Phosphorylation of Histone H3 at Serine 10 Is Indispensable for Neoplastic Cell Transformation

Phosphorylation of histone H3 has been linked with mitotic chromatin condensation. Histone H3 is phosphorylated during mitosis on at least two serine residues, Ser¹⁰ (1, 2) and Ser²⁸ (3). Phosphorylation at Ser¹⁰ begins in early G₂ in the pericentromeric heterochromatin of each chromosome (4) and by metaphase has spread throughout all chromosomes, whereas phosphorylation on Ser²⁸ starts to be evident only in early mitosis (3). Phosphorylation of histone H3 at Ser¹⁰ seems to be involved in the initiation of mammalian chromosome condensation but not the maintenance (5). Increased phosphorylation of histone H3 at Ser¹⁰ was found in mitogen-stimulated and oncogene-transformed mouse fibroblasts (6). Moreover, the phosphorylation of histone H3 at NH₂-terminal Ser¹⁰ is closely related to the induction of immediate-early response genes, including proto-oncogenes c-fos and c-jun (2, 7, 8). These and other immediate-early genes are rapidly and transiently expressed in response to extracellular stimuli. The immediate-early gene response has been implicated in differentiation, mitosis, and disease, such as inflammation and cancer (9, 10). However, much less is known about the role of histone H3 in epidermal growth factor (EGF)–induced neoplastic cell transformation.

The EGF is a well-known tumor promotion agent used to study malignant cell transformation in cell and animal models of cancer (11). EGF induces activation of the transcription factor, activator protein-1 (AP-1; ref. 12). When treated with EGF, JB6 Cl41 skin epidermal cells showed an induction of AP-1 transcriptional activation in promotion-sensitive (P⁺) phenotypes but not in promotion-resistant (P⁻) phenotypes (13). Blocking AP-1 activation causes P⁺ cells to revert to the P⁻ phenotype, indicating a unique requirement for AP-1 activation in EGF-induced cell transformation (14). In previous studies, phosphorylation of histone H3 at Ser¹⁰ was shown to be involved in different signal transduction pathways and to be dependent on the specific stimulation or stress. EGF induces phosphorylation of H3 at Ser¹⁰, which is mediated by RSK2 (15). RSK2 mutation in humans is linked to Coffin-Lowery syndrome and fibroblasts derived from a Coffin-Lowery syndrome patient fail to exhibit EGF-stimulated phosphorylation of histone H3 at Ser¹⁰ (15). In addition, mitogen- and stress-activated protein kinase (MSK1) has been shown to mediate EGF or 12-O-tetradecanoylphorbol-13-acetate (TPA)–induced phosphorylation of histone H3 at Ser¹⁰ (15). These results suggested that histone H3 might be regulated by mitogen-derived signal transduction induced by tumor promoters, such as EGF. Although EGF signaling cascades have been thoroughly studied, the involvement of histone H3 in the neoplastic cell transformation is unknown.

Recently, members of the Aurora kinase family, which are overexpressed in many human cancer cells, were shown to be responsible for mitotic phosphorylation of histone H3 at Ser¹⁰ in yeast and nematodes (16–18). Moreover, increased H3 histone phosphorylation as a result of AIM-1 (Aurora B) overexpression is a major precipitating factor of chromosome instability and thus may play a role in carcinogenesis (19). Because mitosis is the phase of the cell cycle most susceptible to various chemical and physical agents, the consistent observation of histone H3 phosphorylation at Ser¹⁰ and/or overexpression in human tumor cell lines (20) suggested that regulation of mitosis by phosphorylation of histone H3 at Ser¹⁰ may be a possible target for cancer treatment. Thus, although AIM-1-mediated mitotic H3 phosphorylation seems to induce carcinogenesis, whether histone H3 directly is essential for neoplastic cell transformation and cancer development is still an intriguing question. In this study, we investigated whether phosphorylation of histone H3 is required for inducing neoplastic cell transformation using knockdown and dominant-negative constructs of histone H3, which were stably transfected with psi-H3 or pV5-H3 mutant constructs (S10A, S28A, or S10/28A), respectively.

