonucleotide to ERK significantly inhibited sodium butyrate- and HC toxin-induced transcription but had no effect on trichostatin A-induced transcription. Inhibition of the JNK and p38 MAPKs, using selective inhibitors, had no effect on sodium butyrate-induced transcription. In cells in which sodium butyrate induction of promoter activation had been inhibited by various concentrations of U0126, constitutively activated ERK2 reversed this inhibition. These results show that the MEK-ERK signal transduction pathway is important in butyrate signaling, which eventually converges in the cell nucleus.
      G-proteins
      guanine nucleotide-binding regulatory proteins
      C/EBPα
      CCAAT box enhancer-binding protein
      DTT
      dithiothreitol
      EMSA
      electrophoretic mobility shift assay
      ERKs
      extracellular-regulated kinases
      HC toxin
      Helminthsporium carbonium toxin
      JNK
      c-Jun N-terminal kinase
      MEK (MAPKK)
      dual specificity mitogen-activated protein kinase kinase
      PBS
      phosphate-buffered saline, PMSF, phenylmethylsulfonyl fluoride
      SB 203580
      [4-(4′-fluorophenyl)-2-(4′-methylsulfinylphenyl)-5-(4′-pyridyl) imidazole]
      Sp1
      promoter-specific factor binding protein 1 (also called stimulatory protein 1)
      TBP
      TATA box-binding protein
      U0126
      1,4-diamino-2,3- dicyano-1,4-bis[2-aminophenyl-thio]butadiene
      MAPK
      mitogen-activated protein kinase
      There is compelling evidence that the α-subunits of heterotrimeric G-proteins1can influence cell differentiation in different ways, depending on the cell type. For example, Gαs has been shown to suppress dexamethasone-induced differentiation of 3T3-L1 cells, leading to suppression of adipogenesis in these cells (
      • Wang H.
      • Watkins D.C.
      • Malbon C.C.
      ). Gα12 and Gα13 have been implicated in the retinoic acid-mediated differentiation of P19 mouse embryonal carcinoma cells (
      • Jho E.-H.
      • Malbon C.C.
      ,
      • Jho E.-H.
      • Davis R.J.
      • Malbon C.C.
      ). During Me2SO-induced neutrophilic differentiation of human myeloid HL-60 cells, the expression of Gα16 is decreased by 90%, whereas the expression of Gαi2 is increased by 160% (
      • Amatruda III, T.T.
      • Steele D.A.
      • Slepak V.Z.
      • Simon M.I.
      ,
      • Wilke T.M.
      • Scherle P.A.
      • Strathmann M.P.
      • Slepak V.Z.
      • Simon M.I.
      ); this suggests an association between cell differentiation and these G-protein α-subunits. In F9 teratocarcinoma cells, the levels of Gαi2 decrease as the cells are induced to differentiate (
      • Watkins D.C.
      • Johnson G.L.
      • Malbon C.C.
      ). Sodium butyrate-induced erythroblastic differentiation of K562 cells requires the presence of Gαi2, since pertussis toxin or an antisense oligonucleotide to a portion of the Gαi2 gene blocks the sodium butyrate-induced effect (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). The expression of genes for some proteins has been reported to be influenced by butyrate (
      • Leder A.
      • Leder P.
      ,
      • Birren B.W.
      • Herschman H.R.
      ,
      • Lazar M.A.
      ,
      • Rius C.
      • Cabanas C.
      • Aller P.
      ,
      • Nakano K.
      • Mizuno T.
      • Sowa Y.
      • Orita T.
      • Yoshino T.
      • Okuyama Y.
      • Fugita T.
      • Ohtani-Fujita N.
      • Matsukawa Y.
      • Tokino T.
      • Yamagishi H.
      • Oka T.
      • Nomura H.
      • Sakai T.
      ,
      • Lu Y.
      • Lotan R.
      ,
      • Benvenuto G.
      • Carpentieri M.L.
      • Salvatore P.
      • Cindolo L.
      • Bruni C.B.
      • Chiariotti L.
      ,
      • Tsuji Y.
      • Moran E.
      • Torti S.V.
      • Torti F.M.
      ), but none of these proteins is a G-protein.
      Although the molecular details of the involvement of G-proteins in cell differentiation have yet to be elucidated, the associated change in G-protein concentration provides an excellent model for exploring the molecular regulation of the expression of the G-proteins themselves. Furthermore, the mechanism by which any cell differentiation inducing agent alters G-protein levels is not known. Sodium butyrate-induced differentiation of K562 cells is accompanied by a 3–4-fold increase in the mRNA levels for Gαi2 (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ), suggesting transcriptional activation of this gene during differentiation. The gene for Gαi2, which contains no TATA box, was isolated several years ago (
      • Weinstein L.S.
      • Spiegel A.M.
      • Carter A.D.
      ,
      • Weinstein L.S.
      • Kats I.
      • Spiegel A.M.
      • Carter A.D.
      ). There are reports that elevation of cAMP has a stimulatory effect on the Gαi2 gene promoter (
      • Kinane T.B.
      • Shang C.
      • Finder J.D.
      • Ercolani L.
      ,
      • Eschenhagen T.
      • Friedrichsen M.
      • Gsell S.
      • Hollmann A.
      • Mittmann C.
      • Schmitz W.
      • Scholz H.
      • Weil J.
      • Weinstein L.S.
      ). There are also reports from Ercolani and co-workers (
      • Holtzman E.J.
      • Soper B.W.
      • Stow J.L.
      • Ausiello D.A.
      • Ercolani L.
      ,
      • Holtzman E.J.
      • Kinane T.B.
      • West K.
      • Soper B.W.
      • Karga H.
      • Ausiello D.A.
      • Ercolani L.
      ,
      • Kinane T.B.
      • Finder J.D.
      • Kawashima A.
      • Brown D.
      • Abbate M.
      • Shang C.
      • Fredericks W.J.
      • Rauscher III, F.J.
      • Sukhatme V.P.
      • Ercolani L.
      ,
      • Kinane T.B.
      • Finder J.D.
      • Kawashima A.
      • Brown D.
      • Abbate M.
      • Fredericks W.J.
      • Sukhatme V.P.
      • Rauscher III, F.J.
      • Ercolani L.
      ) that the Gαi2 gene promoter is regulated in LLC-PK1 renal cells. Apart from these reports, there has been no other study addressing the molecular regulation of the transcription of this gene (
      • Morris A.J.
      • Malbon C.C.
      ).
      Here we have used K562 cells to explore the DNA sequence elements and/or transcription factor(s) involved in the sodium butyrate-induced expression of the gene for Gαi2. We found that sodium butyrate can strongly activate transcription from the Gαi2 promoter and that Sp1 sites at −50/−36 and −92/−85, relative to the putative transcription start site, are involved in this activation. GC-rich binding sites for Sp1 have previously been found in numerous promoters that drive the expression of genes involved in the regulation of a variety of cell functions, including differentiation, proliferation, apoptosis, metabolism, and secretion (
      • Liu C.
      • Calogero A.
      • Ragona G.
      • Adamson E.
      • Mercola D.
      ). Sp1 sites are ubiquitous in mammalian genes. Mice that are homozygous for deletion in the gene for Sp1 exhibit several embryonic malformations and often die during development (
      • Marin M.
      • Karis A.
      • Visser P.
      • Grosveld F.
      • Phillipsen S.
      ). Our finding that Sp1 sites mediate the sodium butyrate-induced transcription from the Gαi2 promoter adds this promoter to a small but growing list of butyrate-regulated genes, including the galectin-1 gene (
      • Lu Y.
      • Lotan R.
      ), the WAF1/Cip1 gene (
      • Nakano K.
      • Mizuno T.
      • Sowa Y.
      • Orita T.
      • Yoshino T.
      • Okuyama Y.
      • Fugita T.
      • Ohtani-Fujita N.
      • Matsukawa Y.
      • Tokino T.
      • Yamagishi H.
      • Oka T.
      • Nomura H.
      • Sakai T.
      ), and the mouse ferritin H gene (
      • Tsuji Y.
      • Moran E.
      • Torti S.V.
      • Torti F.M.
      ), for which Sp1 sites have been shown to be targets of the sodium butyrate effect.
      Inhibition of the MAPKs by specific inhibitors, depletion of the expression of endogenous ERK with an antisense oligonucleotide to ERK, and transfection with plasmids containing genes for constitutively activated ERKs demonstrated that the sodium butyrate effect requires the MEK-ERK signal transduction pathway and does not involve JNK or p38 MAPKs.

      EXPERIMENTAL PROCEDURES

      Chemicals

      Sodium butyrate, protease inhibitor mixture, hydroxyurea, and trichostatin A were purchased from the Sigma Chemical Company (St. Louis, MO). Restriction enzymes, T4 DNA ligase, DNA polymerase I (large fragment), and Klenow fragment were purchased from New England Biolabs Inc. (Beverly, MA). Plasmid pGL3-basic, Sp1 oligonucleotide, anti-ACTIVE MAPK antibody, anti-ACTIVE p38 antibody, cell culture lysis reagent, and the MAPK inhibitors, U0126 and SB 203580, were purchased from Promega (Madison, WI). PD 169316 was purchased from Calbiochem-Novabiochem Corp. (La Jolla, CA). Antisense oligonucleotide (5′-GCCGCCGCCGCCGCCAT-3′) to ERK and HC toxin were purchased from Biomol (Plymouth Meeting, PA). Anti-Sp1 antibody was a product of Santa Cruz Biotechnology (Santa Cruz, CA). A plasmid containing the full-length Gαi2 promoter sequence (−1214/+115), linked to chloramphenicol acetyltransferase gene (
      • Weinstein L.S.
      • Spiegel A.M.
      • Carter A.D.
      ,
      • Weinstein L.S.
      • Kats I.
      • Spiegel A.M.
      • Carter A.D.
      ) was a gift from Dr. Lee Weinstein (National Institutes of Health). Plasmids harboring genes for constitutively activated ERKs (pCHA-ERK1 and pcDNA3-ERK2) as well as empty vectors (pCHA and pcDNA3) used to clone these genes were gifts from Dr. Michael J. Weber (University of Virginia, Charlottesville). Plasmid Mini and Qiafilter Midi Kits were products of Qiagen Inc. (Valencia, CA). Poly(dI-dC)·(dI-dC) was purchased from Amersham Pharmacia Biotech(Piscataway, NJ). [α-32P]dCTP (3,000 Ci/mmol) was purchased from PerkinElmer Life Sciences (Boston, MA). QuickChange Mutagenesis kit was purchased from Stratagene (La Jolla, CA). Slide-A-Lyzer Dialysis Cassette was purchased from Pierce. FuGENE-6 transfection reagent was purchased from Roche Molecular Biochemicals. Oligonucleotide primers and oligonucleotides were purchased fromPromega (Madison, WI), and from Integrated DNA Technologies, Inc. (Coralville, IA). The sources of all other chemicals and reagents have been described previously (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ).

