the journal of biological 267, no. 13, 5, pp. 9300-9306 ... · 1.3cat, using sphi or barnhi linker,...

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 13, Issue of May 5, pp. 9300-9306,1992 Printed in U. S. A.

Identification of Two Enhancer Elements in the Gene Encoding the Type 1 Glucose Transporter from the Mouse Which Are Responsive to Serum, Growth Factor, and Oncogenes*

(Received for publication, October 25, 1991)

Takashi MurakamiS, Toshihiko NishiyamaSQ, Tetsuya ShirotaniSB, Yasuo ShinoharaSlI, Masaharu KanSII, Kazuo Ishiil, Fumihiko Kanail, Shoichi Nakazurul, and Yousuke EbinaS** From the $Department of Enzyme Genetics, Institute for Enzyme Research, YFaculty of Pharmaceutical Science and 11 Department of Urology, The University of Tokushima, Kuramoto-cho, Tokushima 770 and the §Department of Metabolic Medicine, Kumamoto University Medical School, Kumamoto 860, Japan

The type 1 glucose transporter (GLUTl) gene en- codes an integral membrane glycoprotein responsible for facilitating transfer of glucose across plasma membrane and is rapidly activated by serum, growth factors, and by oncogenic transformation. To elucidate the molecular mechanisms of regulation of GLUTl gene expression, we isolated and characterized the mouse GLUTl gene. DNA elements regulating transcription of the gene were analyzed in transient expression assays after transfection of NIH/3T3 cells with a low background chloramphenicol acetyltransferase (CAT) vector system pSVOOCAT. We identified two enhancer elements; the first one is located 2.7 kilobases upstream of the cap site of the gene which contains the homologous sequences with two 12-0- tetradecanoylphorbol-13-acetate-responsive elements (TREs), a serum response element, a cyclic AMP-responsive element (CRE) and three GC boxes, and the second one is located in the second intron of the gene which contains the homologous sequences with two TREs and one CRE. With the promoter alone the transcription of the gene is activated by src, only slightly activated by ras and is not activated by serum and platelet-derived growth factor. When the gene is ac- companied by one of these enhancers, the transcription is activated by all these stimuli.

Oncogenic transformation of mammalian cells is associated with complex changes in the expression of numerous cellular genes( 1,2). To better understand molecular mechanisms gov- erning oncogenic transformation, it is essential to identify genes with products of known function and whose transcrip- tions are altered directly as a consequence of oncogenic transformation. Most tumor cells show increased rates of respira- tion, glucose uptake, and glucose metabolism, as compared to untransformed cells (3). An accelerated rate of glucose trans-

* This work was supported by research grants from the Ministry of Education, Science and Culture of Japan, a grant for Diabetes Research from the Ministry of Health and Welfare, and a grant for diabetes research from Otsuka Pharmaceutical Co., Ltd. (to Y. E.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

** To whom correspondence should be addressed Dept. of Enzyme Genetics, Institute for Enzyme Research, The University of Toku- shima, Kuramoto-cho, Tokushima 770, Japan. Tel.: 0886-31-3111, ex. 2540: Fax: 0886-33-1845.

port is among the most characteristic biochemical markers of the transformed phenotype.

Most animal cells take up glucose by the process which depends on the gradient of glucose concentration not requiring cellular energy (4). The molecules responsible for this process are integral membrane glycoproteins. Recent clonings of cDNAs encoding five different types of glucose transporters have provided the necessary tools for a molecular analysis of the regulation of the gene expression as well as details on the primary structures (5). Among five known types of glucose transporters, the type 1 (GLUT1)’ gene is activated in cells transformed by oncogenes such as hs, src, and ras, as a result of the transcriptional activation of the gene (6,7). This is one of the earliest known effects of oncogenesis on the expression of a gene encoding a protein of well-defined function. The addition of serum, peptide growth factors, and agents which elevate intracellular cyclic AMP (CAMP) in quiescent fibro- blasts also increases GLUTl mRNA levels by a rapid and transient activation of the gene (8, 9). The effect of serum is a direct response not requiring intermediary new protein synthesis (8). Thus, the GLUTl gene is one of the so-called immediate early genes and seems to be an ideal transcription unit to examine mechanisms of signal transductions of oncogene products and growth factors. In the present work, we isolated the mouse GLUTl gene, characterized, and identified two enhancer elements which respond to serum, platelet- derived growth factor (PDGF), and to the oncogenes ras and src. We made use of a low background chloramphenicol acetyltransferase (CAT) assay system pSVOOCAT (10).