Reagents and antibodies. Eagle's MEM, l-glutamine, and gentamicin were purchased from Invitrogen (Carlsbad, CA). EGF was purchased from EMD Biosciences, Inc. (San Diego, CA) and fetal bovine serum (FBS) from Gemini Bio-Products (Calabasas, CA). Polyvinylidene difluoride (PVDF) membrane was from Millipore (Bedford, MA). The DNA ligation kit (version 2.0) was purchased from Takara Bio, Inc. (Shiga, Japan). Antibodies against histone H2A or the V5 epitope were purchased from Upstate (Lake Placid, NY) or Invitrogen (Carlsbad, CA), respectively. Antibodies against c-Jun or c-Fos were purchased from Cell Signaling Technology, Inc. (Beverly, MA) or EMD Biosciences, respectively. Control small interfering RNA (siRNA), c-Jun siRNA (mouse) or c-Fos siRNA (mouse), was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) and jetSI-ENDO was from Qbiogene (Irvine, CA).

Cell culture and transfection. JB6 Cl41 mouse epidermal cells were cultured in MEM supplemented with 5% FBS. psi-H3 or pV5-H3 stably transfected JB6 Cl41 cells were established in MEM supplemented with 5% FBS containing 600 μg/mL G418 and maintained in MEM supplemented with 5% FBS containing 200 μg/mL G418. Cells were transfected by a cationic polymer transfection method, jetPEI (Qbiogene, Carlsbad, CA). Transfection of siRNA duplex with jetSI-ENDO was done following the manufacturer's protocol.

Construction of mammalian expression vector and small interfering RNA vector. A cDNA fragment encoding human histone H3 was inserted in-frame into the BamHI/EcoRI sites of the pcDNA3.1/V5-His vector (Invitrogen, San Diego, CA) to produce the V5 epitope–tagged construct, pV5-H3. cDNAs encoding the S10A, S28A, and S10/28A mutants of histone H3 were generated using the QuickChange II site-directed mutagenesis kit (Stratagene, La Jolla, CA) and subcloned into the pcDNA3.1/V5-His vector to produce the V5 epitope–tagged constructs, pV5-H3 S10A, pV5-H3 S28A, and pV5-H3 S10/28A, respectively. All of the positive clones containing cDNA inserts were identified by restriction enzyme mapping and sequenced at Genewiz, Inc. (North Brunswick, NJ). To construct the siRNA-histone H3 (psi-H3), the pU6pro vector (provided by David L. Turner, University of Michigan, Ann Arbor, MI) was digested with XbaI and BbsI. The annealed synthetic primers were then introduced: sense siRNA 5′-TTTGCAGACAGCTCGGAAATCCATTCAAGAGATGGATTTCCGAGCTGTCTGTTTTT and antisense siRNA GTCTGTCGAGCCTTTAGGTAAGTTCTCTACCTAAAGGCT-CGACAGACAAAAAGATC-5′ following the recommended protocols (21). The psi-H3 was confirmed by agarose gel electrophoresis and DNA sequencing. pcDNA3.1-H-Ras^G12V were provided by University of Missouri-Rolla cDNA Resource Center (Rolla, MO).

3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfonyl)-2H-tetrazolium assay. To estimate cell proliferation, JB6 Cl41 cells transfected with pU6pro, psi-H3, pV5, or pV5-H3 were seeded (1 × 10³) in 96-well plates in 100 μL of 5% FBS-MEM at 37°C in a 5% CO₂ incubator. After culturing for 12 hours, CellTiter 96 Aqueous One Solution (20 μL; Promega, Madison, WI) were added to each well, and cells were then incubated for 1 hour at 37°C in a 5% CO₂. Absorbance was measured at 492 and 690 nm.

Anchorage-independent cell transformation assay (soft agar assay). EGF-induced cell transformation was investigated in mock, psi-H3, or pV5-H3 stably transfected cells. In brief, cells (8 × 10³/mL) were exposed to EGF (0.1-10 ng/mL) in 1 mL of 0.3% basal medium Eagle's agar containing 10% FBS. The cultures were maintained at 37°C in a 5% CO₂ incubator for 10 days, and the cell colonies were scored using a microscope and the Image-Pro PLUS computer software program (Media Cybernetics, Silver Spring, MD; ref. 22).

Focus-forming assay. Transformation of Swiss 3T3 cells was done following standard protocols (23). Swiss 3T3 cells were plated in 60 mm dishes at a density 1 × 10⁴ cells. Cells were transiently transfected with 100 ng H-Ras^G12V, 2.5 μg pV5, or 2.5 μg pV5-H3 plasmid. Cells were kept in Eagle's MEM with 2% FBS. The medium was changed every 3 days for a period of 3 weeks. Foci were enumerated by staining the monolayer with methanol for fixation and 0.4% crystal violet. Data were obtained from three independent plates for each transfection.