      Plasmid Constructs

      A sequence (−1214/+115) containing Gαi2 full-length promoter was removed from a plasmid consisting of Gαi2 promoter linked to the chloramphenicol acetyltransferase structural gene (obtained as a gift from Dr. Lee Weinstein), by using appropriate restriction enzymes (KpnI and SmaI). The resulting fragment was sub-cloned into pGL3-basic containing the luciferase gene to generate pGαi2(−1214/+115)-luc. This plasmid was then used to generate the five truncations of the Gαi2 promoter shown in Fig. 1. The −1214/−784, −236/+115 and −1214/−784, −717/+115 truncations were made by digesting pGαi2(−1214/+115)-luc with SmaI and religating. The −667/+115 modification was made by deleting part (−1214/−668) of the full-length promoter with restriction enzymeXmnI; −184/+115 was generated by deleting thePstI fragment from the full-length promoter; and −79/+115 was made by sub-cloning the NcoI/Tsp45I fragment (251 base pairs) into pGL3-basic, usingNcoI/SmaI. After amplification in bacteria grown in LB medium, all plasmids were isolated with Plasmid Mini or Qiafilter Midi Kits (Qiagen, Inc.), checked for purity by the ratio of absorbance at 260/280 nm, and after separation of the DNA by electrophoresis on 0.8% agarose gels, followed by visualization under UV light. The truncations in the Gαi2 promoter were confirmed by restriction enzyme digestion.
      Figure thumbnail gr1
      Figure 1Sodium butyrate activates transcription from the Gαi2 gene promoter. K562 cells were grown in 24-well plates and transfected as described under “Experimental Procedures.” Each plasmid was tested in replicate cultures on at least six different occasions. Values shown are means ± S.E. for six experiments. The relative luciferase (LUC) activities (relative luciferase activity/µg protein) of the cell extracts are expressed as fold stimulation, relative to cells that were not treated with sodium butyrate. The plasmid designated as pGαi2(−1214/+115)-luc contained the full-length promoter for the gene for Gαi2. The indicated truncations were derived from pGαi2(−1214/+115)-luc as described under “Experimental Procedures.” *, statistically different (p < 0.002) compared with the full-length promoter.

      Site-directed Mutagenesis

      Specific nucleotides in the full-length Gαi2 gene promoter were mutated or deleted by using the QuickChange Mutagenesis kit purchased from Stratagene (La Jolla, CA). Briefly, a pair of primers (GGAGCGGAGTGGGTCTTTCGGGGCCGAGCC) was used to mutate the putative Sp1-binding site (+68/+75, GGGCGGGG), designated as site 1, to generate mutant pM1 (see M series in Fig. 2). To generate mutant pM2, a different pair of primers (CCCCACCCCCGAACCGCCCCGCCG) was used to mutate the putative Sp1-binding site (−50/−36, CCCCCGGCCCGCCCC), designated as site 2; this site contains an overlapping pair of consensus Sp1 sequence motifs. Another pair of primers (CCTGCAAGCACGAACCGGCCCAGTCACAGG) was used to mutate the putative Sp1-binding site (−92/−85, CCCGCCCC), designated as site 3. The underlined nucleotides were introduced into the mutant constructs, using the QuickChange Mutagenesis kit from Stratagene (La Jolla, CA). The double mutants designated as pM1,2 and pM1,3 were made by starting the mutation protocol with pM1; the double mutant designated as pM2,3 was made by starting the mutation protocol with pM2. The mutant designated pM1,2,3 was made from pM1,2. For deletions, the sequence CCCCCGGCCCGCCCCGC (−50/−34), which contains the putative Sp1-binding site 2 (−50/−36), was deleted from the construct pGαi2(−1214/+115)-luc, to generate pD2. The sequence GCCCCGCCTGCAAGCCCGCCCCG (−106/−84), which contains the putative Sp1-binding site 3 (−92/−85), was deleted from pGαi2(−1214/+115)-luc to generate pD3. Both sequences were absent in the double deletion mutant pD2,3; this mutant was derived from pD2.
      Figure thumbnail gr2
      Figure 2Sodium butyrate-activated transcription from a series of mutant Gαi2promoters. A, the indicated mutations (M series) were derived from pGαi2(−1214/+115)-luc, as described under “Experimental Procedures.” The underlined nucleotidesrepresent mutations. The results are expressed as in Fig. and are means ± S.E. for six samples for each plasmid. *, statistically different (p < 0.02) compared with the full-length promoter. **, statistically different (p < 0.002) compared with the full-length promoter. B, plasmids containing single or double deletions in the Gαi2 gene promoter were made as described under “Experimental Procedures.” To generate the single deletion mutants, the putative Sp1 site at −50/−36 or −92/−85 was removed from the full-length promoter. In the double mutant, both of these sites were deleted. The results (means ± S.E. of eight experiments) are expressed as fold stimulation, relative to cells that were not treated with sodium butyrate. *, statistically different (p < 0.003) compared with the full-length promoter. **, statistically different (p < 0.001) compared with the full-length promoter.LUC, luciferase.
      The plasmids resulting from the seven different substitution mutations and three deletions were isolated, and the substitutions/deletions were confirmed by restriction digestion and by DNA sequencing. DNA sequencing was performed by the Case Western Reserve University Molecular Biology Core Laboratory.

      Cell Line and DNA Transfection Studies

      K562 cells were obtained from the American Type Culture Collection (Manassas, VA) and maintained in culture as described previously (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). Briefly, the cells (1 × 105 cells in 1 ml of medium/well) were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum and antibiotics (50 units of penicillin and 50 µg of streptomycin per ml) at 37 °C in 95% air, 5% CO2 atmosphere in 24-well plates for 24 h before transfection. The cells were then transfected with plasmid DNA (0.5 µg containing the Gαi2 promoter construct or mutant) and 1.5 µl of FuGENE-6 transfection reagent (Roche Molecular Biochemicals) for 1 h, followed by addition of sodium butyrate or other cell differentiation inducers. Co-transfections with antisense oligonucleotide to ERK or with plasmids containing genes for constitutively activated ERKs were similarly carried out by appropriately adjusting the amount of DNA/oligonucleotide and the FuGENE-6 transfection reagent. When used, inhibitors (i.e.U0126, PD 169316, or SB 203580) were added 30 min prior to the addition of cell differentiation inducers. After 24 h, the cells were harvested by centrifuging (12,000 × g, 45 s) in 1.5-ml microcentrifuge tubes and washed once with 1× PBS, pH 7.4. The cell pellets were lysed with 150 µl of 1× Cell Culture Lysis Reagent (Promega, Madison, WI). After centrifugation for 2 min to remove cell debris, the luciferase activity and protein content of the cell extracts were measured.
      For luciferase activity, 10 µl of the extracts were used to measure the integrated light units over 10 s, using the luciferase assay system (Promega, Madison, WI) and a luminometer (Tropix, Inc., Bedford, MA), as recommended by the manufacturers. The protein content of the extracts was determined by the Bradford protein assay method (Bio-Rad), using bovine serum albumin as a standard.

      Preparation of Nuclear Extracts

      The general procedure outlined by Dignam et al. (
      • Dignam J.D.
      • Lebovitz R.M.
      • Roeder R.G.
      ) was followed, with several modifications. Essentially, 1 × 107 cells were transferred to a 15-ml conical bottom tube and centrifuged (500 ×g) in a clinical centrifuge for 1.5 min. The resulting pellet was washed by resuspension in 5 ml of ice-cold 1× PBS and recentrifuged; the recovered pellet was resuspended in 1 ml of ice-cold 1× PBS, transferred to a 1.5-ml microcentrifuge tube, and centrifuged at 12,000 × g for 15 s to pellet the cells. The pelleted cells were resuspended in 100 µl of buffer A, containing 10 mm Hepes-KOH (pH 7.9), 1.5 mmMgCl2, 10 mm KCl, 0.5 mm DTT, 0.5 mm PMSF, and freshly added 10 µl of protease inhibitor mixture (Sigma), and gently mixed with a pipette. After 15 min of incubation on ice, the cells were lysed by adding a 2% solution of Nonidet P-40 to achieve a final detergent concentration of 0.05%, followed by pipetting up and down 5 times to mix the solution. The solution was centrifuged at 12,000 × g for 20 s to obtain nuclei (pellet). This pellet was then resuspended in 100 µl of ice-cold buffer B, containing 20 mm Hepes-KOH (pH 7.9), 1.5 mm MgCl2, 0.2 mm KCl, 0.2 mm EDTA, 0.5 mm DTT, 0.2 mm PMSF, 25% glycerol, and 5 µl of protease inhibitor mixture as in buffer A, and placed in ice for 5 min. DTT, PMSF, and protease inhibitors were added just before use. To lyse the nuclei, buffer C (0.9 mKCl) was then added dropwise (about 50 µl) to achieve a final concentration of 0.3 m KCl. The mixture was placed on ice for 30 min with occasional gentle shaking and then centrifuged at 12,000 × g for 15 min at 4 °C to obtain the nuclear extract (supernatant). This extract was then dialyzed against 100 volumes of buffer D (dialysis buffer) for 2 h at 4 °C, using Slide-A-Lyzer Dialysis Cassette purchased from Pierce. Buffer D (modified from Khana-Gupta et al. (
      • Khana-Gupta A.
      • Zibello T.
      • Simkevich C.
      • Rosmarin A.G.
      • Berliner N.
      )) contained 20 mm Hepes-KOH (pH 7.9), 100 mm KCl, 0.5 mm DTT, 0.2 mm PMSF, 20% glycerol, and 10 µl of protease inhibitor mixture per 100 ml. The dialyzed nuclear extract was recovered into a 1.5-ml tube and centrifuged at 12,000 ×g for 20 min to remove precipitations. The protein concentration of the supernatant solution (nuclear extract) was measured, using the Bradford protein assay method (Bio-Rad), and 40-µl aliquots containing 0.7 µg of protein/µl were stored at −70 °C until used for electrophoretic mobility shift assays.

