Regulation of elongation factor-1alpha expression by growth factors and anti-receptor blocking antibodies.

The epidermal growth factor (EGF) family and its receptors regulate normal and cancerous epithelial cell proliferation, a process that could be suppressed by anti-receptor blocking antibodies. Polypeptide elongation factor-1alpha (EF-1alpha) is a multifunctional protein whose levels are positively correlated with the proliferative state of cells. To identify genes, whose expression may be modulated by anti-receptor blocking antibodies, we performed a differential display screening and isolated differentially expressed cDNAs. Isolates from one clone were 100% identical to human EF-1alpha. Both EGF and heregulin-beta1 (HRG) induced EF-1alpha promoter activity and mRNA and protein expression. Growth factor-mediated EF-1alpha expression was effectively blocked by pretreatment with humanized anti-EGF receptor antibody C225 or anti-human epidermal growth factor receptor-2 (HER2) antibody herceptin. Mutants and pharmacological inhibitors of p38(MAPK) and MEK, but not phosphatidylinositol 3-kinase, suppressed both constitutive and HRG-induced stimulation of EF-1alpha promoter activity in MCF-7 cells. Deletion analysis of the promoter suggested the requirement of the -393 to -204 region for growth factor-mediated transcription of EF-1alpha. Fine mapping and point mutation studies revealed a role of the SP1 site in the observed HRG-mediated regulation of the EF-1alpha promoter. In addition, we also provide new evidence to suggest that HRG stimulation of the EF-1alpha promoter involves increased physical interactions with acetylated histone H3 and histone H4. These results suggest that regulation of EF-1alpha expression by extracellular signals that function through human EGF receptor family members that are widely deregulated in human cancers and that growth factor regulation of EF-1alpha expression involve histone acetylation.

Growth factors and their receptors play an important role in the regulation of epithelial cell growth. Abnormalities in the expression, structure, or activity of proto-oncogene products contribute to the development and the pathogenesis of cancer. For example, the human epidermal growth factor receptor (HER1) 1 is overexpressed in a number of epithelial tumor cells (1). HER2, the second member of the HER family, shares extensive sequence homology with the tyrosine kinase domain of HER1 (1,2) and is overexpressed and/or amplified in a number of human malignancies, including breast, ovarian, colon, lung, prostate, and cervical cancers. Recently, HER3 and HER4 have been added to the family, as they share sequence homology with the tyrosine kinase domain of HER1 (2). Regulation of these receptor family members is complex, and they can be transactivated in a ligand-dependent manner. For example, binding of heregulin-␤1 (HRG) to HER3 or HER4 can activate HER2 receptor as a result of HER2/HER3 or HER4/HER2 heterodimeric interactions (3,4). HER1 and HER2 have been shown to induce transformation in recipient cell, possibly because of excessive activation of signal transduction pathways. In contrast, transformation by HER3 or HER4 requires the presence of HER1 or HER2 (3,4).
Since growth factors regulate the proliferation of cancer cells by activating receptors on the surface of the cells, one approach to controlling cell proliferation is to use anti-receptor blocking monoclonal antibodies that interfere with growth factor receptor-mediated autocrine/paracrine growth stimulation. The humanized antibody C225 against the EGF receptor (EGFR) blocks binding of ligand and prevents ligand-induced activation of receptor tyrosine kinase (5,6). C225 is currently being used in phase IIA multicenter clinical trials alone and in combination with chemotherapy or radiation to treat to patients with head and neck, lung, or prostate carcinomas (7)(8)(9). Similarly, the humanized form of anti-HER2 monoclonal antibody HCT (herceptin) inhibits the growth of breast cancer cells overexpressing HER2 (10,11) and is currently being used as an effective drug against some forms of breast cancer (12). Antireceptor antibodies are known to inhibit many processes, including mitogenesis, cell cycle progression, invasion and metastasis, angiogenesis, and DNA repair.
Mitogenic growth factors stimulate protein synthesis in eukaryotic cells. Polypeptide elongation factor-1␣ (EF-1␣) is a ubiquitously expressed protein that plays a key role in the elongation cycle during translation. EF-1␣ forms a complex with aminoacyl-tRNA and GTP that transfers the aminoacyl-tRNA group to the 80 S ribosome and hydrolyzes GTP (13). EF-1␣ is also involved in cytoskeleton reorganization (14,15) and proliferation (16). It can regulate embryogenesis (17), actin bundling, and microtubule severing and is associated with the centrosome and mitotic machinery (14,18,19). EF-1␣ is also one of the actin-associated activators of phosphatidylinositol 4-kinase, which regulates the levels of phosphatidylinositol 4-phosphate and phosphatidylinositol 4,5-bisphosphate (20). These phospholipids regulate the capping of actin filaments by actin-binding protein (18). In brief, EF-1␣ regulates cellular functions that are both dependent on and independent of translational controls.