Reporter gene assays. The reporter gene assay for firefly luciferase activity was done using lysates from transfected cells. In addition, the reporter gene vector pRL-SV40 (Promega) was cotransfected into each cell line and the Renilla luciferase activity generated by this vector was used to normalize the results for transfection efficiency. Cell lysates were prepared by first washing the transfected JB6 Cl41 cells (grown in 60 mm diameter dishes) once in PBS at 37°C. After removing the PBS completely, passive lysis buffer (500 μL; Promega Dual Luciferase Reporter Assay System) was added; then, cells were incubated for 1 hour with gentle shaking. The lysate was then transferred into a reaction tube and the cellular debris was removed by centrifugation. The supernatant fraction was used for the measurement of firefly and Renilla luciferase activities. Cell lysates (20 μL each) were mixed with 100 μL Luciferase Assay II reagent and firefly luciferase light emission was measured by a Luminoskan Ascent (Thermo Electron Corp., Helsinki, Finland). Subsequently, coelenterazine reagent (100 μL) containing the substrate for Renilla luciferase light emission was mixed to normalize the firefly luciferase data. The results are expressed as relative c-Fos, c-Jun, or AP-1 activity (fold or percentage) and are presented as luciferase activity relative to the c-Fos, c-Jun, or AP-1-only transfected control cells. c-Fos promoter luciferase (pFos-WT GL3) and c-Jun promoter luciferase (pJC6 GL3) constructs were kindly provided by Dr. Ron Prywes (Columbia University, New York, NY). The AP-1 luciferase reporter plasmid (−73/+63 collagenase-luciferase) was constructed as reported previously (24, 25).

Isolation of histone proteins. For the isolation of histone proteins, cells (2 × 10⁷-5 × 10⁷) were homogenized in 1 mL nuclear preparation buffer [10 mmol/L Tris-HCl (pH 7.6), 150 mmol/L NaCl, 1.5 mmol/L MgCl₂, 0.65% NP40, 1 mmol/L phenylmethylsulfonyl fluoride (PMSF)] in the presence of protein phosphatase inhibitors (10 mmol/L NaF, 1 mmol/L sodium orthovanadate, 25 mmol/L β-glycerophosphate). Nuclei were recovered by centrifugation at 1,500 × g for 10 minutes. All centrifugations were carried out at 4°C. Nuclei were resuspended in 0.3 mL resuspension buffer [10 mmol/L Tris-HCl (pH 7.6), 3 mmol/L MgCl₂, 10 mmol/L NaCl, 1 mmol/L PMSF, protein phosphatase inhibitors]. Nuclei were extracted with 0.4 N H₂SO₄ to isolate total histones. The samples were precipitated with trichloroacetic acid and then resuspended in double-distilled H₂O.

Immunoblotting. The proteins were resolved by SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked and hybridized with the appropriate primary antibody overnight at 4°C. Histone H3, V5 epitope histone H3, or histone H2A was detected with an antibody against anti–histone H3, anti-V5, or anti–histone 2A, respectively. Protein bands were visualized by the chemiluminescence detection kit (enhanced chemiluminescence, Amersham Biosciences Corp., Piscataway, NJ) after hybridization with the horseradish peroxidase–conjugated secondary antibody from rabbit or mouse.