      Electrophoretic Mobility Shift Assay (EMSA)

      Annealed 5′-overhang oligonucleotides containing the sequence between −51/−34 and −45/−29 of the full-length Gαi2 gene promoter (ACCCCCGGCCCGCCCCGC and CGACGGCGGGGCGGGCC) were labeled with [α-32P]dCTP, using the Klenow fragment (3′ → 5′ exo) (New England Biolabs, Beverly, MA). The underlined nucleotides represent the overhangs. The reaction mixture (20 µl) contained 2 µl of 10× EcoPol Buffer (New England Biolabs), 3.5 pmol of oligonucleotide, 10 µCi of [α-32P]dCTP, 2 units of Klenow fragment, and 2 µl of 10× dNTP mix (dGTP, 500 µm; dCTP, 60 µm; dATP, 200 µm; dTTP, 250 µm). The reaction was allowed to proceed for 30 min at 25 °C and then stopped with 2 µl of 0.2m EDTA (pH 8.0). The labeled probe was purified by passing the reaction mixture through a Sephadex G-25 column (Amersham Pharmacia Biotech). The reaction mixture (25 µl) for the EMSA contained 5 µl of an EMSA 5× buffer from Promega (Madison, WI), 5 µg of nuclear extract protein, 0.2 µg of poly(dI-dC)·(dI-dC) (Amersham Pharmacia Biotech), and 20,000 cpm of the labeled oligonucleotide probe with or without competitors, as indicated in Fig. 3. The reaction was carried out at 25 °C for 20 min, and the product of the reaction was resolved by electrophoresis on a 4% non-denaturing polyacrylamide gel. After electrophoresis, the gel was dried and then exposed to a phosphorscreen for 24 h or less and visualized on a PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA).
      Figure thumbnail gr3
      Figure 3Electrophoretic mobility shift assay.EMSA was performed using nuclear extracts from control and sodium butyrate-treated K562 cells. The cells were harvested after 24 h in culture with sodium butyrate. The annealed oligonucleotide was labeled by using the Klenow fragment fill-in reaction, as described in detail under “Experimental Procedures.” The probe contains the consensus Sp1 sequence that we designated as site 2 in the Gαi2 gene promoter; two overlapping Sp1-binding motifs are present at that site. The reaction was carried out with 5 µg of nuclear extract protein. Competition experiments were carried out with 50-fold excess of (a) unlabeled oligonucleotide, ATTCGATCGGGGCGGGGCGAGC (from Promega), containing consensus Sp1 sequence (lanes 3 and 9); (b) an unlabeled oligonucleotide identical to the labeled probe except that nucleotides GC (in the probe) were replaced with AA (mutant) (lanes 4 and 10); and (c) an unrelated oligonucleotide, GATCGAACTGACCGCCCGCGGCCCGT (from Promega), (lanes 5 and 11), which contains a consensus activator protein 2 sequence (AP-2 consensus seq.).Lanes 6 and 12 represent supershifts with anti-Sp1 antibody. The position for Sp1 binding is indicated with ahorizontal arrow. NaBu, sodium butyrate.

      Immunoblotting of Gαi2 and Activated MAPK

      For measurement of Gαi2 levels, K562 cells (1 × 106) were preincubated with or without 20 µmU0126 for 30 min and then treated with 2.5 mm sodium butyrate for 24 h. The cells were washed twice with PBS and lysed for 20 min on ice, with a lysis buffer composed of 10 mmTris/HCl (pH 7.5), 1 mm EDTA, 150 mm NaCl, 4 mm MgCl2, 10 mm NaF, 5 mm DTT, 1% Triton X-100, 0.5% Nonidet P-40, 2 mm sodium orthovanadate, 1 µm leupeptin, 3 mm benzamidine, 0.1 unit/ml aprotinin, and 0.1 mm PMSF. The lysed cells were centrifuged at 6,700 ×g in a microcentrifuge for 10 min, and the supernatant solution (whole-cell lysates) was used for analysis of Gαi2 by immunoblotting as described previously (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ).
      For MAPK activation, K562 cells (5 × 105) were preincubated with or without 20 µm U0126 for 30 min and then treated with 2.5 mm sodium butyrate for various times up to 120 min. The cells were washed twice with cold PBS containing 1 mm sodium orthovanadate and then lysed as described above. Whole-cell lysates were then subjected to SDS-polyacrylamide gel electrophoresis, and separated proteins were transferred to Immobilon-P membranes as described previously (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). Phospho-ERKs and phospho-p38 MAPK were analyzed by immunoblotting with anti-ACTIVE ERK antibody (1:2,500 dilution) and anti-ACTIVE p38 antibody (1:2,000 dilution), respectively, using 40 µg of cell-lysate protein. About 2 µg of cell-lysate protein were used for blotting for total ERK (non-phosphorylated ERK). Bands were detected by chemiluminescence (PerkinElmer Life Sciences).

      RESULTS

      Sodium Butyrate Activates Transcription from the Gαi2 Gene Promoter

      Sodium butyrate increases Gαi2 levels in differentiating K562 cells (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). To understand further the molecular mechanism underlying the effect of sodium butyrate on Gαi2 gene expression, we used a reporter gene assay to monitor transcription. When a plasmid containing the full-length Gαi2 gene promoter linked to a luciferase reporter gene (pGαi2(−1214/+115)-luc) was transfected into K562 cells, the addition of sodium butyrate led to a 15.5-fold increase in transcription from this promoter compared with cells that were not treated with sodium butyrate (TableI). To decipher whether the increase in transcription from the Gαi2 gene promoter can be triggered by the differentiation process per se, the effects of various erythroid differentiation inducers (
      • McCaffrey P.G.
      • Newsome D.A.
      • Fibach E.
      • Yoshida M.
      • Su M.S.-S.
      ,
      • Adunyah S.E.
      • Chander R.
      • Barner V.K.
      • Cooper R.S.
      ) were tested; erythroid differentiation was verified by measuring induction of hemoglobin (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ) as a differentiation marker. As can be seen from Table I, HC toxin and trichostatin A (at low concentrations) caused a 4–5-fold increase in transcription from this promoter, whereas hydroxyurea, another potent erythroid differentiation inducer, had no effect on transcription. Interestingly, treatment with the phorbol ester, phorbol 12-myristate 13-acetate (10 nm), which causes these cells to differentiate toward monocytic and megakaryocytic lineages (
      • Racke F.K.
      • Lewandowska K.
      • Goueli S.
      • Goldfarb A.N.
      ,
      • Ajenjo N.
      • Aaronson D.S.
      • Ceballos E.
      • Richard C.
      • Leon J.
      • Crespo P.
      ), also induced transcription from the Gαi2 gene promoter (Table I). These results indicate that although the increase in transcription from the Gαi2 gene promoter can be triggered by the differentiation process per se, the effect is inducer-specific.
      Table IEffect of inducers of erythroid and megakaryocytic differentiation on Gαi2 promoter activity in K562 cells
      AdditionHemoglobin contenti2promoter activity
      pg/cell-Fold change
      -fold stimulation
      Erythroid differentiation inducer
      None0.41 ± 0.03 (9)1.01.0 (8)
      Sodium butyrate, 2.5 mm0.67 ± 0.04 (9)1.615.5 ± 1.3 (8)
      Hydroxyurea, 250 µm0.91 ± 0.07 (4)2.20.9 ± 0.1 (6)
      HC toxin, 10 nm0.67 ± 0.04 (5)1.64.7 ± 0.4 (7)
      HC toxin, 40 nm10.0 ± 1.0 (9)
      Trichostatin A, 66 nm0.72 ± 0.04 (5)1.84.4 ± 0.9 (4)
      Megakaryocyte differentiation inducer
      PMA, 10 nm43.7 ± 4.0 (6)
      K562 cells were seeded in 24-well plates (1 × 105 cells per well in 1 ml of medium) and cultured for 24 h as described under “Experimental Procedures.” Erythroid differentiation was determined 24 h after the addition of inducers, as the induction of hemoglobin expression (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). Each condition was replicated three times for the number of separate cell cultures indicated in parentheses. The expression of integrin β3 (CD61) was measured as a marker of megakaryocyte differentiation, using Western blotting. To measure promoter activity, cells were transfected with 0.5 µg of plasmid DNA (pGαi2(−1214/+115)-luc) containing the full-length promoter for the Gαi2 gene, 1 h before the addition of the cell differentiation inducers tested. The cells were harvested for luciferase assay 24 h later. Each condition was replicated four times for the number of separate cell cultures indicated in parentheses. Promoter activity was measured as the relative luciferase activity (relative luciferase activity/µg protein) of the cell extracts and is expressed as -fold stimulation, relative to cells that were not treated with any inducer. Values shown are means ± S.E. PMA, phorbol 12-myristate 13-acetate.
      To decipher what region(s) of the promoter mediated the induction, several truncated constructs of the promoter were made and tested in the transfection assay. The truncations were designed to avoid cutting through any of the several Sp1-like sites (at least seven, one of which is present in the 5′-untranslated region) (
      • Weinstein L.S.
      • Spiegel A.M.
      • Carter A.D.
      ,
      • Weinstein L.S.
      • Kats I.
      • Spiegel A.M.
      • Carter A.D.
      ,
      • Quandt K.
      • Frech K.
      • Karas H.
      • Wingender E.
      • Werner T.
      ) that are contained in the Gαi2 promoter. Transcriptional activity was not affected when −1214 through −184 fragment was deleted from the promoter. However, when the deletion was extended to −79, there was a 60% reduction in transcriptional activity (Fig.1). This suggests that a region (or regions) of the promoter between −184 and −79 is/are required for the full response of the Gαi2 gene promoter to sodium butyrate.
      We next tested substitution mutations and block deletions in the upstream region in order to delineate the specific DNA sequence(s) involved in the sodium butyrate effect on the Gαi2 gene transcription. These substitution mutations/deletions involved three of the seven Sp1-like sequence motifs. Mutations affecting only the putative Sp1 site at +68/+75 in the 5′-untranslated region (mutant pM1) had no effect on the sodium butyrate-induced activation of transcription, but mutation at this site in conjunction with mutation at −50/−36 decreased transcription (pM1,2in Fig. 2 A). Mutations at the putative Sp1 sites at −50/−36 and −92/−85 (mutants pM2and pM3) decreased transcription by about 30 and 38%, respectively (Fig. 2 A). Mutating both of these Sp1 sequences, in the same construct, reduced transcription even further (>50% reduction) (pM2,3 in Fig. 2 A), indicating the importance of these sites (especially the Sp1 site at −50/−36, designated as site 2) in the sodium butyrate-induced activation of transcription. This becomes even more obvious when single and double deletions involving these sites were tested (Fig.2 B). The data in Fig. 2 B complement those in Fig.2 A; therefore, we conclude that the putative Sp1 sites at −92/−85 and −50/−36 play an important role in the sodium butyrate response.