Next only to actin, EF-1␣ is the most abundant protein in normal cells, accounting for 1-2% of total protein. Regulation of its levels may be important for normal cell function; rapidly growing cells usually exhibit a large increase in their EF-1␣ mRNA levels (21); overexpression of EF-1␣ correlates with metastasis (22); and EF-1␣ mRNA levels decrease during murine erythroleukemic cell differentiation (23). EF-1␣ expression can be regulated at both the transcriptional and post-transcriptional levels (24,25). EF-1␣ mRNA levels have been shown to be up-regulated by oncogenes and induced by phytohemagglutinin in human blood lymphocytes (22,26). In addition, overexpression of EF-1␣ in fibroblasts leads to increased susceptibility to oncogenic transformation (27). In addition to its predominant cytoplasmic presence, EF-1␣ has been reported in the nucleus (28) where it binds to RNA polymerase (29). Recently, EF-1␣ has been shown to be physically associated with the novel zinc finger protein ZPR1 in A-431 cells and to be translocated to the nucleus in an EGF-dependent manner (30). Furthermore, insulin can regulate translational elongation activity of EF-1␣ by regulating its phosphorylation by multipotential S6 kinase (31)(32)(33). Yet, despite knowledge of these cellular functions, the possible regulation of EF-1␣ by the EGF family of growth factors and by therapeutic anti-receptor antibodies remains unexplored.
To identify genes whose expression may be down-regulated by anti-receptor blocking antibodies, presumably owing to ligand-induced activation of receptor tyrosine kinase and/or interference of receptor-associated functions, we used differential display approach to isolate differentially expressed cDNAs. We report that one of these clones had 100% identity with human elongation factor-1␣ (EF-1␣). Both EGF and HRG induced EF-1␣ promoter activity and mRNA and protein expression that could be effectively blocked by pretreatment with antireceptor monoclonal antibodies. Our results also suggest involvement of specific signaling pathways in the base-line regulation of EF-1␣ transcription. In addition, we also provide new evidence to suggest that HRG stimulation of EF-1␣ promoter requires the SP1 site and that the EF-1␣ promoter undergoes histone acetylation in response to HRG.  (6) were maintained in Dulbecco's modified Eagle's medium/F-12 (1:1) supplemented with 10% fetal calf serum. Recombinant HRG-␤1 was purchased from Neomarkers Inc. Anti-vinculin antibody and recombinant EGF were purchased from Sigma.
Differential Display and Cloning of EF-1␣ cDNA-Differential display was performed according to the method described in Ref. 36. In brief, MDA-MB435 and MDA-MB231 cells were treated with or without C225 or herceptin. Total RNA was isolated using Trizol reagent (Life Technologies, Inc.). Total RNA was digested with RNase-free DNase (Promega) and purified by phenol/chloroform extraction. First strand cDNA was synthesized by reverse transcriptase reaction containing 200 ng of total RNA using four different degenerate anchored oligo(dT) primer set (dT12VA, dT12VG, dT12VC, dT12VT; Operon Technologies Inc.). Reactions were performed in a 20-l volume using Moloney murine leukemia virus reverse transcriptase (Promega). Amplification of cDNA fragments was performed using 2 l of the cDNA in reaction buffer containing 2 l of 10ϫ PCR buffer, 10 Ci of ␣-35 S-dATP, dNTPs (2 M), 1 unit of Taq polymerase (Roche Molecular Biochemicals), the same 3Ј-degenerate oligo(dT) primer, and 1 of the 10 5Ј-orbitary decamers (OP-26 -01 to OP-26 -10, Operon Technology, Inc.). PCR cycles were as follows: denaturation 95°C for 5 min, 40 cycles at 94°C for 30 s, 40°C for 2 min, 72°C for 30 s with a final extension of 72°C for 10 min. PCR products were separated on a 6% polyacrylamide gel and developed by autoradiography. Bands of interest were excised, extracted, and reamplified with same set of primers. Amplified DNA was separated on an agarose gel, and bands were purified and cloned into PCR2.1 vector using Topocloning kit (Invitrogen). Five to ten independent clones were isolated, miniprepared, and sequenced at the M. D. Anderson Cancer Core facility. Sequences were compared with GenBank TM sequences using BLAST search.