Epidermal growth factor–induced neoplastic cell transformation is suppressed by knockdown of histone H3. To decrease endogenous histone H3 protein level by the introduction of U6 hairpin siRNA, we synthesized a duplex length of 19 nucleotides with a 9-nucleotide loop specific for histone H3 mRNA using the sites conserved on both the human and the mouse genes (Fig. 1A and B). Our designed siRNA target sequence for histone H3 was based on an expression vector with a mouse U6 promoter, in which we could insert a sequence after the first nucleotides of the U6 transcript (a G). The mouse U6pro vector is expected to express 21-nucleotide complementary single-stranded RNAs with 19 nucleotides corresponding to the sense or antisense strands. In addition to JB6 Cl41 cells stably expressing histone H3 siRNA (psi-H3), we constructed JB6 Cl41 cells stably overexpressing the mock vector (pU6pro) as a control. In our initial experiments, we observed up to a 50% silencing of histone H3 induced by the introduction of siRNA compared with cells expressing mock vector only (Fig. 1C). To test the knockdown effect of H3 in the cell proliferation, the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfonyl)-2H-tetrazolium (MTS) assay was used to examine the proliferation in cells stably transfected with pU6pro or psi-H3. psi-H3 stably transfected cells showed a marked decrease in the rate of proliferation compared with pU6pro stably transfected cells (P < 0.005; Fig. 1D). Using these cell lines, we studied differences in EGF-promoted cell transformation using our previously developed methods (26). Mock vector (pU6pro) cells and cells expressing siRNA against histone H3 (psi-H3) were treated separately with EGF (0.1, 1.0, or 10 ng/mL) in soft agar and incubated at 37°C in a 5% CO₂ incubator for 10 days. Our results showed that psi-H3 cells significantly inhibited the formation of EGF-promoted colonies compared with mock vector control cells (Fig. 1E). The inhibition was evident not only in colony number but also in colony size. Untreated pU6pro and psi-H3 cells showed no colony formation. These data indicated that histone H3 is involved in JB6 Cl41 cell proliferation as well as cell transformation as a positive regulator.

The overexpression of histone H3 enhanced epidermal growth factor–induced neoplastic cell transformation. EGF stimulates the proliferation of several epithelial cells both in vivo and in vitro and is implicated in epithelial tumor formation (27). Mitogen-activated protein kinase (MAPK) and histone H3 were rapidly and transiently phosphorylated on EGF stimulation. An accumulation of recent evidence showed that phosphorylation of histone H3 is mediated by EGF (28). To better determine the role of histone H3 in neoplastic cell transformation, we cloned the human histone H3 cDNA, including the open reading frame, and recombined it into the mammalian expression vector, pcDNA3.1/V5 (pV5-H3). This plasmid was then introduced into JB6 Cl41 mouse skin epidermal cells, and neoplastic cell transformation was assessed. Overexpression of histone H3 was detected in histone H3 stably transfected cells (pV5-H3) using an anti-V5 antibody (Fig. 2A). The effect of histone H3 on cell proliferation was tested in the pV5-H3 cells compared with mock cells because knockdown of histone H3 by siRNA decreased cell proliferation as well as neoplastic cell transformation. The MTS assay results indicated that histone H3–overexpressing cells proliferated at a significantly higher rate compared with mock vector stably transfected control cells (P < 0.005; Fig. 2B). This result confirmed that histone H3 is involved in cell proliferation. Next, cells (mock and pV5-H3) were employed to determine whether they were susceptible to cell transformation induced by EGF. Cells were treated with EGF (0.1, 1.0, or 10 ng/mL) in soft agar and incubated at 37°C in a 5% CO₂ incubator for 10 days and colony numbers determined as described. Results indicated increased colony numbers as well as colony sizes in cells overexpressing wild-type histone H3 (H3 WT) promoted by EGF compared with control cells stimulated by EGF under the same conditions (Fig. 2C). The overexpression of histone H3 caused a dramatic increase in cell transformation without addition of EGF (Fig. 2C). This result strongly suggests that histone H3 also might induce cell transformation independent of EGF-induced signal transduction. To clarify this, we next decided to assess the focus-forming activity elicited by histone H3 in a reduced serum condition (2% FBS) without EGF. Swiss 3T3 cells were transiently transfected with each expression vector (mock, pV5-H3, or H-Ras^G12V). After ∼3 weeks, the formation of small cell clusters became clearly visible. Ras^G12V-transfected cells showed a large number of transformed foci, and very importantly, the transfection of histone H3 (pV5-H3) into Swiss 3T3 cells readily induced the appearance of foci of transformation compared with mock cells (Fig. 2D). Taken together, these data strongly support the idea that histone H3 contributes to neoplastic cell transformation in epidermal mouse skin cells stimulated or not stimulated with EGF.