      Detection of Sp1 Binding to Gαi2 Gene Promoter

      To confirm the involvement of Sp1 in the sodium butyrate response, electrophoretic mobility shift assays were performed (Fig.3) with a labeled synthetic double-stranded DNA probe containing the putative Sp1-binding sequence present at −50/−36. This sequence was chosen as probe because mutations within this sequence or its deletion resulted in the greatest reduction in transcription (Fig. 2), and because the −50/−36 segment contains two overlapping Sp1-binding motifs (
      • Quandt K.
      • Frech K.
      • Karas H.
      • Wingender E.
      • Werner T.
      ). A consensus Sp1 oligonucleotide completely abolished the binding of nuclear proteins to the labeled DNA (Fig. 3, lane 3), whereas mutation of the two bases GC to AA within this sequence had no effect on the binding (Fig. 3, lane 4). A similar competitive effect was noted for the consensus Sp1 oligonucleotide in extracts prepared from sodium butyrate-treated cells (lane 9); again the mutated oligonucleotide did not compete for binding to the DNA (lane 10). Similarly, an unrelated oligonucleotide (activator protein 2 oligonucleotide) had no effect (lanes 5 and 11). In the presence of an antibody to Sp1, a marked supershift was evident (lanes 6 and 12) in the band position for Sp1. We conclude that the sequence −50/−36 within the Gαi2 gene promoter functions as a binding site for Sp1. The intensity of the Sp1 signal was greater in sodium butyrate-treated cells than in control cells, suggesting either an increased affinity of nuclear proteins for the labeled DNA or an increased nuclear content of this transcription factor in the sodium butyrate-treated cells.

      Effect of Trichostatin A and HC Toxin on Transcription from the Full-length or Mutant Gαi2 Gene Promoter

      Sodium butyrate inhibits histone deacetylation (
      • Riggs M.G.
      • Whittaker R.G.
      • Neumann J.R.
      • Ingram V.M.
      ). This effect is the basis for the long-standing concept that the action of sodium butyrate on gene transcription is related to alterations on chromatin structure resulting from butyrate-induced hyperacetylation of histones (
      • Kruh J.
      ). All of the compounds that induce erythroid differentiation that were tested in Table I are histone deacetylase inhibitors, except hydroxyurea. Therefore, we determined the influence of histone hyperacetylation on the overall activation of transcription from the Gαi2gene promoter, by assessing the effect of trichostatin A and HC toxin, both of which are potent histone deacetylase inhibitors (
      • Yoshida M.
      • Kijima M.
      • Akita M.
      • Beppu T.
      ,
      • Brosch G.
      • Ransom R.
      • Lechner T.
      • Walton J.D.
      • Loidl P.
      ). Trichostatin A is particularly interesting because its action on gene transcription is reported to involve Sp1 sites (
      • Sowa Y.
      • Orita T.
      • Minamikawa S.
      • Nakano K.
      • Mizuno T.
      • Nomura H.
      • Sakai T.
      ,
      • Sowa Y.
      • Orita T.
      • Hiranabe-Minamikawa S.
      • Nakano K.
      • Mizuno T.
      • Nomura H.
      • Sakai T.
      ,
      • Doetzlhofer A.
      • Rotheneder H.
      • Lagger G.
      • Koranda M.
      • Kurtev V.
      • Brosch G.
      • Wintersberger E.
      • Seiser C.
      ). Fig.4 A shows that trichostatin A activated transcription from the Gαi2 promoter in a dose-dependent manner, when used at up to 50 ng/ml. Because this compound is known to be cytotoxic at high concentrations (see Fig.4 B), it was important to carry out subsequent transfection experiments with low concentrations (20–30 ng/ml) of this drug. With 30 ng of trichostatin A per ml, and using the single and double deletion mutants studied in Fig. 2, the reporter gene activity was only 0.3–0.5-fold increased above background (empty vector) (Fig.5 A), a pattern almost similar to that seen with sodium butyrate (see Fig. 2 B). With HC toxin, the reporter gene activity was also reduced to only 33 and 50% of the full-length promoter, when the double deletion mutant (pD2,3) and the double substitution mutant (pM2,3), respectively, were tested (Fig. 5 B). This pattern was essentially similar to that noted for either trichostatin A (Fig. 5 A) or sodium butyrate (Fig.2 B), suggesting some commonality in the mode of action of these compounds in inducing transcription from the Gαi2gene promoter.
      Figure thumbnail gr4
      Figure 4Effect of trichostatin A on cell viability and on transcription from the Gαi2gene promoter. K562 cells were grown in RPMI 1640 medium in the absence or presence of various concentrations of trichostatin A for 24 h. The promoter activity (A) is indicated as fold induction compared with cell cultures that were not treated with trichostatin A. Cell numbers (B) were determined by using a hemocytometer. The results are means ± S.E. of six experiments.
      Figure thumbnail gr5
      Figure 5Substitution mutations in, or deletion of, putative Sp1 sites at −50/−36 and −92/−85 decrease trichostatin A- and HC toxin-induced transcription from the Gαi2 gene promoter. The plasmids are the same as in Fig. . The results (means ± S.E. of 6–8 experiments) are expressed as fold stimulation, relative to cells that were not treated with trichostatin A (30 ng/ml) (A) or HC toxin (40 nm) (B). *, statistically significant (p < 0.002) compared with the full-length promoter.

      Involvement of the MAPK Pathway in Butyrate-induced Activation of the Gαi2 Promoter