EXPERIMENTAL PROCEDURES

Isolation and Characterization of Genomic Clones-The mouse 3T3- L1 cell genomic library was prepared as follows. A partial Sau3AI digest of DNA from 3T3-Ll cells was treated with calf intestine phosphatase and ligated with a BamHI digest of bacteriophage X EMBL4. After packaging in vitro, recombinant phages that carried 3T3-Ll DNA inserts were selected by use of the restrictive host strain, Escherichia coli NM539. At first, this library was screened with oligonucleotides corresponding to the 5’ sequence of rat GLUTl cDNA (11) and then with DNA fragments derived from some of the isolated genomic clones which are free of repetitive sequences. Phage DNAs of positive clones were characterized by restriction mapping and by Southern blot hybridization analysis (12). Subclones were constructed into plasmid pUC19 as a vector, and the nucleotide

The abbreviations used are: GLUT1, type 1 glucose transporter; CAT, chloramphenicol acetyltransferase; kb, kilobase pairs; bp, base pairs; TPA, 12-0-tetradecanoylphorbol-13-acetate; TRE, TPA-responsive element; SRE, serum response element; CRE, cyclic AMP- responsive element; PDGF, platelet-derived growth factor; PIPES, piperazine-N,N”bis(2-ethanesulfonic acid).

9300

Enhancers in the GLUT 1 Gene from the Mouse 9301

sequences were determined by the dideoxynucleotide chain-termination method (13, 14), using synthetic oligonucleotide primers complementary to the vector or mouse cDNA sequence (15).

Analysis of the 5' End of GLUTl rnRNA-Total and poly(A)+ RNA were prepared from 3T3-Ll cells by the guanidine thiocyanate-CsC1 method (16, 17), followed by oligo(dT)-cellulose chromatography.

For the S1 nuclease assay, we used the internally labeled single strand DNA which covered the PstI-BarnHI site (-190 to +158). This single strand DNA was generated as follows. A synthesized and purified 18-residue oligonucleotide complementary to nucleotides +141 to +158 was annealed to the alkali-denatured plasmid containing the mouse GLUTl promoter, then the internally labeled DNA was synthesized by the Klenow fragment of E. coli DNA polymerase I, using as substrates [LX-~'P]~ATP and three unlabeled dNTPs. The DNA product was digested with PstI, and the single strand DNA probe was purified by polyacrylamide gel electrophoresis under de- naturing conditions. Hybridization was carried out with 0.006 pmol of probe (1 X lo6 dpm) and 70 pg of total RNA, 1.5 pg of poly(A)+ RNA, or 70 pg of yeast tRNA (as a control) a t 58 "C for 16 h in 80% formamide, 40 mM PIPES, pH 6.4, 1 mM EDTA, 0.4 M NaC1, and the mixture was subjected to nuclease S1 digestion (17), performed at 30 "C with 500 units of the nuclease (Boehringer Mannheim) for 90 min.

For primer extension analysis, synthesized and purified 30-residue oligonucleotide complementary to nucleotides +58 to +87 was end- labeled with [y3'P]ATP and T4 polynucleotide kinase. Annealing was carried out with 0.1 pmol of primer (9 X 10' dpm) and 3 pg of poly(A)+ RNA or yeast tRNA (as a control) at 32 "C for 12 h in 80% formamide, 40 mM PIPES, pH 6.4, 1 mM EDTA, 0.4 M NaC1, and after ethanol precipitation the mixture was subjected to reverse transcription (17). To directly compare the products with the genomic nucleotide sequence, the sequences were obtained by the dideoxy nucleotide chain-termination technique using the same primer as used for making single strand DNA probe (for S1 nuclease assay) or that used for primer extension analysis, respectively. Portions of samples were electrophoresed in a 6% acrylamide, 7 M urea gel.

Construction of Plasmids-The basal promoterless CAT plasmid pSV002CAT, which is constructed from pSVOOCAT (lo), was a kind gift from Dr. E. Araki (Kumamoto University). This plasmid has multicloning sites (5'-EcoRV-BglII-KpnI-SphI-XhoI-HindIII-NruI- Sad-3') at the 5' of the CAT gene. The TaqI site of the mouse GLUTl gene, positioned at +137 relative to the transcription start site, was converted to the HindIII site using the Klenow fragment and the HindIII linker. The plasmid pGT1-1.3CAT was constructed by ligating SphI (-1.3 kilobases (kb))-HindIII(Taq1) (+137) fragment of mouse GLUTl gene to SphI-Hind111 double-digestedpSVOO2CAT.

The pGT1-1.3CAT/SBb or pGT1-1.3CAT/BXb was constructed as follows. The SphI (-3.3 kb)-BbeI (-2.7 kb) fragment or the BamHI (+16.7 kb)-XbaI (+18.0 kb) fragment was inserted into the SphI site or the BglII site at the 5' site of the GLUTl promoter of pGT1- 1.3CAT, using SphI or BarnHI linker, respectively. To make the pGT1-1.3CAT/BXb/SBb, the BarnHI-XbaI fragment was inserted into the BglII site of the pGT1-1.3 CAT/SBb, using the BamHI linker.

To prepare CAT constructs with deletion mutants from 5' or 3' of these enhancer elements, some polymerase chain reaction (18) products were made by synthesized oligonucleotides corresponding to the sequence of the enhancer. These products or several restriction enzyme fragments were inserted into the same sites as that of the original plasmids (/SBb or/BXb), using SphI (/SBb), BarnHI, or BglII linker (/BXb). Sequences of the polymerase chain reaction products were determined by the dideoxynucleotide chain-termination method, using synthetic oligonucleotide primers. Fragments were inserted in the same direction relative to the CAT transcription unit.

ptkCAT/SBb or ptkCAT/BXb was constructed as follows. The SphI (-3.3 kb)-BbeI (-2.7 kb) fragment or the BarnHI (+16.7 kb)- XbaI (+18.0 kb) fragment was inserted into the BarnHI site of a plasmid ptkCAT (19), using the BarnHI linker.