Motif of histone H3 phosphorylation responsible for neoplastic cell transformation promoted by epidermal growth factor. Ser¹⁰ and Ser²⁸ on the histone H3 tail are each proceeded by the same three amino acids (alanine-arginine-lysine) and both of these phosphorylatable motifs are very highly conserved, being identical in yeast and human (29). Two distinct modes of histone H3 phosphorylation have been characterized, mitotic and stimulus-inducible phosphorylation, both of which occur at these two serine residues. Furthermore, MAPK-mediated phosphorylation of Ser¹⁰ and Ser²⁸ has also been shown on stimulation with EGF (28), UVB (30, 31), TPA, or anisomycin (32). To study whether the histone H3 phosphorylatable motif at Ser¹⁰ or Ser²⁸ specifically induces cell transformation promoted by EGF, we replaced Ser¹⁰ of histone H3 with alanine (S10A), Ser²⁸ with alanine (S28A), or both Ser¹⁰ and Ser²⁸ with alanine (S10/28A) and then subcloned these constructs into the pcDNA3.1/V5-His vector (Fig. 3A). These plasmids were then introduced into JB6 Cl41 cells. Expression of constructs was confirmed with an anti-V5 antibody against V5-histone H3 (Fig. 3B). Stable transfectant H3 WT or mutants, S10A, S28A, or S10/28A, were then employed to determine their susceptibility to cell transformation promoted by EGF. Cells were treated with EGF (1 ng/mL) in a soft agar matrix and incubated at 37°C in a 5% CO₂ incubator for 10 days. Results indicated that colony numbers promoted by EGF were greatly decreased in S10A, but not S28A, mutant JB6 Cl41 cells compared with cell transformation in JB6 Cl41 cells overexpressing H3 WT (Fig. 3C). These data indicate that Ser¹⁰ of histone H3 is most likely a critical site for mediating neoplastic cell transformation promoted by EGF.

Induction of the c-jun and c-fos promoter is critical for neoplastic cell transformation elicited by H3. c-Jun and c-Fos are nuclear proto-oncoproteins whose expression is stimulated by a variety of growth-promoting agents and activated oncogenes (9). Phospho-acetylated histone H3 occurs on nucleosomes associated with the active, but not the inactive, c-fos and c-jun proto-oncogene promoters under physiologic conditions using EGF as a stimulus (33). To investigate whether cell transformation elicited by overexpression of H3 results from direct stimulation of the c-jun or c-fos promoter, we took advantage of the availability of the reporter plasmid carrying the luc gene under the control of the murine c-jun or c-fos promoter (34, 35). Twenty-four hours after transfection with these reporters in mock or H3 WT–overexpressing JB6 Cl41 cells, cells were starved for another 24 hours by incubating in serum-deprived MEM at 37°C in a 5% CO₂ atmosphere. Cells were then treated or not treated with EGF (1, 10, 30, 50, and 100 ng/mL). As shown in Fig. 4A and B, EGF-promoted c-jun or c-fos transcriptional activity was significantly induced in JB6 Cl41 cells overexpressing H3 WT compared with cells expressing pcDNA3.1/V5 (mock). These data support our notion that induction of the c-jun or c-fos promoter by EGF is one of the mechanisms explaining the increased anchorage-independent growth of JB6 Cl41 cells overexpressing H3 WT. In contrast, the histone H3 mutant, S10A (H3 S10A), but not S28A (H3 S28A), suppressed EGF-promoted c-jun or c-fos gene induction compared with H3 WT cells (Fig. 4C). Cotransfection of the reporters with the histone H3 cDNA in JB6 Cl41 cells revealed that histone H3 can strongly induce the activity of the c-jun or c-fos promoter (Fig. 4D). These data provide evidence that the induction of the c-jun and c-fos genes is involved in neoplastic cell transformation induced by overexpression of the H3 WT protein.

We next decided to assess the effect of histone H3 on the expression of the endogenous c-Jun or c-Fos proteins. For these experiments, mock, H3 WT, or H3 S10A cells were treated with EGF (30 ng/mL) in a time-dependent way (1, 2, or 3 hours); then, the endogenous c-Jun or c-Fos protein level was analyzed by immunoblotting using a specific antibody against c-Jun or c-Fos, respectively. As shown in Fig. 4E, only H3 WT (middle), but not H3 S10A, cells dramatically increased endogenous c-Jun and c-Fos protein levels induced by EGF compared with mock cells. Taken together, these data therefore indicate that histone H3 can regulate the expression of c-Jun and c-Fos and, in turn, that c-Jun and c-Fos are potential candidates to mediate the neoplastic cell transformation elicited by histone H3.