Construction of EF-1␣ Promoter Deletion Constructs-Construction of deletion constructs pEF (construct numbers 1-7) was described (37). Sub-fragmentation of the insert of construct 6 was done using restriction sites HindIII/SphI, HindIII/StyI, and AvaII/SphI. Cloning of bluntended fragments into the XbaI site of pBLCAT5 (38) yielded constructs 8, 9, and 12, respectively. The mutations in construct 9 were made using the QuickChange kit (Stratagene) according to the instructions.
Cell Extracts, Immunoblotting, and Immunoprecipitation-To prepare cell extracts, cells were washed three times with phosphate-buffered saline and lysed in buffer (50 mM Tris-HCl, pH 7.5, 120 mM NaCl, 0.5% Nonidet P-40, 100 mM NaF, 200 mM NaVO 5 , 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 10 g/ml aprotinin) for 15 min on ice. The lysates were centrifuged in an Eppendorf centrifuge at 4°C for 15 min. Cell lysates containing equal amounts of protein were resolved by SDS-PAGE, transferred to nitrocellulose, and probed with the appropriate antibodies. An equal number of cells were metabolically labeled for 4 -8 h with 100 Ci/ml [ 35 S]methionine in methionine-free medium containing 2% dialyzed fetal bovine serum in the absence or presence of indicated treatments. Cell extracts containing equal trichloroacetic acid-precipitable counts were immunoprecipitated with the desired antibody, resolved on SDS-PAGE, and analyzed (34).
Northern Hybridization-Total cytoplasmic RNA was isolated using the Trizol reagent (Life Technologies, Inc.), and 20 g of RNA was analyzed by Northern hybridization using a 1.8-kilobase pair cDNA fragment of human EF1-␣. rRNA (28 S and 18 S) was used to assess the integrity of the RNA, and for RNA loading and transfer control, the blots were routinely reprobed with glyceraldehyde-3-phosphate dehydrogenase cDNA.
Transfection and Promoter Assays-Cells were split in 100-mm tissue culture dishes (Falcon) 24 h before transfection. Subconfluent cells were transiently transfected with pEF⌬1090CATSp1 or with other constructs as needed or control pSVb-Gal vector using LipofectAMINE method (Life Technologies, Inc.). After 5 h of transfection, medium was changed to Dulbecco's modified Eagle's medium containing 10% serum. After 24 h, cultures were shifted to 0% serum (for growth factor treatment) or 2% serum (for antibody treatment) for 12 h before harvesting. CAT activity was measured 48 h after transfection using a CAT assay kit (Promaga) (39). When indicated, cells were treated with HRG or EGF (30-ng/ml medium) or herceptin or C225 (50 nM final concentration). In some experiments, cells were pretreated with 20 M PD098059 (a MEK inhibitor), 20 M SB203580 (a p38 MAPK inhibitor), and 20 M LY294002 (PI3K inhibitor) for 1 h before HRG treatment. Each experiment was repeated two to five times and transfection efficiency varied between 30 and 50%.