      It has been reported that early effects of sodium butyrate on the differentiation of K562 cells might involve activation of the MAPK pathway (
      • Rivero J.A.
      • Adunyah S.E.
      ). Because we have shown previously that the sodium butyrate-induced expression of Gαi2 protein parallels the differentiation event (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ), it seemed reasonable to ask whether events upstream of Sp1 activation might contribute to the sodium butyrate-induced expression of Gαi2 and whether such events might involve components of a known signal transduction cascade, such as the MAPK cascade. As shown in Fig. 6 A, incubation of these cells with sodium butyrate (2.5 mm) resulted in rapid (within 5 min) activation of ERK, primarily ERK2; the signals for ERK1 were not detectable, an indication that these cells contain low levels of this ERK isoform. U0126, an MEK inhibitor (
      • Favata M.F.
      • Horiuchi K.Y.
      • Manos E.J.
      • Daulerio A.J.
      • Stradley D.A.
      • Feeser W.S.
      • Van Dyk D.E.
      • Pitts W.J.
      • Earl R.A.
      • Hobbs F.
      • Copeland R.A.
      • Magolda R.L.
      • Scherle P.A.
      • Trzaskos J.M.
      ), completely inhibited the sodium butyrate-dependent activation of ERK (Fig.6 A, lanes 8 and 9). Blotting of these same samples with anti-ACTIVE p38 antibody failed to detect any phosphorylated p38, indicating that p38 MAPK was not activated by the sodium butyrate treatment. It should be noted that both sodium butyrate- and HC toxin-induced erythroid differentiation, measured as hemoglobin accumulation, was completely suppressed by U0126 (TableII). In contrast, this inhibitor did not affect the erythroid differentiation-inducing effect of trichostatin A, indicating a clear difference in the mode of action of this compound compared with that of sodium butyrate and HC toxin. Western blotting analyses of cell extracts of cultures treated with 20 µmU0126 for 30 min before the addition of sodium butyrate showed that Gαi2 levels were only 30% of the levels in cells that were not treated with U0126 (Fig. 6 B). These data strongly suggest the involvement of the MEK-ERK signal transduction pathway in the sodium butyrate-induced expression of Gαi2.
      Figure thumbnail gr6
      Figure 6Western blot analysis of ERKs and Gαi2 levels in sodium butyrate-treated K562 cells and the effect of U0126 on sodium butyrate-induced Gαi2 and phospho-ERK levels. A, time course of changes in the activation state of ERKs after addition of sodium butyrate. ERKs were detected as described under “Experimental Procedures.” Inlanes 8 and 9, the effect of MEK inhibitor U0126 on sodium butyrate-induced phospho-ERK levels was measured. K562 cells (5 × 105) were preincubated with or without 20 µm U0126 for 30 min and then treated with 2.5 mm sodium butyrate for 5–120 min. Whole-cell lysates were prepared, and the activation of ERK was analyzed by Western blotting, using anti-ACTIVE ERK antibody. To detect phosphorylated ERK and non-phosphorylated ERK, 40 and 2 µg of protein, respectively, from the cell lysate were used in the blotting protocol. B, K562 cells (1 × 106) were preincubated with or without 20 µm U0126 for 30 min and then cultured in the presence or absence of 2.5 mm sodium butyrate for 24 h. Gαi2 levels in whole-cell lysates were analyzed by Western blotting, using 10 µg of protein per sample, as described previously (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). The results shown are means ± S.E. for four experiments. The fold change over control levels was measured as arbitrary units of densitometric scans of Gαi2 signals detected on Western blots. Inset, representative Western blot. Control, no sodium butyrate; NaBu, 2.5 mm sodium butyrate; NaBu + U0126 , 2.5 mm sodium butyrate + 10 or 20 µm U0126. *, significantly different (p < 0.05) compared with sodium butyrate alone.
      Table IIEffect of MEK inhibitor U0126 on the induction of hemoglobin by HC toxin, trichostatin A, and sodium butyrate
      ConditionExperimental series I (n = 4), hemoglobinExperimental series II (n = 3), hemoglobin
      pg/cell% controlpg/cell% control
      Control0.50 ± 0.011000.37 ± 0.01100
      Sodium butyrate0.70 ± 0.031400.55 ± 0.02149
      Sodium butyrate + U01260.51 ± 0.021020.39 ± 0.06105
      Trichostatin A0.78 ± 0.03156
      Trichostatin A + U01260.79 ± 0.01158
      HC toxin0.60 ± 0.04162
      HC toxin + U01260.35 ± 0.0295
      K562 cells were cultured for 24 h as described in Table I. The cells were treated with or without 10 µm U0126 for 30 min before the addition of HC toxin (10 nm), trichostatin A (66 nm), or sodium butyrate (2.5 mm). The hemoglobin content of the cells was measured 24 h later (
      • Davis M.G.
      • Kawai Y.
      • Arinze I.J.
      ). The results are expressed as the means ± S.E. for three or four experiments.
      To test if the MEK-ERK signal transduction pathway is involved in the transcriptional activation of the Gαi2 gene promoter, the luciferase reporter gene activity was measured in cell extracts prepared from cultures treated with U0126. At 10 or 20 µm, this inhibitor drastically inhibited (∼61% inhibition) the sodium butyrate-induced transcriptional activation of the Gαi2 gene promoter but had no effect on trichostatin A-induced transcription (Fig.7 A). U0126 also inhibited HC toxin-induced promoter activity (Fig. 7 A), suggesting that this inducer, like sodium butyrate, may also be activating the MEK-ERK signal transduction pathway. To confirm this notion, an antisense oligonucleotide to ERK was used to decrease the endogenous levels of ERK, an approach that has been used to deplete ERK expression in 3T3-L1 cells (
      • Sale E.M.
      • Atkinson P.G.
      • Sale G.J.
      ). When the cultures were treated with an antisense oligonucleotide to ERK (Fig. 7 B), the results were qualitatively similar to those obtained with U0126. In both cases, transcriptional activation by sodium butyrate and HC toxin was inhibited, but no change was noted for transcriptional activation by trichostatin A. These results indicate that the actions of butyrate and HC toxin require the MEK-ERK signal transduction pathway.
      Figure thumbnail gr7
      Figure 7Effects of inhibitors of MAPKs and antisense oligonucleotide to ERK on cell differentiation inducer-activated transcription from the Gαi2 gene promoter. K562 cells were transfected with pGαi2(−1214/+115)-luc as described under “Experimental Procedures.” After 1 h, U0126 (A), SB 203580 (C), or PD 169316 (D) was added, and the cells were incubated for 30 min, followed by the addition of sodium butyrate (2.5 mm), trichostatin A (30 ng/ml), or HC toxin (40 nm). B, 105 cells/ml were co-transfected with 0.5 µg of pGαi2(−1214/+114)-luc and different amounts of antisense oligonucleotide (5′-GCCGCCGCCGCCGCCAT-3′) to ERK, as indicated in the figure. Unrelated oligonucleotides-1 (O-1) and -2 (O-2) (5′-CACCGACTTCTTGGCTT-3′ and 5′-CAAAGAGCAGCGAGAAG-3′) used as controls (50 nm) were also co-transfected with pGαi2(−1214/+115)-luc (0.5 µg) 30 min before addition of sodium butyrate (2.5 mm). These sequences were preferred as controls instead of a randomized sequence of the antisense oligonucleotide because the high GCC content in a randomized antisense material might have high potential to target the desired mRNA molecules. In all cases, the cells were harvested 24 h after the addition of the cell differentiation inducing agent, and the luciferase activity of cell lysates was determined as described under “Experimental Procedures.” Promoter activity is expressed as fold stimulation over the appropriate controls. Values shown are means ± S.E. for replicate assays for 5–7 experiments. NaBu, sodium butyrate; TSA, trichostatin A; HC, HC toxin; SB, SB 203580; PD, PD 169316. *, significantly different (p < 0.03) from cells treated with sodium butyrate or HC toxin without inhibitor or antisense oligonucleotide. **, significantly different (p < 0.01) from cells treated with sodium butyrate or HC toxin without inhibitor or antisense oligonucleotide.
      To ascertain whether other MAPK signal transduction pathways are involved in the transcriptional activation by butyrate, inhibitors of p38 and JNK MAPKs were tested. SB 203580, an inhibitor of some p38 MAPKs (
      • Cuenda A.
      • Rous J.
      • Doza Y.N.
      • Meier R.
      • Cohen P.
      • Gallagher T.F.
      • Young P.R.
      • Lee J.C.
      ), did not inhibit transcription from this promoter (Fig.7 C), irrespective of which inducer was tested. This result indicates that p38 MAPK does not mediate the transcriptional activation measured here. We also tested the effect of PD 169316, an inhibitor of both the JNK and p38 MAPKs (
      • Kummer J.L.
      • Rao P.K.
      • Heidenreich K.A.
      ,
      • Assefa Z.
      • Vantieghem A.
      • Declercq W.
      • Vandenabeele P.
      • Vandenheede J.R.
      • Merlevede W.
      • de Witte P.
      • Agostinis P.
      ,
      • Zhang Y.
      • Zhong S.
      • Dong Z.
      • Chen N.
      • Bode A.M.
      • Ma W.
      • Dong Z.
      ). This inhibitor had no effect on sodium butyrate-induced transcription, but it inhibited HC toxin-induced transcription (Fig. 7 D), indicating that JNK is involved in the signaling induced by HC toxin but not in the signaling induced by sodium butyrate. Thus, these results implicate the MEK-ERK signal transduction pathway in the overall induction of Gαi2 by sodium butyrate.
      If activation of the MEK-ERK signal transduction pathway leads to activation of transcription from the Gαi2 gene promoter, then the use of constitutively activated ERK1/2 should activate transcription from this promoter. To test this idea, we co-transfected K562 cells with plasmid DNA containing either constitutively activated ERK1 (pCHA-ERK1) or ERK2 (pcDNA3-ERK2), together with the Gαi2 gene promoter. At all concentrations tested, the pCHA-ERK1 had no effect on transcription from this promoter, whereas up to 4.2-fold induction of transcription was obtained with 1 µg of pcDNA3-ERK2 (Fig. 8 A), a strong indication that activation of the MEK-ERK (presumably ERK2) pathway can lead to increased transcription from the Gαi2gene promoter. At higher concentrations of ERK2 (e.g. 1.5 µg), promoter activation was somewhat reduced (1.4-fold) but was still higher than that noted for ERK1. To determine whether ERK2per se activates transcription through the Sp1 sites on the Gαi2 gene promoter, transfections with pcDNA3-ERK2 were carried out with mutant promoters M2,3 and D2,3. The results (Fig. 8 B) show that, as with sodium butyrate, deletion of Sp1 sequences at −50/−36 and −92/−85 caused drastic reduction (53–60%) in transcription, suggesting that the same Sp1 sequences are mediating both ERK2-dependent and sodium butyrate-dependent activation of transcription from the Gαi2 gene promoter.
      Figure thumbnail gr8
      Figure 8Constitutively activated ERK2, but not ERK1, induces transcription from the Gαi2gene promoter. A, K562 cells were transfected with up to 1.5 µg of plasmid DNA harboring sequences for activated ERK1 or ERK2. Promoter activity was measured by the luciferase reporter gene assay as described under “Experimental Procedures.” B,mutations in the Gαi2 promoter decrease the promoter activation effect of constitutively activated ERK2. The mutant promoters used were the same as in Fig. . C, constitutively activated ERK2 overrides U0126-inhibited butyrate activation of transcription. The amount of added ERK2 was the same (1 µg) at each inhibitor concentration. *, significantly different (p< 0.05) from cells treated with sodium butyrate and U0126. **, significantly different (p < 0.01) from cells treated with sodium butyrate and U0126.
      Inhibition of sodium butyrate-dependent activation of transcription from the Gαi2 gene promoter by U0126, but not by SB 203580 or PD 169316, is a clear indication that the MEK-ERK signal transduction pathway is involved in the inductive effect of sodium butyrate (Fig. 7). The results of the experiments with antisense oligonucleotide to ERK (Fig. 7 B) are consistent with this conclusion. In additional experiments, we show that in cells in which sodium butyrate induction of transcription had been inhibited by various concentrations of U0126, transfection of constitutively activated ERK2 (pcDNA3-ERK2) was capable of reversing such inhibition (Fig. 8 C). These data clearly support the conclusion that the MEK-ERK, albeit ERK2, signal transduction pathway is involved in the sodium butyrate-induced activation of transcription from the Gαi2 gene promoter.