Activator protein-1 activity is enhanced in H3-overexpressing cells. The AP-1 transcription factor is a dimeric complex that comprises members of the Jun, Fos, activating transcription factor, and musculoaponeurotic fibrosarcoma protein families (36). The regulation of cell proliferation by AP-1 might be of crucial importance for the multistage development of tumors (37, 38). AP-1 is induced by several external stimuli, such as EGF, which increase MAPK activity (39). To determine whether the overexpression of H3 WT in JB6 Cl41 cells is responsible for the AP-1 activation response to EGF, we next cotransfected the AP-1 luciferase reporter plasmid and the pRL-SV40 gene into H3 WT–overexpressing or control JB6 Cl41 cells. At 24 hours after transfection, cells were starved for an additional 24 hours by incubating in serum-deprived MEM at 37°C in a 5% CO₂ atmosphere. They were then treated or not treated with EGF (1, 10, 30, 50, and 100 ng/mL). The EGF-induced AP-1 activation response in JB6 Cl41 cells was much more pronounced in H3 WT–overexpressing cells (pV5-H3 WT) than mock vector control cells (Fig. 5A). In contrast, no significant gains of AP-1 transactivation were observed in histone H3 mutant S10A (pV5-H3 S10A)– or S10/28A (pV5-H3 S10/28A)–overexpressing JB6 Cl41 cells (Fig. 5B). When H3 WT was transiently overexpressed in JB6 Cl41 cells, AP-1 luciferase activity was increased in a dose-dependent way by overexpression of H3 WT (pV5-H3 WT), similar to that observed for c-jun or c-fos activity (Fig. 5C).

A knockdown of c-Jun and c-Fos inhibits H3-induced cell transformation. To evaluate the function of c-Jun and c-Fos in neoplastic cell transformation induced by overexpression of histone H3, we tested whether siRNA-mediated knockdown of c-Jun or c-Fos in pV5-H3 WT cells would inhibit the cell-transforming activity. Immunoblot analysis of c-Jun or c-Fos siRNA-transfected pV5-H3 cells revealed that a maximal protein knockdown of up to 90% or 70%, respectively, was induced by c-Jun-specific or c-Fos-specific siRNA but had no effect on β-actin protein expression (Fig. 6A and B). We next examined whether knockdown of c-Jun or c-Fos affected the ability of histone H3 to form colonies in soft agar. Mock or pV5-H3 WT cells were treated with 100 pmol each of control siRNA, c-Jun-specific, or c-Fos-specific siRNA and plated in soft agar with or without EGF. Cells were incubated at 37°C in a 5% CO₂ incubator for only 4 days. The overexpression of histone H3 caused a dramatic increase in cell transformation with addition of EGF (Fig. 6C). In contrast, both c-Jun-specific and c-Fos-specific siRNAs almost totally inhibited the number of colonies formed in the H3-enhanced cell transformation by EGF (Fig. 6C). Finally, to determine whether c-Jun-specific or c-Fos-specific siRNA inhibits cell transformation in the histone-overexpressing cells through the AP-1 transactivation activity, we treated cells with 100 pmol each of control siRNA, c-Jun-specific, or c-Fos-specific siRNA. After 24-hour treatment of siRNAs, we transfected AP-1-luciferase reporter plasmid in the mock or pV5-H3 cells and used these transfectants for analysis of AP-1 transactivation activities. At 24 hours after transfection, cells were starved for an additional 24 hours by incubating in serum-deprived MEM at 37°C in a 5% CO₂ atmosphere. They were then treated or not treated with EGF (30 ng/mL) for 3 hours. The EGF-induced AP-1 activation response in JB6 Cl41 cells was much more pronounced in H3 WT–overexpressing cells (pV5-H3 WT) than mock vector control cells (Fig. 6D). In contrast, no significant gains of AP-1 transactivation were observed in c-Jun- or c-Fos- siRNA treated pV5-H3 cells (Fig. 6D).

Mitotic phosphorylation of histone H3 at Ser¹⁰ is critical for proper chromosome condensation and segregation (40). The phosphorylation status of histone H3 at Ser¹⁰, in turn, influences methylation at Lys⁹ of histone H3 by the SUV39H1 methylase (41). Methylation of histone H3 at Lys⁹ by SUV39H1 has also been implicated in recruiting heterochromatin protein 1 to heterochromatin sites, where it mediates gene silencing (42–44). This observation has led us to believe that chromatin-based regulatory mechanisms may be involved in epigenetic alteration of cancer cells leading to tumor establishment and progression. However, the intriguing question is whether the modification in the NH₂-terminal tails of histone H3 directly affects neoplastic cell transformation.