The plasmid pSRav-src was constructed as follows. The plasmid pSRA-2 (20), in which the whole genome of the avian sarcoma virus (SRA-2) was subcloned, was a kind gift from Dr. J. M. Bishop (University of California, San Francisco). The 1.9-kb NcoI-EcoRI fragment of pSRA-2 which contains the entire coding region of the v-src gene was inserted into the unique EcoRI site of the expression plasmid pcDL-SRa296 (21; a kind gift from Dr. Y. Takebe, DNAX Research Institute, Palo Alto), using a EcoRI linker.

AlolacZ was constructed as follows. The BglII-Hind111 fragment of the plasmid AloCAT,(22), which contains the SV40 early promoter,

was inserted into the plasmid pAc-lacZ (23, kindly provided by Dr. J. Miyazaki, University of Tokyo), in place of the XhoI-Hind111 chicken p-actin promoter.

The Dlasmids were DreDared from E. coli MC1061 RecA- bv the alkaliisodium dodecyl 'suffate method and purified by two sets of centrifugation to equilibrium in ethidium bromide-CsC1 gradients (17). Purities of the plasmids were checked by agarose gel electrophoresis.

Cell Culture and Transient Expression Assays-To test the enhancer activities, the mouse fibroblast cell line NIH/3T3 were grown in Dulbecco's modified Eagle's medium supplemented with 10% calf serum. 7 X lo6 cells were plated in 90 mm-diameter culture dishes 48 h before transfection. Transfections were carried out by the calcium phosphate precipitation method (24), using a total of 13 pg of DNA mixture containing 10 pg of the CAT gene recombinant plasmid and 3 pg of an internal standard plasmid pAc-lacZ bearing the P-galactosidase gene under the control of the chicken @-actin gene promoter (23). Four h later the cells were shocked for 2 min with 12.5% glycerol and were harvested 48 h after the transfection. Cells were broken by three cycles of freezing and thawing, followed by sonication. The cell lysates were assayed for 0-galactosidase activities and CAT activities, as described elsewhere (25). The count of each spot of the CAT assays was detected using a Bio-image-analyzer BAS-2000 (Fuji Film insti- tution), and CAT activity was normalized for transfection efficiency, dividing CAT activity by @-galactosidase activity.

To test serum or PDGF inducibilities, 4 X lo6 NIH/3T3 cells were plated in 60-mm diameter culture dishes 48 h before transfection. Transfection was carried out by lipofection (26). Twenty pg of DNA mixture containing 16 pg of DNA of the CAT gene recombinant plasmid and 4 pg of an internal standard plasmid A&cZ were diluted to 100 gl in water, and 80 pg of Lipofectin reagent (purchased from Bethesda Research Laboratories) was diluted to 100 pl in water, respectively. We combined 100 p1 of each of the DNA and the Lipofectin reagent dilutions in a 15-ml polystyrene tube, mixed them gently, and left them to stand for 15 min at room temperature. Next, 6.5 ml of serum-free Dulbecco's modified Eagle's medium was added, and the preparation was mixed. Three ml each of the medium containing DNA-Lipofectin complex was added to each of the two dishes of cells, which had been previously washed twice with serum-free medium. The cells were incubated for 7 h, then 1 ml of the medium supplemented with 2% calf serum was added (final 0.5% calf serum). This preparation was incubated for 10 h and the medium was replaced with that supplemented with 0.5% calf serum, then incubation was allowed to proceed for 38 h. Cells of one dish were scraped off and cells of the other were treated with calf serum (final concentration was 15%) or PDGF (purchased from Takara Shuzo; final concentration was 4 units/ml) for 10 h and then scraped off. Cell lysates were assayed for protein concentration, and the same amounts of the protein of each cell lysate, about 100 pg, were assayed for P-galactosidase and CAT activities. The fold induction of the @-galactosidase activities ranges from 0.8 to 1.2 (data not shown).

For the activated ras or v-src cotransfection assay, we used pcDSRac-Ha-ras'""'(27) or pSRav-src, respectively. NIH/3T3 cells (4 X lo5 cells) were plated in 60-mm diameter culture dishes 48 h before transfection. Transfection were carried out by the calcium phosphate precipitation method, using a total of 15.5 pg of DNA mixture containing 4 pg of the CAT gene recombinant plasmid, 1.5 pg of an internal standard plasmid AlolacZ, and 10 pg of pcDSRac- Ha-ra.sVa112, pSRav-src, or pUC18. Cells were harvested 48 h after the transfection. Cell lysates were assayed for protein concentration and same amounts of the protein of each cell lysates were assayed for @- galactosidase and CAT activities. The fold induction of the P-galactosidase activities ranges from 0.6 to 1.8 (data not shown).

RESULTS

Isolation and Characterization of the Mouse GLUTl Gene- We constructed the 3T3-Ll cell-EMBL4 genomic library and cloned the mouse GLUTl gene as described under "Experi- mental Procedures." Eight independent clones were isolated and analyzed by restriction enzyme digestion. These clones overlapped and spanned about 48 kb (Fig. 1).