Chromatin Immunoprecipitation Assays-MCF-7 cells were split in 100-mm tissue culture dishes (Falcon). About 70% confluent dishes were serum-starved for 24 h followed by overnight treatment with HRG (30 ng/ml). Quantitative chromatin immunoprecipitation assay was done as described previously (40 -42) with some modifications. Approximately 10 6 cells were treated with formaldehyde (1% final concentration) for 10 min at 37°C to cross-link histones to DNA. The cells were washed twice with phosphate-buffered saline, pH 7.4, containing protease inhibitor mixture (Roche Molecular Biochemicals). Cells were lysed and sonicated as described (40). Sonicated lysate was centrifuged for 10 min at 12,000 rpm at 4°C. Supernatant was diluted 10-fold by dilution buffer containing 0.01% SDS, 1.1% Triton X-100, and protease inhibitor mixture (Roche Molecular Biochemicals). A portion (1%) of the chromatin solution was kept to check the amount of input DNA in different samples before immunoprecipitation. Chromatin solutions were precleared with 80 l of protein A-Sepharose beads (60 mg/ml) saturated with salmon sperm DNA and bovine serum albumin for 30 min at 4°C before immunoprecipitating with either anti-acetylated histone H3 or anti-acetylated H4 antibody (Upstate Biotechnology, Inc.) at 4°C overnight. Immunocomplexes were recovered with 60 l of protein A-Sepharose beads at 4°C for 1 h. Beads were washed as described (40) on a rotating platform before eluting the immunocomplexes by incubation with 400 l of 1% SDS containing 0.1 M NaHCO 3 . The elution was heated to 65°C for 6 h to reverse the formaldehyde cross-links. Phenol/chloroform extraction was performed, and the supernatant was ethanol-precipitated (using 20 g of glycogen as an inert carrier). DNA was resuspended in 50 l of 10 mM Tris, 1 mM EDTA, pH 8.0. Quantitative PCR was done with 10 l of DNA sample restricted to 25 cycles. The EF-1␣ DNA sequence of the 5Ј primer was 5Ј GATTT-GTCCCGGACTAGCGAG and of the 3Ј primer was 5Ј TCTTCTCCAC-CTCAGTGATGACG 3Ј. The PCR products were resolved on a 1.5% agarose gel and stained with ethidium bromide.

RESULTS AND DISCUSSION
Identification of Human Polypeptide Elongation Factor-1␣ as an Anti-receptor Antibody-regulated Gene-In an attempt to identify genes whose expression may be modulated in human cancer cells by anti-receptor blocking antibodies, total RNA was isolated from two highly invasive human breast cancer cell lines, MDA-MB435 and MDA-MB231, and treated with or without C225 and herceptin for 10 h. Although MDA-MB435 and MDA-MB231 cells have normal levels of EGFR and HER2, the in vitro invasive properties of these cells were inhibited by herceptin and C225 (43). A total of 160 reactions was performed using four 3Ј-degenerate oligo(dT) primers and 10 5Ј-random primers for each antibody treatment. Analysis of gels showed amplification of a number of bands ranging from 100 to 600 base pairs, and the majorities of the bands were of equal intensity. By using these bands as internal control, we analyzed for the bands with differences in intensity in C225-or herceptin-treated lanes. This analysis resulted in the identification of five differentially expressed gene products ranging in size from 120 to 350 base pairs. A representative portion of the gel is shown in Fig. 1A.
Each differentially expressed gene product was amplified, cloned, and sequenced. The resulting sequence was compared with sequences deposited in GenBank TM . For each gene product five clones were sequenced, and all five sequences from one band were 100% identical to human elongation factor-1␣ from 1580 to 1694 (EF1␣, GenBank TM accession number X16869) (Fig. 1B). To determine whether EF-1␣ expression can be modulated by anti-receptor blocking antibodies C225 and herceptin on EF-1␣ mRNA levels in tumor cells, we did Northern blot hybridization using the 1.8-kilobase pair EF-1␣ cDNA as a probe. Treatment of human breast carcinoma cells (SK-BR-3, BT-474, and MDA-MB468) and vulvar carcinoma cells (A-431) with C225 or herceptin was accompanied by a significant decrease in EF-1␣ mRNA levels (Fig. 1C).
Regulation of EF-1␣ mRNA Expression by Anti-receptor Antibodies and Growth Factors-A-431 cells, which overexpress EGFR, are growth-stimulated by autocrine transforming growth factor-␣ (TGF-␣) and growth-inhibited by C225 (6). To determine whether TGF-␣ is involved in the regulation of EF-1␣, we asked whether C225, which blocks TGF-␣ from binding to EGFR, could down-regulate the steady-state level of EF-1␣ mRNA. Treatment of A-431 cells with C225 was accompanied by a gradual decrease in EF-1␣ mRNA expression ( Fig. 2A). Similarly, herceptin treatment of BT-474 cells, which overexpress HER2, was associated with reduced expression of EF-1␣ mRNA (Fig. 2B). The observed suppression was not related to high levels of EGFR, as C225 was effective in selectively reducing EF-1␣ levels (but not EF-1␤) in breast cancer MDA-MB231 cells, which have normal levels of EGFR and HER2 (Fig. 2C).