In this study, we found that overexpression of histone H3 induced neoplastic cell transformation and cell proliferation in JB6 Cl41 cells. In contrast, knockdown of histone H3 expression using the siRNA technique suppressed cell transformation and cell proliferation. Surprisingly, mutation of Ser¹⁰ of histone H3 with alanine suppressed cell transformation promoted by EGF. Numerous oncogenes act by mimicking normal growth signaling and many of the cell surface receptors that transduce growth-stimulatory signals into the cell interior are themselves targets of deregulation during tumor pathogenesis. For example, the EGF receptor is up-regulated in the stomach, brain, and breast tumors (45), and the SOS-Ras-Raf-MAPK cascade plays a central role in tumorigenesis stimulated by growth factors, such as EGF (46). In ∼25% of human tumors, Ras proteins are present in structurally altered forms that enable them to release a flux of mitogenic signals into cells without stimulation by their normal upstream regulators (47). Constitutive activation of the Ras-MAPK signaling pathway in mouse fibroblasts transformed with oncogenes (e.g., H-ras) elevates the level of phosphorylated H3 (6). The phosphorylation of histone H3 at Ser¹⁰ is essential for maintenance of proper chromosome dynamics during mitosis (40). As indicated earlier, the elevation of phosphorylated H3 (Ser¹⁰) levels may contribute to the aberrant gene expression observed in the oncogene-transformed cells, resulting from persistent activation of the Ras-MAPK pathway and MSK1 (48). Increased mitotic H3 (Ser¹⁰) phosphorylation was also observed in various colorectal tumor cells with high AIM-1 expression levels (19). AIM-1, a mammalian Ipl1/aurora kinase involved in H3 phosphorylation, is overexpressed in many tumor cell lines, including human colorectal tumors of advanced grade and stage. This accumulated evidence suggested that the post-translational modification of histone H3 at the NH₂-terminal tail might not only regulate gene expression, chromosomal condensation, and mitosis but also have oncogenic effects. These previous studies strongly supported our finding that histone H3, especially the Ser¹⁰ motif, might have an important role in carcinogenesis, resulting from the EGF-induced activation of MSK1 and the Ras-Raf-MAPK pathway.

For signaling pathway–activated genes, such as c-fos and c-jun, the coupling of histone phosphorylation at specific loci may contribute to mechanisms that allow them to be rapidly activated in response to external stimuli (28). Overexpression of the Fos and Jun proteins was found to correlate with a positive effect on cell transformation (33). c-Jun is believed to be primarily a positive regulator of cell proliferation because c-Jun-deficient fibroblasts have a marked proliferation defect in vitro (49). To fully promote cell cycle progression, the c-Jun protein is activated by c-Jun NH₂-terminal kinases (50). Subsequently, the activated c-Jun-containing AP-1 complex induces the transcription of positive regulators of cell cycle progression, including cyclin D1, or represses negative regulators, including the tumor suppressor p53 and the cyclin-dependent kinase INK4A (36). c-Fos might have a more important role than c-Jun during late-stage tumorigenesis (51). The importance of c-Fos in tumor invasion has been supported in vivo, as the progression of chemically induced papillomas to invasive squamous cell carcinomas is impaired in c-Fos-deficient mice (52).

In this study, although many factors have been shown to interact with and activate histone H3, we found that the c-fos or c-jun gene is a common target of histone H3 leading to induction of AP-1 activity, which may include its transactivating potential and DNA-binding capacity and the stability of AP-1 components. The phosphorylation of the ternary complex factor by the Ras-Raf-MAPK/extracellular signal-regulated kinase (ERK) kinase-ERK signal cascade (12) and the stimulation of the activity of the c-fos SRE by histone H3 phosphorylation (Ser¹⁰; ref. 28) might promote c-Fos expression and thereby stabilize the c-Fos/c-Jun heterodimer leading to expression of AP-1-regulated genes (53). By linking histone H3 phosphorylation to AP-1 transactivating potential, these finding provide an attractive explanation of how post-translational modification of histone H3 can be related to neoplastic cell transformation. Thus, these results suggest that histone H3 may be a crucial target for cancer chemotherapy or genetic therapy in the future.

Grant support: The Hormel Foundation and NIH grants CA77646 and CA81064.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Dr. Ron Prywes for the c-Jun-Luc (pJC6 CL3) and c-Fos-Luc (c-Fos WT CL3) promoter.