To define the positions and boundaries of the exon blocks, subclones were constructed into plasmid pUC19, as a vector, and the nucleotide sequences were determined by the dideoxynucleotide chain-termination method using synthetic oli-

9302 Enhancers in the GLUT 1 Gene from the Mouse

FIG. 1. Physical map of the mouse GLUTl gene. The scale is indicated by the 1-kb bar in the lower right hand

cleases BamHI ( B ) , EcoRI ( E ) , and corner. Sites for the restriction endonu-

Hind111 (H) are noted. The closed burs indicate coding sequences and the open bars the transcribed, noncoding regions. The genomic DNA fragments contained in the phage clones are shown at the bottom.

-1080 -960 -840

-600 -720

-480 -360 -240 -120

1 121 241

E E E E E H E E E H E B E H E E W E E H E B E E E H H E

Exon 1 2

U I

H Ikb

~ A G A G ~ ~ A C A C A G C G A G ~ G C C C C ~ A A ~ G G G T G C C C ~ ~ C C ~ ~ T I T T ~ A T T ~ A G C ~ ~ ~ ~ ~ ~ C C ~ C G ~ C C T C C C ~ C T T C C C ~ ~ C CACGCCTCTCAGAGTT

lHEmM"" V C ~ C T I T G G G C C C A C C T A C A C C C C A W G G C G G T C ~ A G C T C C G C G C G C T C T T C C C C TITI

G C A C A G C T A G A G C T I T G G G C cccGCTITcTcAcGcAGCcGcc

FIG. 2. Nucleotide sequence of the mouse GLUTl gene promoter. The two major transcription initiation sites predicted by S1 nuclease mapping and primer-extension analysis are shown by arrowheads. The A, positioned 151 bases upstream from the initiation codon of the translation, is numbered +I (see "Results"). GC boxes, potential APP-binding site, TRE, and CCAAT and TATA boxes are underlined. The coding region of the first exon is boxed.

gonucleotide primers complementary to the vector or mouse GLUTl cDNA sequence (5). The results are shown in Fig. 1. The 28-kb long gene is divided into 10 exons. All the exon borders lie in a position identical to that in the rat (28) and human (29) GLUTl genes. The 10 exons range from 96 base pairs (bp) (exon 2) to 1083 bp (exon lo), and the nine introns range from 93 bp (intron 5) to 12 kb (intron 2). All of the splice donor and acceptor sites conformed to the GT/AG rule (30) for nucleotides immediately flanking exon borders (data not shown). The sequence of the mRNA predicted from that of the gene differs from the cDNA sequence (5) in several instances: amino acid 52 is Tyr (TAC) rather than Ile (ATC), amino acids 193 and 195 are Ile (ATC) rather than Val (GTC), amino acid 403 is Ala (GCT) rather than Arg (CGT). The gene sequence encoding the 3"untranslated region of the cDNA also contains two different sequences (data not shown). Whether these differences reflect sequence polymorphisms or are due to cloning artifacts remains unknown.

Characterization of the 5' End of the GLUTl Gene-DNA sequences around the 5' end of the GLUTl gene are shown in Fig. 2. The 5' end of the mRNA was determined by nuclease S1 mapping and primer extension (Fig. 3). The internally labeled single noncoding strand DNA which covered the sequence from seven bases downstream to 341 bases upstream from the initiation codon of the translation was used as the probe for the S1 mapping. The primer labeled at a position 65 bases upstream from the initiation codon of the translation was used in primer extension analysis. The two distinct bands positioned at A and G, 151 and 149 bases upstream from the initiation codon of the translation, respectively, were detected in both S1 mapping and primer extension analysis (see arrowheads in Figs. 2 and 3). There were also several bands seen in the S1 mapping analysis. However, these bands could not be detected clearly in the primer extension analysis (Fig. 3). The one which positioned at A, 151 bases upstream from the initiation codon of the translation, had the most intensity. Since the generally preferred cap site is residue A preceded by C (30), the A, one of the major cap sites, is numbered +l.

Examination of the GLUTl gene promoter sequence re- vealed elements that may play a role in transcription. Around the major transcription initiation site, there are several different types of transcriptional control elements, including a

TATAA box, a CCAAT box, a potential AP2-binding site (31), a 12-0-tetradecanoylphorbol-13-acetate (TPA)-responsive element (TRE) (32), and two GC boxes (33). In this promoter region there are no homologous sequences with the serum response element (SRE) (34).

Two Enhancer elements of the GLUTl Gene-To ascertain whether the mouse GLUTl promoter is sufficient for inductions by serum growth factors or by oncogenic transformation, we constructed a plasmid pGT1-1.3CAT which contains the mouse GLUTl gene 5'-flanking sequence, -1.3 kb to +137 (bp), linked to the CAT gene. This pGT1-1.3CAT was transfected into NIH/3T3 cells and 15% calf serum was added after serum starvation. Before (for the control) or 10 h after serum stimulation, the cells of each dish were scraped off and CAT activities were determined. As the promoter region was not sufficient for serum induction (Fig. 4B), we examined various fragments of the mouse GLUTl gene for possible enhancer activities.