Since MDA-MB231 cells are known to constitutively secrete HRG, a combinatorial ligand for HER3 and HER4 that can transactivate EGFR and HER2, the above results raised the possibility that EF-1␣ mRNA expression could be induced by HRG. Indeed, both HRG and EGF were potent inducers of EF-1␣ mRNA (but not EF-1␤) in MDA-MB231 cells (Fig. 2D). EGF regulation of EF-1␣ mRNA expression was confirmed using a mouse NIH3T3 cell line (HER14) that stably expressed human EGFR and responded to exogenous recombinant EGF by growth stimulation (35). EGF treatment of HER14 cells for 8 h was accompanied by a significant increase in EF-1␣ mRNA levels (Fig. 2E). Taken together, these results suggest that EF-1␣ mRNA levels in a number of cell types are modulated by EGF, HRG, and anti-receptor monoclonal antibodies that interfere with EGFR and HER2.
Regulation of EF-1␣ Protein Expression by Anti-receptor Antibodies and Growth Factors-Western blot analysis was performed to determine whether the modulation of EF-1␣ mRNA levels by growth factors and anti-receptor monoclonal antibodies was associated with a corresponding modulation in the expression of EF-1␣ protein. Results demonstrated that A-431 cells and human colon carcinoma DiFi cells, which have a functional TGF-␣ autocrine loop (44), expressed a lower level of the 51-kDa EF-1␣ protein, after C225 treatment (Fig. 3, A and  B). Similarly, herceptin inhibited EF-1␣ protein levels in BT-474 cells (Fig. 3C). In contrast, treatment of MCF-7 cells with EGF or HRG significantly increased the level of EF-1␣ protein (Fig. 3D).
To validate further the modulation of EF-1␣ protein expression by growth factors, we examined the effects of growth factors and anti-receptor monoclonal antibodies on the level of newly synthesized EF-1␣ protein in cells metabolically labeled with [ 35 S]methionine. Similar to our results from Western analysis, treatment with C225 or herceptin resulted in a reduction of 35 S-labeled EF-1␣ protein in A-431, BT-474, and MDA-MB231 cells (Fig. 4, A-C). In contrast, HRG treatment caused an increase in 35 S-labeled EF-1␣ protein in BT-474 (Fig. 4A) and MDA-MB231 (Fig. 4D) cells and EGF in HER14 cells (Fig.  4E). The observed induction of 35 S-labeled EF-1␣ protein by HRG and EGF was mediated through HER2 and EGFR, as it was effectively suppressed by herceptin and C225 (Fig. 4, A and  E). These results indicate the ability of growth factors to induce expression of EF-1␣ and of growth factor receptor antibodies to block it.
Regulation of the EF-1␣ Promoter by Anti-receptor Antibodies and Growth Factors-The regulatory elements in the human EF-1␣ are not completely understood. Recently, Clark and colleagues (37) have shown the significance of specific elements (44), located in the first intron and also close to the TATA box, in the regulation of EF-1␣ transcription. To examine the effect of growth factor-blocking monoclonal antibodies on EF-1␣ promoter activity, tumor cells were transfected with the EF-1␣ promoter construct pEF⌬1090CATSp1, which contains all regulatory elements upstream of the human EF-1␣ TATA box fused to the thymidine kinase promoter (37). Treatment of A-431 and MDA-MB231 cells with C225 (Fig. 5, A and B) and of MDA-MB231 and BT-474 cells with herceptin (Fig. 5, B and C) resulted in a significant inhibition of EF-1␣ promoter-driven reporter transcription. Conversely, exposure of HER14 cells to EGF was accompanied by 2-4-fold stimulation of EF-1␣ promoter-driven transcription, and EGF receptor antagonist C225 blocked the EGF-mediated stimulation of EF-1␣ promoter activity (Fig. 4D). Similarly, stimulation of EF-1␣ promoter activity by HRG was also suppressed by pretreatment of MCF-7 cells with herceptin (Fig. 4E).