      DISCUSSION

      Sodium butyrate induces differentiation as well as apoptosis in several cell types (
      • Kruh J.
      ,
      • Heerdt B.G.
      • Houston M.A.
      • Augenlicht L.H.
      ,
      • Janson W.
      • Brandner G.
      • Siegel J.
      ). It can also affect gene transcription in a positive (
      • Leder A.
      • Leder P.
      ,
      • Birren B.W.
      • Herschman H.R.
      ,
      • Lazar M.A.
      ,
      • Rius C.
      • Cabanas C.
      • Aller P.
      ,
      • Nakano K.
      • Mizuno T.
      • Sowa Y.
      • Orita T.
      • Yoshino T.
      • Okuyama Y.
      • Fugita T.
      • Ohtani-Fujita N.
      • Matsukawa Y.
      • Tokino T.
      • Yamagishi H.
      • Oka T.
      • Nomura H.
      • Sakai T.
      ,
      • Lu Y.
      • Lotan R.
      ,
      • Benvenuto G.
      • Carpentieri M.L.
      • Salvatore P.
      • Cindolo L.
      • Bruni C.B.
      • Chiariotti L.
      ,
      • Tsuji Y.
      • Moran E.
      • Torti S.V.
      • Torti F.M.
      ) or negative (
      • Heruth D.P.
      • Zimstein G.W.
      • Bradley J.F.
      • Rothberg P.G.
      ) manner, depending on the gene. Metabolically, the interest in sodium butyrate stems from the fact that it is a major short chain fatty acid produced in the human by bacterial fermentation activity in the colon (
      • Cummings J.H.
      ,
      • Roediger W.E.W.
      ). It is a major dietary lipid in cow's milk, and it is the short chain fatty acid that exerts significant effects on colonic epithelial cells in vitro andin vivo (
      • Scheppach W.
      • Bartram P.
      • Richter A.
      • Richter F.
      • Liepold H.
      • Dusel G.
      • Hofstetter G.
      • Rüthlein J.
      • Kasper H.
      ,
      • McBain J.A.
      • Eastman A.
      • Nobel C.S.
      • Mueller G.C.
      ). In erythroid cells, butyrate induces the accumulation of fetal hemoglobin (
      • Constantoulakis P.
      • Knitter G.
      • Stamatoyannopoulos G.
      ,
      • Glauber J.G.
      • Wandersee N.J.
      • Little J.A.
      • Ginder G.D.
      ). Indeed, butyrate has been used as a booster of fetal hemoglobin production in sickle cell disease (
      • Dover G.J.
      • Humphries R.K.
      • Moore J.G.
      • Ley T.J.
      • Young N.S.
      • Charache S.
      • Nienhuis A.W.
      ), and failure to oxidize it has been implicated in ulcerative colitis (
      • Roediger W.E.W.
      ).
      The precise mechanism of action of butyrate in cell differentiation, apoptosis, and gene expression is not understood. Because sodium butyrate inhibits histone deacetylase (
      • Kruh J.
      ), and because hyperacetylation of histones can lead to alterations in chromatin structure, resulting in conditions that favor accessibility of transcription factors to DNA, the transcriptional and other effects of butyrate are often ascribed to its ability to effect histone hyperacetylation (
      • Kruh J.
      ). However, other mechanisms, such as enrichment of available acetyl groups (resulting from butyrate metabolism), acetylation of non-histone proteins, or binding to proteins other than histone deacetylase, may play a role in some of the observed effects of butyrate. In the absence of histones, acetyl-CoA induces conformational changes in the transcription factor IID-transcription factor IIA-DNA complex in vitro, leading to activation and enhancement of transcription in vitro (
      • Galasinski S.K.
      • Lively T.N.
      • Grebe de Barron A.
      • Goodrich J.A.
      ). Whether this phenomenon occursin vivo is unknown, but this finding suggests that butyrate (through butyryl-CoA) might produce a similar effect on transcription or other cellular processes in the nucleus. The butyrate-activated transcription of at least three genes (
      • Nakano K.
      • Mizuno T.
      • Sowa Y.
      • Orita T.
      • Yoshino T.
      • Okuyama Y.
      • Fugita T.
      • Ohtani-Fujita N.
      • Matsukawa Y.
      • Tokino T.
      • Yamagishi H.
      • Oka T.
      • Nomura H.
      • Sakai T.
      ,
      • Lu Y.
      • Lotan R.
      ,
      • Tsuji Y.
      • Moran E.
      • Torti S.V.
      • Torti F.M.
      ) involves Sp1 sites. For example, sodium butyrate activation of transcription from the galectin-1 gene promoter involves an Sp1 site proximal to the transcription start site (
      • Lu Y.
      • Lotan R.
      ). Sp1 sites also seem to mediate sodium butyrate-induced transcription from the ferritin H gene promoter (
      • Tsuji Y.
      • Moran E.
      • Torti S.V.
      • Torti F.M.
      ), and the butyrate activation of the WAF1/CiP1 gene in a p53-negative human colonic cell line involves Sp1 sites (
      • Nakano K.
      • Mizuno T.
      • Sowa Y.
      • Orita T.
      • Yoshino T.
      • Okuyama Y.
      • Fugita T.
      • Ohtani-Fujita N.
      • Matsukawa Y.
      • Tokino T.
      • Yamagishi H.
      • Oka T.
      • Nomura H.
      • Sakai T.
      ). In the present study, we show that sodium butyrate-induced expression of Gαi2 involves Sp1 sites at −92/−85 and −50/−36 in the Gαi2 gene promoter but does not involve the putative Sp1 site (+68/+75) within the 5′-untranslated region.
      Sp1, originally discovered on the basis of its ability to activate selectively transcription from the viral SV40 promoter (
      • Dynan W.S.
      • Tjian R.
      ,
      • Dynan W.S.
      • Tjian R.
      ), is a ubiquitous transcription factor that has been implicated in the control of cell cycle-regulated genes such as thymidine kinase (
      • Karlseder J.
      • Rotheneder H.
      • Wintersberger E.
      ), B-myb (
      • Lin S.-Y.
      • Black A.R.
      • Kostic D.
      • Pajovic S.
      • Hoover C.N.
      • Azizkhan J.C.
      ), and dihydrofolate reductase (
      • Zwicker J.
      • Liu N.
      • Engeland K.
      • Lucibello F.C.
      • Müller R.
      ). Sp1 has also been reported to be involved in the transcriptional regulation of several other genes (
      • Nakano K.
      • Mizuno T.
      • Sowa Y.
      • Orita T.
      • Yoshino T.
      • Okuyama Y.
      • Fugita T.
      • Ohtani-Fujita N.
      • Matsukawa Y.
      • Tokino T.
      • Yamagishi H.
      • Oka T.
      • Nomura H.
      • Sakai T.
      ,
      • Lu Y.
      • Lotan R.
      ,
      • Tsuji Y.
      • Moran E.
      • Torti S.V.
      • Torti F.M.
      ,
      • Khana-Gupta A.
      • Zibello T.
      • Simkevich C.
      • Rosmarin A.G.
      • Berliner N.
      ,
      • Merchiers P.
      • Bulens F.
      • De Vriese A.
      • Collen D.
      • Belayew A.
      ,
      • Netzker R.
      • Fabian D.
      • Weigert C.
      • Brand K.A.
      ,
      • Le H.B.
      • Vaisanen P.A.
      • Johnson J.L.
      • Raney A.K.
      • McLachlan A.
      ,
      • Daniel S.
      • Zhang S.
      • DePaoli-Roach A.A.
      • Kim K.-H.
      ,
      • Zhang W.
      • Geiman D.E.
      • Shields J.M.
      • Dang D.T.
      • Mahatan C.S.
      • Kaestner K.H.
      • Biggs J.R.
      • Kraft A.S.
      • Yang V.W.
      ,
      • Zhang D.E.
      • Hetherington C.J.
      • Tan S.
      • Dziennis S.E.
      • Gonzalez D.A.
      • Chen H.M.
      • Tenen D.G.
      ,
      • Merchant J.L.
      • Shiotani A.
      • Mortensen E.R.
      • Shumaker D.K.