We inserted various BglII, BamHI, and SphI fragments of the mouse GLUTl gene into the BgZII or SphI site in the 5' position to the GLUTl promoter of pGT1-1.3CAT. Frag- ments were inserted in the same direction relative to the CAT transcription unit. These fragments span from -14 kb to +31 kb of the transcription start site. We obtained evidence for two enhancer elements; one is in the 5' upstream region and the other is in the second intron. After narrowing the enhancer regions, we ascertained that the SphI (-3.3 kb)-BbeI (-2.7 kb) fragment caused about an 8-fold increase in CAT activity (pGT1-1.3CAT/SBb), and the BamHI (+16.7 kb)- XbaI (+18.0 kb) fragment in the second intron caused about a 14-fold increase (pGT1-1.3CAT/BXb) (Fig. 4B). By transfection of the plasmid containing both enhancer elements, the CAT activity was elevated about 23-fold (pGTl-l.SCAT/ BXb/SBb). These enhancers are also active when inserted in the opposite direction relative to the CAT transcription unit and in the cell lines, Balb/c 3T3 and Hep G2 (data not shown). To examine whether the two enhancers can activate transcription from the heterologous promoter, each of the two enhancers was inserted into the BamHI site of a plasmid ptkCAT (19), in which the CAT gene is under the control of the herpes simplex virus thymidine kinase gene (ptkCAT/ SBb and /BXb). By transient transfection into NIH/3T3


A B

Y .s

FIG. 3. Determination of the 5‘ end of the mouse GLUTl mRNA by nuclease S1 mapping ( A ) and primer extension (B) . Total and poly(A)’ RNAs from 3T3-Ll cells were analyzed as described under “Experimental Procedures.” A, an internally labeled single noncoding strand DNA which covered PstI-BornHI site (-190 to +158) was used for nuclease S1 mapping. E , 30-residue oligonucleotide complementary to nucleotides +58 to +87 was used for primer extension. To directly compare the products with the genomic nucleotide sequence, the sequence were obtained by dideoxy chain- termination technique, using the same primer as used for making the single strand DNA probe ( A ) or that used for primer extension analysis ( E ) . Arrowheads point to two bands detected in both S1 mapping and primer extension analysis.

cells, each of the enhancers also caused increases in CAT activity, as compared with the enhancerless construct (Fig. 4C). Other regions in the mouse GLUTl gene could not significantly enhance the CAT activities (data not shown).

Serum and PDGF Activates GLUTl Enhancers-To ascertain whether these two enhancer elements possess growth factor-inducible potential, we transfected these four constructs (pGT-1.3CAT, pGT-1.3CAT/SBb, /BXb, or/BXb/ SBb) into NIH/3T3 cells. Following serum starvation, cells were exposed to serum or PDGF, and the CAT activities were measured (Fig. 4B). In cells transfected with pGT-l.SCAT, CAT activities remained unchanged after these stimulations, but in cells transfected with the CAT construct containing one or both of the enhancers, CAT activities were elevated significantly after serum or PDGF stimulation (about 3- or 2- fold with /SBb, about 6- or 4-fold with /BXb and about 3- fold with /BXb/SBb, respectively). We concluded that serum and PDGF activate transcription of the mouse GLUTl gene through these two enhancers. However, there was no evidence for an additive effect on CAT activity by these two enhancer elements, as determined in experiments of serum and PDGF induction after transfection of pGT-1.3CAT/BXb/SBb plasmid (see “Discussion”).

Activated ras and v-src Actiuates GLUT1 Enhancers-In most tumor cells there are increased rates of glucose uptake, as compared to events in untransformed cells (3). This transformed phenotype was explained by the finding that oncogene products stimulate GLUTl gene transcription (6, 7, 35). To determine the manner in which oncogene products stimulate the transcription, we cotransfected the CAT constructs with the activated ras expression plasmid (pcDSRac-Ha-rasva’12)

(27), v-src expression plasmid (pSRav-src) or pUCl8, as a control (Fig. 4B). The activated ras activated GLUTl gene expression mainly through two enhancer elements. The CAT construct without both of the two enhancers was activated only weakly. The v-src activated GLUTl gene expression through its promoter. To determine the distinct effect of v- src on these two enhancers, we used the promoter of the thymidine kinase gene of the herpes simplex virus which is not activated by v-src (see Fig. 4C). With one of the enhancers, the thymidine kinase promoter was activated by v-src (ptkCAT/SBb or /BXb) (Fig. 4C). Therefore, we concluded that the full signal of v-src is transmitted to these enhancers. These results suggest that the accelerated rate of glucose transport of transformed cells by activated ras and v-src is due to activation of the promoter and two enhancers of the GLUTl gene by the activated oncogenes. We obtained no evidence for an additive effect of both enhancer elements (see “Discussion”).

Sequences of GLUT1 Enhancers-We then determined the nucleotide sequences of two enhancer elements. Deletion mutants from 5’ or 3’ of these elements, obtained using restriction enzyme sites or polymerase chain reaction methods, were inserted into the same sites as that of the original plasmids (/SBb or /BXb). CAT assays were done after transfection into NIH/3T3 cells to examine the levels of enhancer activity of the truncated enhancers. The results of the CAT assays were determined by the relative CAT activities to that obtained with pGT-1.3CAT (Fig. 5).