Growth Factor Signaling and Regulation of EF-1␣ Promoter Activity-Distinct signaling pathways regulate different functions of growth factors. For example, HRG utilizes p38 MAPK and PI3K pathways to regulate the spreading and formation of lamellipodia (39,45). A careful analysis of EF-1␣ promoter (GenBank TM accession number E02627) revealed several important motifs, including AP1, SP1, CREB, CRE-BP, and NF-B that are activated by multiple growth factor signaling pathways. To understand the nature of growth factor signaling pathways leading to stimulation of the EF-1␣ promoter, we employed cotransfection of EF1␣ promoter pEF⌬1090CATSp1 reporter with dominant-negative tagged mutants that specifically inhibit p38 MAPK and PI3K activation or dominant-nega- tive mutant of MEK (39,45,46). In these studies, we used HRG as a model growth factor. Mutants of p38 MAPK and MEK, but not PI3K, suppressed both constitutive and the extent of HRGinduced stimulation of EF-1␣ promoter activity in MCF-7 cells (Fig. 6A). The observed inhibitory effects of mutants were not due to variability in the expression levels of the transfected genes in cells treated with or without HRG, as shown by the expression levels of FLAG-tagged p38AF and HA-tagged ⌬p85 by antibodies against FLAG and HA moieties, respectively (Fig. 6B). To confirm these findings further, we next used pharmacological inhibitors, such as PD098059 (for MEK), SB203580 (for p38 MAPK ), and LY294002 (for PI3K), and similar results were obtained (Fig. 6C). In brief, these results suggested a preferential involvement of MEK and p38 MAPK , but not PI3K, in the base-line and HRG-inducible regulation of EF-1␣ promoter activity.
Stimulation of cytoplasmic kinases by growth factors leads to phosphorylation and activation of multiple transcription factors, including c-Jun, ATF-2, NF-B, CREB, and SP1. Since the binding motifs for these transcription factors are present in the EF-1␣ promoter (GenBank TM accession number E02627), it is possible that a combination of these factors may be responsible for optimal EF-1␣ promoter regulation. Recent studies have shown transcriptional regulation of the human EF-1␣ gene by upstream sequences, including a novel C8-stretch element. 2 A Role of SP1 in HRG Regulation of EF-1␣ Promoter-To understand the mechanism by which HRG mediates its effect on EF-1␣ promoter, we employed a series of deletion mutants to map the HRG-responsive regulatory region in the EF-1␣ promoter. All mutants were fussed to a CAT reporter system. Initially, we used constructs 1-6, and we examined the effect of HRG on the EF-1␣ promoter activity. There was minimal effect of the deletions from Ϫ1090 to Ϫ393 of the EF-1␣ promoter on the levels of HRG-mediated up-regulation of EF-1␣ (Fig. 7,  construct numbers 1-6). To delineate further the minimal region required for HRG stimulation of the EF-1␣ promoter, we next used additional mutants (Fig. 7, construct numbers 8, 9,  and 12). Constructs 8 (Ϫ393 to Ϫ204) and 9 (Ϫ313 to Ϫ314) responded to HRG well, whereas constructs 12 (Ϫ204 to Ϫ127) and 7 (Ϫ119 to Ϫ29) showed no regulation in response to HRG. were transiently transfected with a full-length EF-1␣ promoter (pEF⌬1090CATSp1) in the absence or presence of dominant-negative mutants of MEK, p38 MAPK , and PI3K. One set of cultures was treated with or without HRG for 16 h before assaying for CAT activity. Quantitation of CAT activity is shown as fold induction of CAT reporter activity by HRG treatment over the control culture for each mutant group. B, Western blot analysis of expression of FLAG-tagged p38 MAPK and HA-tagged PI3K mutants in lysates from the above panel, using antibodies against FLAG-tagged and HA epitopes, respectively. NS indicates nonspecific band in the same blot. Results shown are representative of two experiments. C, cells were pretreated with specific inhibitors PD098059, SB203580, and LY294002 (20 M each) for 1 h before HRG treatment for 16 h. Quantitation of CAT activity is shown as fold induction of CAT reporter activity by HRG treatment over the control culture in each inhibitor.
The results indicated that Ϫ393 to Ϫ314 region contained the regulatory elements that may confer the HRG-mediated induction of EF-1␣. Analysis of the transcription factor sites present in the Ϫ393 to Ϫ314 revealed presence of an SP1 site. Recently it was shown that HRG regulate the activity of SP1 transcription factor (47). To verify the potential involvement of the SP1 site in HRG induction of EF-1␣ promoter, we next mutated the SP1 site at Ϫ369 to Ϫ363 (construct 11). Point mutation of the SP1 site completely abolished the HRG-mediated induction of the construct containing Ϫ393 to Ϫ314 region. Mutation of another region other than Sp1 site has no effect on HRGmediated induction (construct 10), suggesting a role of SP1 in the HRG regulation of EF-1␣ promoter. Interestingly, construct 8 (Ϫ393 to Ϫ204), which contains two Sp1 sites, demonstrated a significantly higher HRG-inducible activity (4-fold) than the construct containing Ϫ393 to Ϫ314 (construct number 10) with one SP1 site (2.1-fold). Together, these observations suggested a role of SP1 site in the observed HRG-mediated regulation of EF-1␣ promoter.