      • Abraczinskas D.R.
      ) as well, including at least four genes coding for enzymes in key metabolic pathways,viz. pyruvate kinase M (
      • Merchiers P.
      • Bulens F.
      • De Vriese A.
      • Collen D.
      • Belayew A.
      ), aldolase (
      • Netzker R.
      • Fabian D.
      • Weigert C.
      • Brand K.A.
      ), phosphofructokinase P2 (
      • Le H.B.
      • Vaisanen P.A.
      • Johnson J.L.
      • Raney A.K.
      • McLachlan A.
      ), and acetyl-CoA carboxylase (
      • Daniel S.
      • Zhang S.
      • DePaoli-Roach A.A.
      • Kim K.-H.
      ). Thus, studies directed at how Sp1 mediates butyrate activation of gene expression should be useful in our understanding of various cellular processes including metabolism, development, and differentiation.
      Like many transcription factors, Sp1-mediated transcription involves interaction with other transcriptional co-activators or repressors. For example, a cooperative interaction between Sp1 and nuclear factor-κB is required for human immunodeficiency virus, type I, activation in Jurkat T cells (
      • Perkins N.D.
      • Edwards N.L.
      • Duckett C.S.
      • Agranoff A.B.
      • Schmid R.M.
      • Nabel G.J.
      ). Sp1-dependent transcription is also influenced by two members of the retinoblastoma protein family, retinoblastoma protein and p107, in a cell cycle-dependent manner (
      • Chen L.I.
      • Nishinaka T.
      • Kwan K.
      • Kitabayashi I.
      • Yokoyama K.
      • Fu Y.-H.F.
      • Grunwald S.
      • Chiu R.
      ,
      • Datta P.K.
      • Raychaudhuri P.
      • Bagchi S.
      ). The TBP-associated factor 110 and TBP itself also associate with Sp1 and Sp1-like transcription factors (
      • Lania L.
      • Majello B.
      • De Luca P.
      ). Other co-activators for transcriptional activation by Sp1 have been reviewed (
      • Naar A.M.
      • Ryu S.
      • Tjian R.
      ). However, the role of specific Sp1-like transcription factors in transcription from a given promoter remains to be elucidated. It may be relevant to note that for the lactoferrin gene promoter, which contains a C/EBP site flanked by two Sp1 sites, there is a functional interaction between C/EBPα and Sp1 in mediating lactoferrin gene expression during myeloid differentiation (
      • Khana-Gupta A.
      • Zibello T.
      • Simkevich C.
      • Rosmarin A.G.
      • Berliner N.
      ). However, the authors were unable to demonstrate physical interaction between the two transcription factors, using gel shift assays. Doetzlhofer et al. (
      • Doetzlhofer A.
      • Rotheneder H.
      • Lagger G.
      • Koranda M.
      • Kurtev V.
      • Brosch G.
      • Wintersberger E.
      • Seiser C.
      ) have reported that Sp1 is tightly associated with histone deacetylase 1 and that the two proteins may be part of one complex. Their studies on the trichostatin A-mediated activation of the thymidine kinase gene promoter show that Sp1 is a target for histone deacetylase 1-mediated transcriptional repression (
      • Doetzlhofer A.
      • Rotheneder H.
      • Lagger G.
      • Koranda M.
      • Kurtev V.
      • Brosch G.
      • Wintersberger E.
      • Seiser C.
      ). Thus, inhibition of histone deacetylase by trichostatin A releases an inhibitory constraint on Sp1, making it possible for this transcription factor to associate with other accessory proteins (e.g.ECF-2) to effect transcription of the thymidine kinase gene. By analogy, it is also possible that hyperacetylation of transcription components other than histones may occur in the presence of butyrate, although this remains to be tested.
      In this study, we noted that deletion of critical Sp1 sites from the Gαi2 gene promoter did not completely suppress transcription in our cellular transfection system (Fig. 2 and Fig. 5). Although other transcription factors can be expected to be involved, events upstream of Sp1 activation also appear to be important in the activation of this promoter by sodium butyrate. We show here that the MEK-ERK signal transduction pathway is involved in the action of butyrate and HC toxin. U0126 specifically inhibits MEK1/2 (
      • Favata M.F.
      • Horiuchi K.Y.
      • Manos E.J.
      • Daulerio A.J.
      • Stradley D.A.
      • Feeser W.S.
      • Van Dyk D.E.
      • Pitts W.J.
      • Earl R.A.
      • Hobbs F.
      • Copeland R.A.
      • Magolda R.L.
      • Scherle P.A.
      • Trzaskos J.M.
      ), thereby inhibiting activation of ERK1 and ERK2 that mediate signaling downstream of MEK1/2 in the Raf-MEK-ERK signal transduction pathway. When this MAPK signal transduction pathway was inhibited by U0126, the sodium butyrate-induced increase in Gαi2 levels was inhibited (Fig. 6 B), as was the ability of sodium butyrate to induce differentiation (Table II). Furthermore, the transcriptional effect of butyrate or HC toxin, measured with the reporter gene assay, was also inhibited. In sharp contrast, U0126 had no effect on trichostatin A-induced transcription (Fig. 7 A). The use of antisense oligonucleotide to ERK confirmed these results. Thus, the pathways involved in the actions of these transcriptional activators are different; the actions of sodium butyrate and HC toxin involve the MEK-ERK signal transduction pathway, that of trichostatin A does not. We used the effects of PD 169316 to distinguish the mode of action of sodium butyrate from that of HC toxin. Furthermore, because SB 203580, an inhibitor of the p38 MAPK signal transduction pathway, had no inhibitory effect on transcription induced by any of the transcriptional activators tested, we interpret these results to mean that the MEK-ERK signal transduction pathway is involved in sodium butyrate signaling which eventually converges in the cell nucleus. The involvement of the MEK-ERK signal transduction pathway was also confirmed by transfection experiments in which we used plasmids harboring constitutively activated ERKs (Fig. 8).
      Based on detection of phospho-ERKs on Western blots after butyrate treatment, activation of MEK-ERK has been implicated in the butyrate-induced transcription of the choline acetyltransferase gene in CHP126 neuroblastoma cells (
      • Espinos E.
      • Weber M.J.
      ). In the present study, we also demonstrate activation of ERK induced by treatment with sodium butyrate; blockade of this activation resulted in inhibition of butyrate-induced transcription of the Gαi2 gene promoter. To our knowledge, the result we present here is the first study linking the expression of Gαi2, or for that matter any G-protein gene, to the action of Sp1 and the signal transduction pathway involving MEK-ERK. More work will be needed to delineate the components of this pathway that may be involved in butyrate-induced signaling. By using differential display procedures, Courilleau et al. (
      • Courilleau D.
      • Chastre E.
      • Sabbah M.
      • Redeuilh G.
      • Atfi A.
      • Mester J.
      ) have recently identified a novel protein, named B-ind1 (butyrate-induced protein 1), that appears to mediate Rac1 signaling in Balb/c mouse fibroblasts treated with sodium butyrate. Whether such a protein might be part of the phenomenon observed in our study remains to be tested.