The important region of the enhancer element (enhancer 1) of the 5’ region of GLUTl gene has three transcriptional element-like sequences, two TREs (one is in an antisense strand) (32) and SRE (34) which exists in the center of 16-bp dyad symmetrical sequence. When both of these three elements were deleted from enhancer 1, the enhancer lost the activity (Fig. 5A). One cyclic AMP-responsive element (CRE)-like sequence (36) and three GC boxes (33) were also found in enhancer 1. The enhancer element in the second intron (enhancer 2) has two TRE-like sequences. When one of the TRE-like sequences was deleted, the enhancer activity was dramatically reduced (Fig. 58). These TRE-like sequences seem to play important roles in activity of enhancer 2. There was also one CRE-like sequence in enhancer 2 with two nucleotide mismatches with the CRE consensus sequence, TGACGTCA (36). The CRE-like sequences in these two enhancers do not play an important role in the enhancer activities because CAT activities were not significantly reduced by deleting these CRE-like sequences.

DISCUSSION

We have identified two enhancer elements of the mouse GLUTl gene; one is in the 5‘ region (enhancer 1) and the other is in the second intron (enhancer 2). These two enhancers function additively in elevating transcriptional activity of the gene, under the noninduced condition (Fig. 4 8 ) . However, these two enhancers did not show this additive effect in the CAT assay system during the induction by growth factors and oncogenes (Fig. 4B). The relative position and distance of these enhancers in the GLUTl gene seem to be important for the additive role of these enhancers in the case of GLUTl gene activation by these stimuli.

There are sequences which have homologies to important transcriptional control elements, one SRE, two TREs, one CRE, and three GC boxes in enhancer 1, and one CRE and two TREs in enhancer 2. The transcription of the mouse GLUTl gene is induced by growth factors and oncogenic transformation. This process does not require intermediary


A

B

Exon 1 2 3 4 5 6 7 8 9 10

U I I1 BBBEEHB E E H E B E H B E W E E H B B E B B H H E

. . . . , .

Enhancer 1

+ 16.7k mi:: BamHI

Enhancer 2

pGT1-1.3CAT " 1 1 1 2 12 (O.8-1.0)[41 (0.9-1.3)[3] (1.1-2.4)[5] (3.1-18.9)[3]

pGT1-1.3CAT lSBb CAT^-^ (6.0-1 1.4)[6] (1.3-3.9)[4] (1.3-2.9)[3] (2.3-46.2)[5] (3.6-8.0)[3]

3 2 18 6 Enhancer 1

POT1 -1 3CAT , , 23 " (12.1-36.6)[4] (1.4-4.1)[4] (1.2-4.5)[3] (6.2-50.0)[5] (0.8-3.1)[3]

3 3 20 2 IBXblSBb

Enhancer 2 Enhancer 1

C Enhancer Actlvlty

plkCAT .. Pr CAT 1 I.,.

1

ptkCAT1SBb I I ..I. 7

(4.3-9.2)[2]

ptkCATIBXb 5

(3.9-5.5)[2]

lnductlon V - O M

(Fold Increase)

1 (0.2-1.2)[3]

(3.6-6.6)[3] 5

(2.9-67.8)[3] 25

FIG. 4. CAT assay of the two inducible enhancer elements of the mouse GLUTl gene. A, the locations of the two enhancer elements of the GLUTl gene. The hatched and cross-hatched boxes represent two enhancer elements (Enhancer 1 and Enhancer Z ) , and the solid box represents the GLUTl promoter used in the following experiments. B and C, structures of seven CAT constructs alld results of transient expression analyses. One or both of the enhancer elements were inserted into the plasmid pGT1-1.3CAT, bearing the mouse GLUTl gene 5' region from -1.3 kb to +137 bp, or ptkCAT, bearing the herpes simplex virus thymidine kinase gene promoter. Enhancer activities are expressed relative to the CAT activity obtained with pGT-1.3CAT, or ptkCAT. The other CAT activities are expressed as fold increases of inductions, after addition of serum or PDGF, or the cotransfection experiment with pcDSRac-Ha-ras'"'12 or pSRav-src relative to that with pUC18. Each entry represents the average from indeDendent exDeriments. the numbers in brackets mean the number of the independent experiments, and those in parenthesis give the extremes.

new protein synthesis (8). The c-fos proto-oncogene is also a member of the cellular immediate early gene family, and the transcriptional regulation of the gene is well characterized (37). The c-fos gene is activated in a protein-synthesis independent manner by these stimulations, and this activation occurs mainly through a SRE in the 5' region of the gene (34). The core region of SRE is a 10-bp sequence which consists of CC (A/T)G GG and locates in the center of a dyad symmetrical element. At least two factors, SRF and p62-TCF, bind to this element and mediate transcriptional activation by serum growth factors (38). The SRE homologue in the GLUTl upstream enhancer (enhancer 1) exists in the center of a 16-bp complete dyad symmetrical element. The sequences of two dyad symmetrical elements of c-fos and GLUTl gene show little resemblance. The 16-bp dyad symmetry in the GLUTl gene enhancer 1 is perfect to the extent that we think trans-acting factors such as SRF bind to this element and mediate transcriptional activation by growth factors. It is known that TRE mediates transcriptional activation in response to serum, TPA, and oncogenes in the manner of both dependence on and independence of the protein-synthesis (32, 39-43). Therefore, the TRE homologues in the two GLUTl enhancers may also contribute to transcriptional activation by serum. However, these three elements (one SRE and two

TREs) are not sufficient for full activity of the enhancer 1 (Fig. 5A). The other cis-element in the enhancer 1 seems to contribute to full activity of enhancer 1.