Involvement of Histone Acetylation in HRG Regulation of EF-1␣-The eukaryotic genome is compacted with histone and other proteins to form chromatin, which consists of repeating units of nucleosomes (48,49). For transcription factors to access DNA, the repressive chromatin structure needs to be remodeled. Dynamic alterations in the chromatin structure can facilitate or suppress the access of the transcription factors to nucleosomal DNA, leading to transcription regulation. One way to achieve this is through alterations in the acetylation state of nucleosomal histones. Hyperacetylated chromatin is generally associated with transcription activation, whereas hypoacetylated chromatin is associated with transcriptional repression (48,49). To investigate whether the HRG regulation of EF-1␣ expression involves histone acetylation on the EF-1␣ gene, we next performed chromatin immunoprecipitation (ChIP) assay of the target gene, i.e. EF-␣ around a target sequence in the promoter (Ϫ535 to Ϫ209) which has multiple SP1 sites by using antibodies specific for acetylated forms of H3 and H4. Representative results from several independent ex-periments are shown in Fig. 8. HRG treatment of MCF-7 cells was accompanied by a significant enhancement in the association of EF-1␣ promoter region with the acetylated histones H3 and H4 (5.0-and 3.5-fold induction of associated-acetylated H3 and -acetylated H4 by HRG as compared with untreated cells). In addition, there was also easily detectable levels of EF-1␣ promoter association with the acetylated H3 and H4 in control untreated cells (Fig. 8, lanes 1 and 3), implying a potential role of histone acetylation in the base-line expression of EF-1␣. A, deletion constructs were made as described under "Experimental Procedures." Fold of induction by HRG over untreated control cells is shown in the right-hand columns. Construct numbers 10 -12 have similar sequence, except that number 10 is wild type and has one Sp1 site at Ϫ379 to Ϫ374 position; construct number 11 was mutated nonspecifically at Ϫ387 to Ϫ385 where CCC were replaced with AGA, and construct number 12 was mutated at Sp1 site at Ϫ391 to Ϫ389 where CGC were replaced with 5Ј TTA 3Ј. B, regulation of EF-1␣ deletion constructs by HRG (16 h). Results shown are representative of three experiments.
Earlier reports have shown a close correlation of up-regulation of histone H3 and H4 acetylation with an increased transcriptional activity (48,49). In brief, our findings clearly demonstrated the involvement of histone acetylation in HRG-mediated stimulation of EF-␣ gene expression.
In summary, we have presented new evidence that treatment of tumor cells with growth factors significantly increases EF-1␣ promoter activity and mRNA and protein expression and that HRG increases the acetylation of H3 and H4 on the EF-1␣ promoter. Since constitutive EF-1␣ expression was not well correlated with EGFR or HER2 overexpression in different cell types, ligand-activated cellular pathways, rather than receptor levels, may be significant in the regulation of EF-1␣ expression. These views are supported by a recent report showing no effect of EGFR and HER2 overexpression on levels of related EF-1␦ in human keratinocytes (50). However, despite the lack of correlation with receptor levels, the ligand-induced up-regulation of EF-1␣ expression was mediated by a specific ligand-receptor interaction, as the anti-receptor blocking antibodies C225 and herceptin could effectively reduce it. Our study also shows the potential roles of MEK and p38 MAPK in the constitutive regulation of EF-1␣ promoter activity.
There are a number of possible functional implications for growth factor-regulated EF-1␣ expression as follows: 1) promotion of polypeptide elongation and thus potential contribution to increased translation of mRNA encoding growth-related proteins; 2) increased reorganization of the cytoskeleton, since it is one of the earliest phenotypic responses of most cells to growth factors and since EF-1␣ regulates actin bundling; 3) potential undefined roles in the nucleus, due to the ability of EF-1␣ to form a complex with ZPR1 (30). In summary, our findings have clearly demonstrated for the first time a potential role of EF-1␣ in the actions of HER growth factors that are widely deregulated in human cancers and that ligand-dependent EF-1␣ expression was effectively inhibited with humanized anti-receptor blocking antibodies C225 and herceptin. In addition, we also provide new evidence to suggest that HRG stimulation of EF-1␣ promoter requires the SP1 site and that the EF-1␣ promoter undergoes histone acetylation in response to HRG.