      Acknowledgments

      We thank Dr. Lee Weinstein (National Institutes of Health) for the generous gift of plasmid pGαi2(−1214/+115)-CAT containing the full-length promoter for the Gαi2 gene, and Dr. Michael J. Weber (University of Virginia, Charlottesville) for providing us with the plasmids harboring genes for constitutively activated ERKs (pCHA-ERK1 and pcDNA3-ERK2) and the empty vectors (pCHA and pcDNA3) used to clone these ERK genes.

      REFERENCES

        • Wang H.
        • Watkins D.C.
        • Malbon C.C.
        Nature. 1992; 358: 334-337
        • Jho E.-H.
        • Malbon C.C.
        J. Biol. Chem. 1997; 272: 24461-24467
        • Jho E.-H.
        • Davis R.J.
        • Malbon C.C.
        J. Biol. Chem. 1997; 272: 24468-24474
        • Amatruda III, T.T.
        • Steele D.A.
        • Slepak V.Z.
        • Simon M.I.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 5587-5591
        • Wilke T.M.
        • Scherle P.A.
        • Strathmann M.P.
        • Slepak V.Z.
        • Simon M.I.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10049-10053
        • Watkins D.C.
        • Johnson G.L.
        • Malbon C.C.
        Science. 1992; 258: 1373-1375
        • Davis M.G.
        • Kawai Y.
        • Arinze I.J.
        Biochem. J. 2000; 346: 455-461
        • Leder A.
        • Leder P.
        Cell. 1975; 5: 319-322
        • Birren B.W.
        • Herschman H.R.
        Nucleic Acids Res. 1986; 14: 853-867
        • Lazar M.A.
        J. Biol. Chem. 1990; 265: 17474-17477
        • Rius C.
        • Cabanas C.
        • Aller P.
        Exp. Cell Res. 1990; 188: 129-134
        • Nakano K.
        • Mizuno T.
        • Sowa Y.
        • Orita T.
        • Yoshino T.
        • Okuyama Y.
        • Fugita T.
        • Ohtani-Fujita N.
        • Matsukawa Y.
        • Tokino T.
        • Yamagishi H.
        • Oka T.
        • Nomura H.
        • Sakai T.
        J. Biol. Chem. 1997; 272: 22199-22206
        • Lu Y.
        • Lotan R.
        Biochim. Biophys. Acta. 1999; 1444: 85-91
        • Benvenuto G.
        • Carpentieri M.L.
        • Salvatore P.
        • Cindolo L.
        • Bruni C.B.
        • Chiariotti L.
        Mol. Cell. Biol. 1996; 16: 2736-2743
        • Tsuji Y.
        • Moran E.
        • Torti S.V.
        • Torti F.M.
        J. Biol. Chem. 1999; 274: 7501-7507
        • Weinstein L.S.
        • Spiegel A.M.
        • Carter A.D.
        FEBS Lett. 1988; 232: 333-340
        • Weinstein L.S.
        • Kats I.
        • Spiegel A.M.
        • Carter A.D.
        Mol. Endocrinol. 1990; 4: 958-964
        • Kinane T.B.
        • Shang C.
        • Finder J.D.
        • Ercolani L.
        J. Biol. Chem. 1993; 268: 24669-24676
        • Eschenhagen T.
        • Friedrichsen M.
        • Gsell S.
        • Hollmann A.
        • Mittmann C.
        • Schmitz W.
        • Scholz H.
        • Weil J.
        • Weinstein L.S.
        Basic Res. Cardiol. 1996; 91: 41-46
        • Holtzman E.J.
        • Soper B.W.
        • Stow J.L.
        • Ausiello D.A.
        • Ercolani L.
        J. Biol. Chem. 1991; 266: 1763-1771
        • Holtzman E.J.
        • Kinane T.B.
        • West K.
        • Soper B.W.
        • Karga H.
        • Ausiello D.A.
        • Ercolani L.
        J. Biol. Chem. 1993; 268: 3964-3975
        • Kinane T.B.
        • Finder J.D.
        • Kawashima A.
        • Brown D.
        • Abbate M.
        • Shang C.
        • Fredericks W.J.
        • Rauscher III, F.J.
        • Sukhatme V.P.
        • Ercolani L.
        J. Biol. Chem. 1994; 269: 27503-27509
        • Kinane T.B.
        • Finder J.D.
        • Kawashima A.
        • Brown D.
        • Abbate M.
        • Fredericks W.J.
        • Sukhatme V.P.
        • Rauscher III, F.J.
        • Ercolani L.
        J. Biol. Chem. 1995; 270: 30760-30764
        • Morris A.J.
        • Malbon C.C.
        Physiol. Rev. 1999; 79: 1373-1430
        • Liu C.
        • Calogero A.
        • Ragona G.
        • Adamson E.
        • Mercola D.
        Crit. Rev. Oncog. 1996; 7: 101-125
        • Marin M.
        • Karis A.
        • Visser P.
        • Grosveld F.
        • Phillipsen S.
        Cell. 1997; 89: 619-628
        • Dignam J.D.
        • Lebovitz R.M.
        • Roeder R.G.
        Nucleic Acids Res. 1983; 11: 1475-1489
        • Khana-Gupta A.
        • Zibello T.
        • Simkevich C.
        • Rosmarin A.G.
        • Berliner N.
        Blood. 2000; 95: 3734-3741
        • McCaffrey P.G.
        • Newsome D.A.
        • Fibach E.
        • Yoshida M.
        • Su M.S.-S.
        Blood. 1997; 90: 2075-2083
        • Adunyah S.E.
        • Chander R.
        • Barner V.K.
        • Cooper R.S.
        Biochim. Biophys. Acta. 1995; 1263: 123-132
        • Racke F.K.
        • Lewandowska K.
        • Goueli S.
        • Goldfarb A.N.
        J. Biol. Chem. 1997; 272: 23366-23370
        • Ajenjo N.
        • Aaronson D.S.
        • Ceballos E.
        • Richard C.
        • Leon J.
        • Crespo P.
        J. Biol. Chem. 2000; 275: 7189-7197
        • Quandt K.
        • Frech K.
        • Karas H.
        • Wingender E.
        • Werner T.
        Nucleic Acids Res. 1995; 23: 4878-4884
        • Riggs M.G.
        • Whittaker R.G.
        • Neumann J.R.
        • Ingram V.M.
        Nature. 1977; 268: 462-464
        • Kruh J.
        Mol. Cell. Biochem. 1982; 42: 65-82
        • Yoshida M.
        • Kijima M.
        • Akita M.
        • Beppu T.
        J. Biol. Chem. 1990; 265: 17174-17179
        • Brosch G.
        • Ransom R.
        • Lechner T.
        • Walton J.D.
        • Loidl P.
        Plant Cell. 1995; 7: 1941-1950
        • Sowa Y.
        • Orita T.
        • Minamikawa S.
        • Nakano K.
        • Mizuno T.
        • Nomura H.
        • Sakai T.
        Biochem. Biophys. Res. Commun. 1997; 241: 142-150
        • Sowa Y.
        • Orita T.
        • Hiranabe-Minamikawa S.
        • Nakano K.
        • Mizuno T.
        • Nomura H.
        • Sakai T.
        Ann. N. Y. Acad. Sci. 1999; 886: 195-199
        • Doetzlhofer A.
        • Rotheneder H.
        • Lagger G.
        • Koranda M.
        • Kurtev V.
        • Brosch G.
        • Wintersberger E.
        • Seiser C.
        Mol. Cell. Biol. 1999; 19: 5504-5511
        • Rivero J.A.
        • Adunyah S.E.
        Biochem. Biophys. Res. Commun. 1996; 224: 796-801
        • Favata M.F.
        • Horiuchi K.Y.
        • Manos E.J.
        • Daulerio A.J.
        • Stradley D.A.
        • Feeser W.S.
        • Van Dyk D.E.
        • Pitts W.J.
        • Earl R.A.
        • Hobbs F.
        • Copeland R.A.
        • Magolda R.L.
        • Scherle P.A.
        • Trzaskos J.M.
        J. Biol. Chem. 1998; 273: 18623-18632
        • Sale E.M.
        • Atkinson P.G.
        • Sale G.J.
        EMBO J. 1995; 14: 674-684
        • Cuenda A.
        • Rous J.
        • Doza Y.N.
        • Meier R.
        • Cohen P.
        • Gallagher T.F.
        • Young P.R.
        • Lee J.C.
        FEBS Lett. 1995; 364: 229-233
        • Kummer J.L.
        • Rao P.K.
        • Heidenreich K.A.
        J. Biol. Chem. 1997; 272: 20490-20494
        • Assefa Z.
        • Vantieghem A.
        • Declercq W.
        • Vandenabeele P.
        • Vandenheede J.R.
        • Merlevede W.
        • de Witte P.
        • Agostinis P.
        J. Biol. Chem. 1999; 274: 8788-8796
        • Zhang Y.
        • Zhong S.
        • Dong Z.
        • Chen N.
        • Bode A.M.
        • Ma W.
        • Dong Z.
        J. Biol. Chem. 2001; 276: 14572-14580
        • Heerdt B.G.
        • Houston M.A.
        • Augenlicht L.H.
        Cancer Res. 1994; 54: 3288-3294
        • Janson W.
        • Brandner G.
        • Siegel J.
        Oncogene. 1997; 15: 1395-1406
        • Heruth D.P.
        • Zimstein G.W.
        • Bradley J.F.
        • Rothberg P.G.
        J. Biol. Chem. 1993; 268: 20466-20472
        • Cummings J.H.
        Gut. 1981; 22: 763-779
        • Roediger W.E.W.
        Gut. 1980; 21: 793-798
        • Scheppach W.
        • Bartram P.
        • Richter A.
        • Richter F.
        • Liepold H.
        • Dusel G.
        • Hofstetter G.
        • Rüthlein J.
        • Kasper H.
        J. Parenter. Enteral Nutr. 1992; 16: 43-48
        • McBain J.A.
        • Eastman A.
        • Nobel C.S.
        • Mueller G.C.
        Biochem. Pharmacol. 1997; 53: 1357-1368
        • Constantoulakis P.
        • Knitter G.
        • Stamatoyannopoulos G.
        Blood. 1989; 74: 1963-1971
        • Glauber J.G.
        • Wandersee N.J.
        • Little J.A.
        • Ginder G.D.
        Mol. Cell. Biol. 1991; 11: 4690-4697
        • Dover G.J.
        • Humphries R.K.
        • Moore J.G.
        • Ley T.J.
        • Young N.S.
        • Charache S.
        • Nienhuis A.W.
        Blood. 1986; 67: 735-738
        • Roediger W.E.W.
        Lancet. 1980; 2: 712-715
        • Galasinski S.K.
        • Lively T.N.
        • Grebe de Barron A.
        • Goodrich J.A.
        Mol. Cell. Biol. 2000; 20: 1923-1930
        • Dynan W.S.
        • Tjian R.
        Cell. 1983; 32: 669-680
        • Dynan W.S.
        • Tjian R.
        Cell. 1983; 35: 79-87
        • Karlseder J.
        • Rotheneder H.
        • Wintersberger E.
        Mol. Cell. Biol. 1996; 16: 1659-1667
        • Lin S.-Y.
        • Black A.R.
        • Kostic D.
        • Pajovic S.
        • Hoover C.N.
        • Azizkhan J.C.
        Mol. Cell. Biol. 1996; 16: 1668-1675
        • Zwicker J.
        • Liu N.
        • Engeland K.
        • Lucibello F.C.
        • Müller R.
        Science. 1996; 271: 1595-1597
        • Merchiers P.
        • Bulens F.
        • De Vriese A.
        • Collen D.
        • Belayew A.
        FEBS Lett. 1999; 456: 149-154
        • Netzker R.
        • Fabian D.
        • Weigert C.
        • Brand K.A.
        Biochim. Biophys. Acta. 1999; 1444: 231-240
        • Le H.B.
        • Vaisanen P.A.
        • Johnson J.L.
        • Raney A.K.
        • McLachlan A.
        DNA Cell Biol. 1994; 13: 473-485
        • Daniel S.
        • Zhang S.
        • DePaoli-Roach A.A.
        • Kim K.-H.
        J. Biol. Chem. 1996; 271: 14692-14697
        • Zhang W.
        • Geiman D.E.
        • Shields J.M.
        • Dang D.T.
        • Mahatan C.S.
        • Kaestner K.H.
        • Biggs J.R.
        • Kraft A.S.
        • Yang V.W.
        J. Biol. Chem. 2000; 275: 18391-18398
        • Zhang D.E.
        • Hetherington C.J.
        • Tan S.
        • Dziennis S.E.
        • Gonzalez D.A.
        • Chen H.M.
        • Tenen D.G.
        J. Biol. Chem. 1994; 269: 11425-11434
        • Merchant J.L.
        • Shiotani A.
        • Mortensen E.R.
        • Shumaker D.K.
        • Abraczinskas D.R.
        J. Biol. Chem. 1995; 270: 6314-6319
        • Perkins N.D.
        • Edwards N.L.
        • Duckett C.S.
        • Agranoff A.B.
        • Schmid R.M.
        • Nabel G.J.
        EMBO J. 1993; 12: 3551-3558
        • Chen L.I.
        • Nishinaka T.
        • Kwan K.
        • Kitabayashi I.
        • Yokoyama K.
        • Fu Y.-H.F.
        • Grunwald S.
        • Chiu R.
        Mol. Cell. Biol. 1994; 14: 4380-4389
        • Datta P.K.
        • Raychaudhuri P.
        • Bagchi S.
        Mol. Cell. Biol. 1995; 15: 5444-5452
        • Lania L.
        • Majello B.
        • De Luca P.
        Int. J. Biochem. Cell Biol. 1997; 29: 1313-1323
        • Naar A.M.
        • Ryu S.
        • Tjian R.
        Cold Spring Harbor Symp. Quant. Biol. 1998; 63: 189-199
        • Espinos E.
        • Weber M.J.
        Brain Res. Mol. Brain Res. 1998; 56: 118-124
        • Courilleau D.
        • Chastre E.
        • Sabbah M.
        • Redeuilh G.
        • Atfi A.
        • Mester J.
        J. Biol. Chem. 2000; 275: 17344-17348