There is evidence that the tumor suppressor gene p53 plays an important role in regulating cell cycles especially from Go or GI to S phase (44). P53 is also a sequence-specific DNA- binding protein (45) and enhances transcription from a mus- cle-specific creatine kinase gene (46). The transcription of the GLUTl gene is activated during transition from Go to S (8, 9). There are several core sequences of the p53-binding site (TGCCT) (45) in GLUTl two enhancers (Fig. 5). Especially in enhancer 1, the SRE homologue is separated by two sets of core sequence of p53-binding site (one is in the antisense strand). The SRE-binding factors and p53 may bind co- operatively or competitively around the SRE homologue and may regulate GLUTl gene expression, through enhancer 1.

v-src activates the mouse GLUTl promoter. The promoter has a sequence motif that matches consensus sequence of the transcription factor AP2-binding site, CCCCAGGC (31). The AP2 mediates transcriptional activation in response to both TPA and CAMP (31). As v-src uses a protein kinase C- mediated pathway to induce the expression of a transformation-related 9E3 gene(47), one of the src activation sites in the promoter may be the AP2-like element. However, the two


T c n n ; c c C T G T C C T G T A r r C A G T A G T A ~ ~ C ~ A O C ~ T G ~ C O C ~ ~ ~ ~ T A C ~ ~ ~ ~ " ~ C ~ C m C C A ~ ~ C " A T

T C C C n ; A A A G G ~ T C A ~ G A T G ~ T G T T A C T G T C W C C ~ ~ T A ~ATCC'IGAGGGCTGAGGWG """ GAGGAAGAAGACCAATWTAGA 18kb 1 3 4

14 C

FIG. 5. Nucleotide sequences of the enhancer elements in the upstream region (A) and in the second intron ( B ) of the mouse G L U T l gene. The CAT activities of the transient expression analyses with deletion mutants of the enhancer elements are expressed relative to the activity of pGT1-1.3CAT. The arrows indicate the deletion sites. GC boxes, potential SRE, TREs, and CREs are underlined by the solid line. Core sequences of p53- binding sites are underlined with a dashed line. A complete dyad symmetrical element situated around the SRE homologue is shown by the broken line. The CAT activities in the parentheses are those of the deletion plasmids contained from the XhoI site in the second stair ( B ) .

enhancers are essential for the full activation of the GLUTl gene by the v-src product. The GLUTl gene promoter itself was not sufficient for serum induction (Fig. 4B). One copy of AP2-binding site may not be enough to induce transcriptional activity of the GLUTl gene by the stimulus.

Acknowledgments-We thank Drs. J. Miyazaki, E. Araki, K. Kai- buchi, H. Kondoh, and Y. Takebe for generous gifts of materials and valuable advice, Drs. T. Ishizuka (Chiba University), M. Takiguchi (Kumamoto University), S. Ohno, and S. Hirai (Tokyo Metropolitan Institute of Medical Science) for useful discussions, Dr. N. Hashimoto (Chiba University), M. Kondoh, M. Nakamura, and M. Kusumi for technical assistance, and M. Ohara for comments.

REFERENCES

1. Schutzbank, T., Robinson, R., Oren, M., and Levine, A. J. (1982)

2. Scott, M. R. D., Westphal, K.-H., and Rigby, P. W. J. (1983) Cell

3. Hatanaka, M. (1974) Biochem. Biophys. Acta 355, 77-104 4. Simpson, I. A., and Cushman, S. W. (1986) Annu. Reu. Biochem.

5. Bell, G. I., Kayano, T., Buse, J. B., Burant, C. F., Takeda, J., Lin, D., Fukumoto, H., and Seino, S. (1990) Diabetes Care 13,198- 208

6. Flier, J. S., Mueckler, M. M., Usher, P., and Lodish, H. F. (1987) Science 235, 1492-1495

7. Birnbaum, M. J., Haspel, H. C., and Rosen, 0. M. (1987) Science

8. Hiraki, Y., Rosen, 0. M., and Birnbaum, M. J. (1988) J. Biol.

9. Hiraki, Y., McMorrow, 1. M., and Birnbaum, M. J. (1989) Mol.

10. Araki, E., Shimada, F., Shichiri, M., Mori, M., and Ebina, Y.

11. Birnbaum, M. J., Haspel, H. C., and Rosen, 0. M. (1986) Proc.

12. Southern, E. M. (1975) J. Mol. Biol. 98,503-517 13. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl.

Cell 30,481-490

34,557-567

55, 1059-1089

235, 1495-1498

Chem. 263,13655-13662

Endocrinol. 3, 1470-1476

(1988) Nucleic Acids Res. 1 6 , 1627

Natl. Acad. Sci. U. S. A. 83,5784-5788

Acad. Sci. U. S. A. 74, 5463-5467

14. Hattori, M., and Sakaki, Y. (1986) Anal. Biochem. 152,232-238 15. Kaestner, K. H., Christy, R. J., McLenithan, J. C., Braiterman,

L. T., Cornelius, P., Pekala. P. H., and Lane, M. D. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 3150-3154

16. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J., and Rutter,

17. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

18. Mullis, K. B., and Faloona, F. A. (1987) Methods Enzymol. 155,

19. Hayashi, S., Goto, K., Okada, T. S., and Kondoh, H. (1987) Genes & Deu. 1,818-828

20. DeLorbe, W. J., Luciw, P. A., Goodman, H. M., Varmus, H. E., and Bishop, J. M. (1980) J. Virol. 36, 50-61

21. Takebe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yoshida, M., and Arai, N. (1988) Mol. Cell. Biol. 8, 466-472

22. Laimins, L. A., Gruss, P., Pozzatti, R., and Khoury, G . (1984) J. Virol. 49, 183-189

23. Miyazaki, J., Takaki, S., Araki, K., Tashiro, F., Tominaga, A., Takatsu, K. & Yamamura, K. (1989) Gene (Amst.) 79, 269- 277

W. J. (1979) Biochemistry 18, 5294-5299

335-350

24. Graham, F. L., and Van der Eb, A. (1973) J. Virol. 52, 456-467 25. Murakami, T., Nishiyori, A., Takiguchi, M., and Mori, M. (1990)

Mol. Cell. Biol. 1 0 , 1180-1191 26. Felgner, P. L., Gadek, T. R., Holm, M., Roman, R., Chan, H. W.,

Wenz, M., Northrop, J. P., Ringold, G. M., and Danielsen, M. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 7413-7417

27. Fukumoto, Y., Kaibuchi, K., Oku, N., Hori, Y., and Takai, Y. (1990) J. Biol. Chem. 265, 774-780

28. Williams, S. A., and Birnbaum, M. J. (1988) J. Biol. Chem. 263,

29. Fukumoto, H., Seino, S., Imura, H., Seino, Y., and Bell, G. I.

30. Breathnach, R., and Chambon, P. (1981) Annu. Reu. Biochem.

31. Imagawa, M., Chiu, R., and Karin, M. (1987) Cell 51, 251-260 32. Lee, W., Mitchell, P., and Tjian, R. (1987) Cell 49, 741-752 33. Kadonaga, J. T., Carner, K. R., Masiarz, F. R., and Tjian, R.

34. Treisman, R. (1986) Cell 46, 567-574 35. Hiraki, Y., Garcia de Herreros, A., and Birnbaum, M. J. (1989)

19513-19518

(1988) Diabetes 37, 657-661

50,349-383

(1987) Cell 51, 1079-1090

Proc. Natl. Acad. Sci. U. S. A. 8 6 , 8252-8256

9306 Enhancers in the GLUT 1 Gene from the Mouse 36. Montminy, M. R., Sevarino, K. A., Wagner, J . A,, Mandel, G., 42. Boyle, W. J., Smeal, T., Defize, L. H. K., Angel, P., Woodgett, J.

37. Wagner, B. J., Hayes, T. E., Hoban, C. J., and Cochran, B. H. 43. Binktruy, B., Smeal, T., and Karin, M. (1991) Nature 351, 122-

127

38. Schroter, H., Mueller, C. G. F., Meese, K., and Nordheim, A. 44. Martinez, J., Georgoff, I., Martinez, J., and Levine, A. J. (1991)

Genes & Deu. 5, 151-159 39. Piette, J., Hirai, s.-I., and Yaniv, M. (1988) Natl, Acad, sei. 45. Kern, S. E., Kinzler, K. W., Bruskin, A., Jarosz, D., Friedman,

P., Prives, C., and Vogelstein, B. (1991) Science 252, 1708- 40. Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R. J., Rahms- 1711

dorf, H. J., Jonat, c. , Herrlich, p., and Karin, M. (1987) Cell 46. Weintraub, H., Hauschka, s., and Ta~scot t , s. J. (1991) PrOC.

41. Schonthal, A., Herrlich, P., Rahmsdorf, H. J., and Ponta, H. 47. Spangler, R., Joseph, C., Qureshi, S. A., Berg, K. L., and Foster, (1988) Cell 64, 325-334 D. A. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 7017-7021

and Goodman, R. H. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, R., Karin, M., and Hunter, T. (1991) Cell 64, 573-584 6682-6686

(1990) EMBO J. 9, 4477-4484

(1990) EMBO J. 9, 1123-1130

U. S. A. 85, 3401-3405

49,729-739 Natl. Acad. Sci. U. S. A . 88, 4570-4571

the journal of biological 267, no. 13, 5, pp. 9300-9306 ... · 1.3cat, using sphi or barnhi linker,...

Documents